Document JoinQuant cost model findings

- pyproject.toml: fail_under 80 → 95
- test_alpha: +79 lines
- test_cli_workflow: +226 lines
- test_derived: +121 lines
- test_downloader_contracts: +169 lines
- test_features: +16 lines
- test_minute_downloader: +81 lines
- test_portfolio: +208 lines

.gitignore

@@ -1,6 +1,7 @@
 __pycache__/
 *.py[cod]
 .pytest_cache/
+.coverage
 *.egg-info/
 .venv/
 venv/

CLAUDE.md

@@ -26,6 +26,10 @@ uv run python cli.py portfolio eval --positions-path portfolio/eq_10m.pq --data-
 
 Add a runtime dep with `uv add <pkg>`, a dev/test dep with `uv add --dev <pkg>` (both update `pyproject.toml` + `uv.lock`).
 
+Backtrader is optional (`uv sync --extra backtrader`) and is not used by the
+current pipeline. Keep `portfolio simulate` as the canonical backtest/execution
+path unless an explicit future adapter is requested.
+
 Note: `tests/test_downloader.py` hits the network (live baostock/akshare); `tests/test_alpha.py` and `tests/test_portfolio.py` are pure and fast.
 
 ## Architecture: one decoupled pipeline
@@ -65,6 +69,8 @@ Data is stored **long/tidy**, not wide, as a Hive-partitioned dataset keyed by `
 
 `portfolio simulate` must execute `position_shares`, not continuous `target_shares`. It fills at the next available open and clips desired deltas through repeatable constraints (`suspension`, `price_limit`, `volume_cap`). `portfolio eval` uses `target_weight` for a continuous research view, so zero-gross carry dates remain flat there. Keep IC/IR out of portfolio metrics too.
 
+Trading cost uses the simplified open-execution proportional cash-cost model in `docs/portfolio_trading_cost_model.md`: `abs(traded_shares * open) * (cost_bps + slippage_bps) / 10000`. Slippage is cash cost only; do not also adjust execution prices for slippage.
+
 ## Alphas: factory + plugin pattern
 
 Each alpha is a class subclassing `BaseAlpha` (`pipeline/alpha/base.py`), living in its own module. It implements `signal(close) -> wide DataFrame` (the raw score); the base class's `to_weights` cross-sectionally z-scores that into position weights (override for custom normalization). Subclasses declare their own typed `__init__` params (e.g. `lookback`, `vol_window`, or anything custom).

README.md

@@ -48,6 +48,13 @@ The env is managed with [uv](https://docs.astral.sh/uv/). `uv sync` builds `.ven
 uv sync
 ```
 
+Backtrader is an optional dependency and is **not used by the current pipeline**.
+Install it only for future experiments or adapter work:
+
+```bash
+uv sync --extra backtrader
+```
+
 ## Quick start
 
 ```bash
@@ -113,6 +120,49 @@ partitions already written. Pass the **dataset directory** (`{output_dir}/{unive
 as `--data-path` to later phases — `pd.read_parquet` reads the whole partitioned
 set. Symbols use the internal `sh600000` / `sz000001` form (exchange prefix + code).
 
+### `derived` — daily custom/derived data
+
+Derived data is daily-only v1 research data keyed by `symbol_id,date`, with one
+or more numeric value columns. It can come from user CSV/parquet files or Python
+plugins, and is written as a single parquet file at `derived/{name}.pq`.
+
+The validator normalizes `date` to the trading day, requires unique
+`symbol_id,date` keys, rejects duplicate columns, and rejects non-numeric value
+columns. Alpha computation consumes derived data through the existing
+`--feature-path` flag.
+
+```bash
+# Validate a user file without writing output.
+uv run python cli.py derived validate --input-path vendor_factor.csv
+
+# Ingest CSV/parquet into the canonical derived/ layout.
+uv run python cli.py derived ingest \
+    --input-path vendor_factor.csv \
+    --derived-name vendor_factor
+
+# List built-in and external derived-data plugin types.
+uv run python cli.py derived list
+uv run python cli.py derived list --derived-module path/to/my_derived.py
+
+# Compute a derived file from daily and/or minute inputs.
+uv run python cli.py derived compute \
+    --minute-path data/minute_bars/sh600000 \
+    --daily-path data/daily_bars/sh600000 \
+    --derived-type minute_daily_summary \
+    --derived-name minute_summary
+
+# Join derived columns into a feature-aware alpha.
+uv run python cli.py alpha compute \
+    --data-path data/daily_bars/sh600000 \
+    --feature-path derived/minute_summary.pq \
+    --alpha-type my_feature_aware_alpha \
+    --alpha-name my_run
+```
+
+For compatibility, `feature list` and `feature compute` remain available and
+delegate to the same derived-data registry. Existing `features/*.pq` files are
+still valid `--feature-path` inputs when they satisfy the daily numeric contract.
+
 ### `alpha list` — show registered alpha types
 
 ```bash
@@ -130,6 +180,7 @@ uv run python cli.py alpha list --alpha-module path/to/my_alpha.py   # include a
 | `--output-dir` | `alphas` | Output directory |
 | `--lookback` | `5` | Lookback days (passed to alphas that accept it) |
 | `--vol-window` | `20` | Volatility window (passed to alphas that accept it) |
+| `--feature-path` | — | Daily derived/feature parquet file or dataset to left-join on `symbol_id,date`; repeatable |
 | `--alpha-module` | — | External module(s) to import first; repeatable. Dotted path or `.py` file |
 | `--param` | — | Extra constructor param as `name=value`; repeatable |
 
@@ -203,7 +254,9 @@ uv run python cli.py portfolio build \
 
 Executes the constructed `position_shares` book at the next available open,
 clipping trades through repeatable constraints. It writes `fills/<name>.pq` and
-`pnl/<name>.pq`.
+`pnl/<name>.pq`. Trading costs use the simplified open-execution proportional
+cash-cost model documented in
+[`docs/portfolio_trading_cost_model.md`](docs/portfolio_trading_cost_model.md).
 
 | Option | Default | Description |
 | --- | --- | --- |
@@ -361,6 +414,8 @@ between phases (data is stored long/tidy):
   OHLC scale under `qfq`/`hfq`; `turn` is turnover %, `pctChg` daily % change,
   `tradestatus`/`isST` are 0/1 flags, and `peTTM`/`pbMRQ`/`psTTM`/`pcfNcfTTM` are
   baostock valuation ratios.)
+- **derived** (`DERIVED_KEY_COLUMNS` + values): required keys `symbol_id, date`;
+  value columns are user/plugin-defined and must be numeric in v1.
 - **alpha** (`ALPHA_COLUMNS`): `symbol_id, date, alpha_name, weight`
 - **combo** (`COMBO_COLUMNS`): `symbol_id, date, combo_name, weight`
 - **portfolio positions** (`POSITION_COLUMNS`): `symbol_id, date, portfolio_name, target_weight, target_value, target_shares, position_shares, position_value, price`
@@ -378,8 +433,11 @@ directory yields an extra `month` (`YYYY-MM`) partition column on top of
 
 - `cli.py` — entry point wiring the file-based phases together
 - `pipeline/data/` — universe resolution + download → `data/daily_bars/{universe}/month=YYYY-MM/*.pq`
+- `pipeline/derived/` — daily derived-data ingestion, validation, plugin registry,
+  and built-in derived computations → `derived/*.pq`
 - `pipeline/alpha/` — `base.py` (`BaseAlpha`), `registry.py` (factory + plugin loader),
   `library/` (built-in alphas), `compute.py` (`compute_alpha` / `evaluate_alpha`)
+- `pipeline/features/` — compatibility wrappers for the derived-data registry
 - `pipeline/combo/` — alpha combination → `combos/*.pq`
 - `pipeline/portfolio/` — construction, A-share lot/limit rules, constraints,
   reference next-open simulator, and research metrics
@@ -401,9 +459,41 @@ constructed positions, fills/costs, P&L, and target-weight research metrics.
 - [x] **Reference execution simulation** — next-open fills over constructed
   `position_shares`, with suspension, price-limit, volume-cap, transaction-cost,
   and slippage controls.
+- [x] **Derived/custom daily data ("Level 2")** — ingest user CSV/parquet files
+  or compute plugin outputs as validated numeric daily datasets under
+  `derived/{name}.pq`; alpha joins continue through `--feature-path`.
+- [ ] **Optional Backtrader adapter** — Backtrader is available as the
+  `backtrader` extra for possible future event-driven/broker-style experiments,
+  but it is not part of the current canonical portfolio workflow.
 - [ ] **Forward / paper trading** — run the same construction logic on live
   daily data, track simulated fills and a running P&L without real capital.
 - [ ] **Intraday / microstructure data** — bid/ask prices & sizes, mid-price,
   and intraday VWAP. These need a tick / L1–L2 quote feed (typically a paid or
   brokerage data tier); the free daily sources here only expose daily bars, so
   this is a separate data phase rather than extra columns on the daily schema.
+
+### Additional TODOs
+
+The following items are intended extensions beyond the current daily
+alpha-to-portfolio pipeline:
+
+- **Long-only portfolio mode** — add a construction option that converts
+  alpha/combo weights into a long-only book while preserving existing lot,
+  price, suspension, and volume-cap handling.
+- **Index-short hedging mode** — support portfolios that hold long A-share
+  names while shorting an index or index proxy for market exposure control.
+- **Expanded universe presets** — add explicit universe aliases for CSI 300,
+  CSI 500, CSI 1000, and CSI 1800, while keeping file-based and comma-separated
+  custom universes available.
+- **Categorical derived data** — extend the numeric-only derived-data v1 contract
+  to support categorical inputs such as industry classifications. In this
+  project, "Level 2" means customized second-level research data produced by
+  users or plugins; it does not necessarily mean exchange order-book/L2 quote
+  feeds.
+- **Minute bar data** — continue extending the raw minute-bar and feature
+  workflow. The initial Baostock 5-minute download and daily feature plugin path
+  exist; intraday execution and replacing canonical daily bars remain out of
+  scope unless explicitly added later.
+- **Industry data** — add industry classification inputs for filtering,
+  grouping, exposure reporting, neutralization, or industry-aware portfolio
+  construction.

cli.py

@@ -3,7 +3,9 @@
 
 Phases:
     data      — Download daily bars to parquet
+    derived   — Ingest or compute daily derived data
     alpha     — Compute alpha weights from data
+    feature   — Compute daily features from minute bars
     combo     — Combine alphas into a single weight
     portfolio — Build tradable positions and simulate execution
 """
@@ -13,9 +15,12 @@ import logging
 import click
 
 from pipeline.data.cli import data
+from pipeline.derived.cli import derived
 from pipeline.alpha.cli import alpha
+from pipeline.features.cli import feature
 from pipeline.combo.cli import combo
 from pipeline.portfolio.cli import portfolio
+from plugins.joinquant.cli import joinquant
 from tools.pqcat import pqcat
 from tools.alphaview import alphaview
 
@@ -39,9 +44,12 @@ def cli(log_level):
 
 
 cli.add_command(data)
+cli.add_command(derived)
 cli.add_command(alpha)
+cli.add_command(feature)
 cli.add_command(combo)
 cli.add_command(portfolio)
+cli.add_command(joinquant)
 cli.add_command(pqcat)
 cli.add_command(alphaview)
 

data/downloader.py

@@ -31,11 +31,68 @@ _BATCH_COLUMNS = [
     "peTTM", "pbMRQ", "psTTM", "pcfNcfTTM",
 ]
 
+# Raw Baostock minute bars. The ``time`` field is usually compact
+# YYYYMMDDHHMMSSmmm; parsing below also tolerates HH:MM:SS strings in tests.
+_MINUTE_FIELDS = "date,time,code,open,high,low,close,volume,amount,adjustflag"
+_MINUTE_NUMERIC = ["open", "high", "low", "close", "volume", "amount"]
+_MINUTE_COLUMNS = [
+    "symbol", "datetime", "date", "time", "frequency",
+    "open", "high", "low", "close", "volume", "amount", "vwap", "adjustflag",
+]
+_MINUTE_FREQUENCIES = {"5", "15", "30", "60"}
+
 
 class _SessionLost(Exception):
     """baostock reported the session was dropped (``用户未登录``)."""
 
 
+def _normalize_minute_frequency(frequency: str | int) -> tuple[str, str]:
+    """Return Baostock frequency and partition label for a minute interval."""
+    raw = str(frequency).strip().lower()
+    if raw.endswith("m"):
+        raw = raw[:-1]
+    if raw not in _MINUTE_FREQUENCIES:
+        raise ValueError(
+            f"Unsupported minute frequency '{frequency}'. "
+            f"Expected one of {sorted(_MINUTE_FREQUENCIES)} minutes."
+        )
+    return raw, f"{raw}m"
+
+
+def _parse_minute_datetime(date: pd.Series, time: pd.Series) -> pd.Series:
+    """Parse Baostock minute timestamps into pandas datetimes."""
+    date_dt = pd.to_datetime(date, errors="coerce")
+    date_compact = date_dt.dt.strftime("%Y%m%d")
+    time_text = time.astype(str).str.strip()
+    time_digits = time_text.str.replace(r"\D", "", regex=True)
+
+    full_digits = time_digits.str.slice(0, 14)
+    from_full = pd.to_datetime(full_digits, format="%Y%m%d%H%M%S", errors="coerce")
+
+    from_short = pd.Series(pd.NaT, index=date.index, dtype="datetime64[ns]")
+    short_time = time_digits.str.len().between(1, 6)
+    if short_time.any():
+        short_digits = (
+            time_digits.loc[short_time]
+            .str.pad(6, side="right", fillchar="0")
+            .str.slice(0, 6)
+        )
+        from_short.loc[short_time] = pd.to_datetime(
+            date_compact.loc[short_time] + short_digits,
+            format="%Y%m%d%H%M%S",
+            errors="coerce",
+        )
+
+    from_text = pd.Series(pd.NaT, index=date.index, dtype="datetime64[ns]")
+    text_time = time_text.str.contains(":", regex=False)
+    if text_time.any():
+        from_text.loc[text_time] = pd.to_datetime(
+            date.astype(str).loc[text_time] + " " + time_text.loc[text_time],
+            errors="coerce",
+        )
+    return from_full.fillna(from_short).fillna(from_text)
+
+
 def _download_akshare(symbol: str, start: str, end: str, adjust: str = "qfq") -> Optional[pd.DataFrame]:
     """Download daily bars from akshare. Returns DataFrame with OHLCV columns."""
     try:
@@ -239,3 +296,104 @@ def download_daily_batch(
         except Exception:
             pass
 
+
+def download_minute_batch(
+    symbols: Iterable[str],
+    start: str,
+    end: str,
+    frequency: str | int = 5,
+    relogin_every: int = 200,
+) -> Iterator[Tuple[str, Optional[pd.DataFrame]]]:
+    """Download raw Baostock minute bars for many symbols.
+
+    Minute bars are intentionally unadjusted (`adjustflag='3'`) because the
+    output is raw intraday market data for downstream feature aggregation, not a
+    tradable daily price series.
+
+    Args:
+        symbols: Internal-form symbols (``sh600000`` / ``sz000001``).
+        start, end: ``YYYY-MM-DD`` bounds.
+        frequency: Baostock minute frequency. ``5``/``"5"``/``"5m"`` all mean
+            5-minute bars.
+        relogin_every: Proactively refresh the baostock session every N symbols.
+
+    Yields:
+        ``(symbol, df)`` where ``df`` has raw minute bars or ``None`` when no
+        data is available.
+    """
+    query_frequency, frequency_label = _normalize_minute_frequency(frequency)
+    adjustflag = _BAOSTOCK_ADJUST["none"]
+
+    def _relogin() -> None:
+        try:
+            bs.logout()
+        except Exception:
+            pass
+        bs.login()
+
+    def _fetch(symbol: str) -> Optional[pd.DataFrame]:
+        """One Baostock minute query; returns df, None, or raises _SessionLost."""
+        code = f"{symbol[:2]}.{symbol[2:]}"
+        rs = bs.query_history_k_data_plus(
+            code=code,
+            fields=_MINUTE_FIELDS,
+            start_date=start,
+            end_date=end,
+            frequency=query_frequency,
+            adjustflag=adjustflag,
+        )
+        if rs.error_code != "0":
+            if "未登录" in (rs.error_msg or ""):
+                raise _SessionLost(rs.error_msg)
+            logger.warning("baostock minute error for %s: %s", symbol, rs.error_msg)
+            return None
+        rows = []
+        while rs.next():
+            rows.append(rs.get_row_data())
+        if not rows:
+            return None
+
+        df = pd.DataFrame(rows, columns=_MINUTE_FIELDS.split(","))
+        df[_MINUTE_NUMERIC] = df[_MINUTE_NUMERIC].apply(pd.to_numeric, errors="coerce")
+        df["datetime"] = _parse_minute_datetime(df["date"], df["time"])
+        bad_timestamps = df["datetime"].isna()
+        if bad_timestamps.any():
+            raise ValueError(
+                f"Could not parse {int(bad_timestamps.sum())} minute timestamp(s)"
+            )
+        df["date"] = pd.to_datetime(df["date"], errors="coerce")
+        df["time"] = df["datetime"].dt.strftime("%H:%M:%S")
+        df["frequency"] = frequency_label
+        df["vwap"] = (df["amount"] / df["volume"]).where(df["volume"] > 0)
+        df["symbol"] = symbol
+        return df[_MINUTE_COLUMNS].sort_values("datetime").reset_index(drop=True)
+
+    bs.login()
+    try:
+        for i, symbol in enumerate(symbols):
+            if i and relogin_every and i % relogin_every == 0:
+                _relogin()
+
+            df: Optional[pd.DataFrame] = None
+            for attempt in (1, 2):
+                try:
+                    df = _fetch(symbol)
+                    break
+                except _SessionLost:
+                    if attempt == 1:
+                        _relogin()
+                        continue
+                    logger.warning("baostock minute session lost for %s after relogin", symbol)
+                except Exception as e:
+                    logger.warning("baostock minute download failed for %s: %s", symbol, e)
+                    break
+
+            if df is not None and not df.empty:
+                yield symbol, df
+            else:
+                yield symbol, None
+    finally:
+        try:
+            bs.logout()
+        except Exception:
+            pass

docs/joinquant_comparison_plugin.md

@@ -0,0 +1,352 @@
+# JoinQuant Comparison Plugin
+
+## Why a Plugin
+
+JoinQuant is an external execution and simulation reference. Keeping this code
+under `plugins/joinquant/` prevents vendor-specific assumptions from entering
+`pipeline/portfolio/`, where the internal reference simulator remains the
+canonical implementation.
+
+## What It Validates
+
+The comparison is for system correctness:
+
+- date alignment
+- internal to JoinQuant symbol mapping
+- target position generation
+- once-per-day open execution timing
+- lot rounding and filled shares
+- position carry
+- trading cost
+- PnL accounting
+- blocked trades from suspension, limit-up, and limit-down conditions
+
+## What It Does Not Validate
+
+It does not validate alpha quality, IC, IR, forecast skill, or whether the
+strategy is economically useful. Differences can be expected when JoinQuant
+uses different fee, slippage, cash, corporate-action, or internal rounding
+rules.
+
+## Historical Backtest Workflow
+
+```bash
+# 1. Build internal portfolio targets.
+uv run python cli.py portfolio build ...
+
+# 2. Export JoinQuant-compatible frozen targets.
+uv run python cli.py joinquant export-targets \
+  --positions-path portfolio/run1.pq \
+  --portfolio-name run1 \
+  --mode target_shares \
+  --execution-calendar-path data/daily_bars/<universe> \
+  --out-dir plugins_output/joinquant/targets
+
+# 3. Generate and copy the wrapper strategy and target files into JoinQuant.
+uv run python cli.py joinquant write-wrapper \
+  --portfolio-name run1 \
+  --mode target_shares \
+  --out-path plugins_output/joinquant/wrapper_strategy_run1.py
+
+# 4. Run the JoinQuant backtest or simulated trading job.
+# 5. Export JoinQuant fills, positions, and daily PnL to CSV.
+
+# 6. Ingest JoinQuant output.
+uv run python cli.py joinquant ingest \
+  --portfolio-name run1 \
+  --fills-csv path/to/jq_fills.csv \
+  --positions-csv path/to/jq_positions.csv \
+  --pnl-csv path/to/jq_pnl.csv
+
+# 7. Reconcile.
+uv run python cli.py joinquant reconcile \
+  --portfolio-name run1 \
+  --targets-dir plugins_output/joinquant/targets/run1 \
+  --our-fills-path fills/run1.pq \
+  --our-positions-path portfolio/run1.pq \
+  --our-pnl-path pnl/run1.pq \
+  --jq-fills-path plugins_output/joinquant/ingested/run1/fills.pq \
+  --jq-positions-path plugins_output/joinquant/ingested/run1/positions.pq \
+  --jq-pnl-path plugins_output/joinquant/ingested/run1/pnl.pq
+```
+
+## Forward-Testing Workflow
+
+After the T-1 close and after the data update:
+
+```bash
+uv run python cli.py portfolio build ...
+uv run python cli.py joinquant export-targets \
+  --positions-path portfolio/run1.pq \
+  --portfolio-name run1 \
+  --mode target_shares \
+  --execution-calendar-path data/daily_bars/<universe> \
+  --start-date T \
+  --end-date T
+```
+
+Before the T open, upload or expose the frozen target file to JoinQuant. During
+the T open, the JoinQuant wrapper reads that file and submits orders, while the
+internal simulator should run against the same frozen target. After T close or
+after JoinQuant results are available, ingest the JoinQuant CSV files and run
+`joinquant reconcile`.
+
+Forward target files must be frozen before execution. Do not regenerate a
+target file after observing open or close data for the same trading date. The
+exporter writes a snapshot JSON with a SHA-256 hash for this reason and refuses
+to overwrite existing target/snapshot files unless `--force` is passed.
+
+For comparisons against the internal simulator, pass
+`--execution-calendar-path` to `joinquant export-targets`. The positions file is
+dated by construction/signal date, while the simulator executes at the next
+available open. The calendar option shifts exported target files to that next
+trading date, so JoinQuant reads the same target on the same execution session.
+
+For JoinQuant 模拟盘, the browser automation has two operational phases:
+
+```bash
+# Before T open: upload frozen target(s), save strategy, and start/restart 模拟盘.
+uv run python cli.py joinquant write-browser-config \
+  --out-path /tmp/chinese-equity-quant-realdata/joinquant_sim_config.json \
+  --strategy-url "https://www.joinquant.com/<your 模拟盘 page>" \
+  --flow sim-trade
+
+uv run python cli.py joinquant run-browser-sim \
+  --manifest-path /tmp/chinese-equity-quant-realdata/joinquant_smoke_manifest.json \
+  --config-path /tmp/chinese-equity-quant-realdata/joinquant_sim_config.json \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json \
+  --headed
+
+# After T close: download/export JoinQuant fills, positions, and pnl, then
+# ingest/reconcile. The same run-browser-sim command can do this if the config
+# includes download actions, otherwise use the ingest/reconcile commands.
+```
+
+The default simulated-trading template includes selectors for saving the
+strategy and clicking simulated-trading controls such as `模拟盘`, `模拟交易`,
+`启动`, and `重启`. These selectors are intentionally configurable because the
+JoinQuant web UI can differ by account and page version.
+
+## Target-Shares Mode
+
+`target_shares` is the default and preferred correctness mode. The exported
+`target_shares` field comes from the internal `position_shares` column produced
+by `portfolio build`, because the internal simulator executes that discretized
+integer book. The generated wrapper calls:
+
+```python
+order_target(jq_symbol, target_shares)
+```
+
+This mode makes filled shares, position carry, and blocked trades easiest to
+compare.
+
+## Target-Value Mode
+
+`target_value` mode exports `target_value` and `target_weight` from the
+portfolio file. The generated wrapper calls:
+
+```python
+order_target_value(jq_symbol, target_value)
+```
+
+This can be useful for portfolio-level comparisons, but JoinQuant may apply its
+own rounding, cash, and lot rules. Differences are often classified as
+`JOINQUANT_INTERNAL_ROUNDING`, `LOT_ROUNDING`, or `CASH_CONSTRAINT` depending
+on the observed output.
+
+## Symbol Mapping
+
+Internal symbols are converted as follows:
+
+```text
+sh600000 -> 600000.XSHG
+sh688001 -> 688001.XSHG
+sz000001 -> 000001.XSHE
+sz300001 -> 300001.XSHE
+```
+
+Reverse mapping is also supported. Invalid exchanges or unsupported A-share
+prefixes raise `ValueError` instead of silently guessing.
+
+## Wrapper Strategy Usage
+
+Generate a configured wrapper:
+
+```bash
+uv run python cli.py joinquant write-wrapper \
+  --portfolio-name run1 \
+  --mode target_shares \
+  --out-path plugins_output/joinquant/wrapper_strategy_run1.py
+```
+
+Copy the generated file and daily CSV target files into JoinQuant. The default
+loader uses JoinQuant `read_file`, which works for uploaded files. If your
+JoinQuant runtime allows HTTP or another storage backend, replace only
+`_read_target_file()` in the generated strategy.
+
+The wrapper is long-only by default:
+
+```python
+ALLOW_SHORT = False
+```
+
+Negative targets are clipped to zero and logged. Use `--allow-short` only if
+the target JoinQuant account supports the required shorting mechanics.
+
+## Ingesting JoinQuant Outputs
+
+The ingest command accepts permissive CSV column names and writes strict plugin
+schemas:
+
+```text
+plugins_output/joinquant/ingested/{portfolio_name}/fills.pq
+plugins_output/joinquant/ingested/{portfolio_name}/positions.pq
+plugins_output/joinquant/ingested/{portfolio_name}/pnl.pq
+```
+
+Missing cost fields default to zero. Missing blocked status defaults to zero.
+Symbols and dates are normalized.
+
+## Reading Reconciliation Reports
+
+The reconcile command writes:
+
+```text
+plugins_output/joinquant/reconcile/{portfolio_name}/daily_reconcile.pq
+plugins_output/joinquant/reconcile/{portfolio_name}/summary.csv
+plugins_output/joinquant/reconcile/{portfolio_name}/summary.md
+```
+
+`daily_reconcile.pq` is per-symbol and includes target shares, internal filled
+shares, JoinQuant filled shares, realized positions, trade prices, costs, PnL,
+and a `diff_reason`. `summary.csv` is the daily portfolio-level view for gross
+exposure, net exposure, cash, total value, PnL, cumulative PnL, turnover, and
+cost.
+
+Difference reasons include:
+
+```text
+MATCH SYMBOL_MAPPING PRICE_MISMATCH LOT_ROUNDING SUSPENSION LIMIT_UP_BLOCK
+LIMIT_DOWN_BLOCK VOLUME_OR_LIQUIDITY COST_MODEL CASH_CONSTRAINT
+SHORT_NOT_SUPPORTED CORPORATE_ACTION JOINQUANT_INTERNAL_ROUNDING
+MISSING_IN_OUR_SYSTEM MISSING_IN_JOINQUANT UNKNOWN
+```
+
+Default tolerances are exact share matching, `1e-4` relative trade-price
+tolerance, and value tolerance `max(1 yuan, 1e-6 * booksize)`. PnL tolerance is
+configurable with `--pnl-tolerance`.
+
+## Minimal Example
+
+Create a 5-stock equal-weight or fixed-share test portfolio:
+
+```text
+sh600000, sz000001, sh600519, sz002594, sz300750
+```
+
+Build positions for a small date range, export `target_shares`, upload the CSV
+files and wrapper to JoinQuant, run the JoinQuant backtest, export fills,
+positions, and PnL, then run ingest and reconcile. Start with one or two days
+before expanding the sample.
+
+For the first one-stock long-only smoke test, the local side can be prepared in
+one command:
+
+```bash
+uv run python cli.py joinquant prepare-smoke \
+  --out-dir /tmp/chinese-equity-quant-realdata
+```
+
+The command downloads a tiny public daily-bar sample, builds a fixed-share
+`sh600000` long-only position file, simulates it internally, exports aligned
+JoinQuant target files, writes a configured wrapper strategy, and creates
+`joinquant_smoke_manifest.json` with all output paths.
+
+## Browser Backtest Automation
+
+JoinQuant's public `jqdatasdk` is a data SDK. It supports authenticated data
+calls such as `auth(username, password)` and `get_price(...)`, but cloud
+strategy upload, backtest execution, and result export are web-application
+workflows. The plugin therefore automates those remote steps through Playwright
+with a saved browser session.
+
+Install the optional browser runner in the uv environment:
+
+```bash
+uv sync --extra joinquant-browser
+uv run playwright install chromium
+```
+
+Save a reusable login state. This opens a browser; log in normally, including
+any CAPTCHA or 2FA, then press Enter in the terminal to save state:
+
+```bash
+uv run python cli.py joinquant browser-login \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json
+```
+
+Create a selector/action config for a historical backtest:
+
+```bash
+uv run python cli.py joinquant write-browser-config \
+  --out-path /tmp/chinese-equity-quant-realdata/joinquant_browser_config.json \
+  --strategy-url "https://www.joinquant.com/<your strategy page>" \
+  --flow backtest
+```
+
+If selectors need tuning, capture the logged-in page:
+
+```bash
+uv run python cli.py joinquant browser-snapshot \
+  --url "https://www.joinquant.com/<your strategy page>" \
+  --out-dir /tmp/chinese-equity-quant-realdata/browser_snapshot
+```
+
+Then run the remote backtest automation:
+
+```bash
+uv run python cli.py joinquant run-browser-backtest \
+  --manifest-path /tmp/chinese-equity-quant-realdata/joinquant_smoke_manifest.json \
+  --config-path /tmp/chinese-equity-quant-realdata/joinquant_browser_config.json \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json \
+  --headed
+```
+
+The config is declarative: actions can navigate, paste the generated wrapper,
+upload all target CSV files, fill dates, click run, wait for completion,
+download result CSVs, and take screenshots. When the configured downloads
+produce fills, positions, and PnL CSVs, the runner automatically calls
+`joinquant ingest` and `joinquant reconcile`.
+
+For forward testing / 模拟盘, create the config with `--flow sim-trade` and run:
+
+```bash
+uv run python cli.py joinquant run-browser-sim \
+  --manifest-path /tmp/chinese-equity-quant-realdata/joinquant_smoke_manifest.json \
+  --config-path /tmp/chinese-equity-quant-realdata/joinquant_sim_config.json \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json \
+  --headed
+```
+
+Do not store raw JoinQuant passwords in this repository. The browser state file
+is created with `0600` permissions and should live under `~/.config`, outside
+the repo.
+
+## Recommended First Sanity Checks
+
+1. One liquid stock with a fixed target share count.
+2. A 10-stock equal-weight long-only portfolio.
+3. A forced suspension, limit-up, and limit-down sample.
+4. A short target in long-only mode to confirm `SHORT_NOT_SUPPORTED`.
+5. A 5-day reversal portfolio after the mechanical checks pass.
+
+## Known Limitations
+
+- JoinQuant internal execution details may differ from the reference simulator.
+- External file loading depends on the JoinQuant environment.
+- Short selling may not be supported.
+- Fee, tax, slippage, and minimum-fee models may differ.
+- Corporate actions may need special handling and should not be hidden.
+- The internal simulator does not currently emit execution price in
+  `FILL_COLUMNS`; price reconciliation uses explicit price columns if supplied.

docs/joinquant_cost_model_findings.md

@@ -0,0 +1,167 @@
+# JoinQuant Cost Model Findings
+
+Generated: 2026-07-06
+
+This report summarizes the JoinQuant trading-cost behavior observed from the
+browser-automated real-data backtests and compares it with the current internal
+simulator model. The JoinQuant cost formula below is inferred from rendered
+transaction tables and strategy logs for this account. Treat it as an observed
+platform default, not as a guaranteed external contract.
+
+## Runs Used
+
+### Longer Comparison Run
+
+- Portfolio: `jq_long_one_stock_long`
+- Window: `2024-01-11` to `2024-02-29`
+- Booksize: `1,000,000 CNY`
+- Instrument: `600000.XSHG`
+- Target: buy and hold `1,000` shares
+- JoinQuant rendered result: completed
+- Local total PnL: `546.22 CNY`
+- JoinQuant total PnL from positions tab: `575.00 CNY`
+- Difference: `28.78 CNY`, or `0.002878` percentage points on the book
+
+The first JoinQuant transaction was:
+
+| Date | Side | Shares | Price | Turnover | Fee |
+|---|---:|---:|---:|---:|---:|
+| `2024-01-11` | Buy | `1,000` | `6.57` | `6,570.00` | `5.00` |
+
+That trade hit a minimum fee. The local simulator charged `6.1415 CNY` because
+the smoke runner used a flat `10 bps` cash cost on adjusted-price turnover.
+
+### Cost Probe Run
+
+- Portfolio: `jq_cost_probe_buy_sell`
+- Window: `2024-01-11` to `2024-01-12`
+- Booksize: `1,000,000 CNY`
+- Instrument: `600000.XSHG`
+- Targets:
+  - `2024-01-11`: buy `100,000` shares
+  - `2024-01-12`: sell to `0` shares
+
+Observed JoinQuant transactions:
+
+| Date | Side | Shares | Price | Turnover | Fee | Implied Fee |
+|---|---:|---:|---:|---:|---:|---:|
+| `2024-01-11` | Buy | `100,000` | `6.57` | `657,000.00` | `197.10` | `3.0000 bps` |
+| `2024-01-12` | Sell | `-100,000` | `6.51` | `651,000.00` | `846.30` | `13.0000 bps` |
+
+The transaction-table numbers match this formula exactly:
+
+```text
+buy fee  = max(5 CNY, turnover * 0.0003)
+sell fee = max(5 CNY, turnover * 0.0003) + turnover * 0.001
+```
+
+In basis points:
+
+- Buy commission: `3 bps`
+- Sell commission: `3 bps`
+- Sell stamp tax: `10 bps`
+- Minimum commission: `5 CNY`
+
+No separate transfer fee was visible in this probe. If a separate transfer fee
+was present as an additional charge, the observed fees would not match the
+formula above exactly. It may still be folded into JoinQuant's displayed
+commission field, so this finding should be read as "not separately observable"
+rather than "impossible".
+
+No slippage was visible. The wrapper submitted open-time market orders with
+`run_daily(..., time="open")`, and JoinQuant filled them at the displayed open
+prices.
+
+## Evidence From Strategy Logs
+
+The JoinQuant order log for the cost probe showed:
+
+```text
+2024-01-11 ... trade price: 6.57, amount:100000, commission: 197.1
+2024-01-12 ... trade price: 6.51, amount:100000, commission: 846.3
+```
+
+The normal transaction tab showed the same fee values. However, the generated
+wrapper's `JOINQUANT_FILL` log records had `trade_cost: 0.0`, even for those
+same fills. That means `get_trades()` did not expose the usable commission
+value through the field the wrapper currently reads.
+
+For reconciliation, use the transaction table or JoinQuant order logs for fee
+details. Do not rely on the wrapper's current `JOINQUANT_FILL.trade_cost`.
+
+## Difference From The Internal Simulator
+
+The current internal simulator cost model is
+`SimpleProportionalCostModel` in `pipeline/portfolio/costs.py`:
+
+```text
+trade_cost = abs(traded_shares * execution_price)
+             * (cost_bps + slippage_bps) / 10000
+```
+
+The smoke runner used:
+
+- `cost_bps = 5`
+- `slippage_bps = 5`
+- combined one-way cash cost: `10 bps`
+
+Important differences:
+
+- The internal simulator uses the same rate for buys and sells.
+- It has no minimum commission.
+- It has no sell-only stamp tax.
+- Slippage is modeled as an extra cash cost.
+- JoinQuant did not show slippage in the observed open-time fills.
+- The local smoke download used `adjust="qfq"`, while the JoinQuant wrapper set
+  `set_option("use_real_price", True)`. That price-scale mismatch also affects
+  PnL and cost comparisons.
+
+## Practical Implications
+
+For a buy-only smoke test, JoinQuant may charge less than the local model when
+the trade is large enough for `3 bps` to apply, but it may charge more on small
+orders because of the `5 CNY` minimum.
+
+For any test with sells, JoinQuant's default sell fee is materially higher than
+the current local flat model because of the inferred `10 bps` stamp tax.
+
+The earlier 30-day buy-and-hold discrepancy was small because only one buy was
+executed. A rebalancing strategy with many sells will show a larger cost-model
+difference unless the local simulator is configured to match JoinQuant.
+
+## Recommended Follow-Ups
+
+1. Add a JoinQuant-style cost model to the internal simulator:
+
+```text
+commission = max(min_commission, turnover * commission_bps / 10000)
+stamp_tax  = turnover * sell_stamp_tax_bps / 10000 for sells only
+trade_cost = commission + stamp_tax
+```
+
+2. Add a CLI option or preset for `portfolio simulate`, for example
+`--cost-model joinquant-stock`.
+
+3. Update JoinQuant reconciliation to parse fee values from the transaction
+table or order logs when CSV exports are unavailable.
+
+4. Run a second local-vs-JoinQuant comparison with:
+
+- raw or real-price local bars, not adjusted-price bars
+- JoinQuant-style costs
+- slippage disabled locally
+
+That test should isolate remaining differences to data alignment, price source,
+rounding, and JoinQuant internal execution behavior.
+
+## Local Artifacts
+
+The temporary artifacts from the investigation are:
+
+- `/tmp/chinese-equity-quant-jq-long/comparison_report.md`
+- `/tmp/chinese-equity-quant-jq-long/parsed_joinquant/daily_pnl_compare_from_positions_tab.csv`
+- `/tmp/chinese-equity-quant-jq-cost-probe/jq_cost_analysis_report.md`
+- `/tmp/chinese-equity-quant-jq-cost-probe/jq_cost_analysis_summary.json`
+- `/tmp/chinese-equity-quant-jq-cost-probe/jq_cost_probe_transactions_parsed.csv`
+- `/tmp/chinese-equity-quant-jq-cost-probe/detail_tabs/transactions.txt`
+

docs/minute_bar_data.md

@@ -0,0 +1,70 @@
+# Minute Bar Data Notes
+
+The minute-bar path downloads raw Baostock intraday bars and stores them as a
+Hive-partitioned dataset:
+
+```bash
+uv run python cli.py data download-minute \
+    --universe sh600000 \
+    --start-date 2024-01-02 --end-date 2024-01-05 \
+    --frequency 5
+```
+
+The default layout is:
+
+```text
+data/minute_bars/{universe}/frequency=5m/month=YYYY-MM/*.pq
+```
+
+Derived-data plugins can aggregate those bars to daily `symbol_id,date` numeric
+files, for example:
+
+```bash
+uv run python cli.py derived compute \
+    --minute-path data/minute_bars/sh600000 \
+    --daily-path data/daily_bars/sh600000 \
+    --derived-type minute_daily_summary \
+    --derived-name minute_summary
+```
+
+The legacy `feature compute` command delegates to the same derived-data
+registry and remains available for existing scripts.
+
+## Daily vs Minute Reconciliation
+
+Baostock's daily raw bars and 5-minute raw bars are close, but they should not
+be treated as perfectly reconstructible from each other.
+
+When checking consistency, compare daily raw bars (`data download --adjust none`)
+against minute bars on the same raw price scale. The minute aggregation should
+use:
+
+- `open`: first minute open
+- `high`: max minute high
+- `low`: min minute low
+- `close`: last minute close
+- `volume`: sum minute volume
+- `amount`: sum minute amount
+- `vwap`: `sum(amount) / sum(volume)`
+
+In a sanity check for `sh600000` from `2024-01-02` through `2024-01-05`, Baostock
+returned 4 daily rows and 192 5-minute bars, exactly 48 bars per day. Open, low,
+and close matched daily exactly on all 4 days. High matched on 3 of 4 days; on
+`2024-01-04`, the daily high was `6.67` while the max 5-minute high was `6.66`.
+Minute-summed volume and amount were higher than daily by roughly `0.16%` to
+`1.23%`. VWAP remained very close, with max relative difference around
+`0.0043%`.
+
+This appears to be a source-level Baostock reconciliation caveat, not a parser
+or ordering issue: the minute bars covered the regular `09:35:00` through
+`15:00:00` range and sorted correctly by timestamp.
+
+Practical guidance:
+
+- Use tolerance-based daily-vs-minute checks; do not require exact equality for
+  high, volume, or amount.
+- Expect open/close alignment to be a stronger sanity check than exact volume
+  reconstruction.
+- Use minute-derived values as separate daily features, not as replacements for
+  the canonical daily bar dataset unless a strategy explicitly wants that
+  source convention.

docs/portfolio_trading_cost_model.md

@@ -0,0 +1,140 @@
+# Portfolio Trading Cost Model
+
+This document describes the trading cost model used by `portfolio simulate`.
+The current implementation is a simplified open-execution proportional cost
+model. It is intentionally small, explicit, and easy to audit.
+
+## Open-Execution Timeline
+
+The simulator runs once per trading day:
+
+1. A constructed portfolio row provides the target book for an execution date.
+   In the current file layout, a target dated `t` is executed at the next
+   available market date `d = next(t)`.
+2. Trades are executed at `open[d]`.
+3. Realized positions are held during the trading day.
+4. Daily PnL is marked from `open[d]` to `close[d]` on the newly realized book,
+   plus any overnight gap from the previous realized holdings.
+5. Trading cost is charged only on actually realized `traded_shares`, after all
+   constraints have clipped the desired trade.
+
+This means a fully blocked order has `traded_shares = 0` and therefore zero
+trading cost.
+
+## Current Formula
+
+For each symbol:
+
+```text
+trade_value_i = abs(traded_shares_i * execution_price_i)
+trade_cost_i  = trade_value_i * (cost_bps + slippage_bps) / 10000
+```
+
+where:
+
+```text
+execution_price_i = open_price_i
+```
+
+`cost_bps` is the proportional explicit trading-cost rate in basis points.
+`slippage_bps` is modeled as an additional cash cost in basis points. The two
+rates are added linearly. The CLI options `--cost-bps` and `--slippage-bps`
+both default to `0.0`.
+
+Both rates are **one-way, per-trade**: the combined `(cost_bps + slippage_bps)`
+is charged on the traded notional of *each* fill, buy and sell alike. A full
+round trip (enter then exit a position) is therefore charged twice — e.g.
+`5 + 5` bps becomes ~20 bps over a complete round trip, not 10. Quote any
+round-trip figure by doubling, or convert a round-trip budget to a per-trade
+rate by halving before passing it in.
+
+Example:
+
+```text
+traded_shares = 1000
+execution_price = 20 yuan
+cost_bps = 10
+slippage_bps = 5
+
+abs(1000 * 20) * 15 / 10000 = 30 yuan
+```
+
+## Slippage Convention
+
+Slippage is not applied by changing the execution price. It is charged only as
+a cash cost through `trade_cost`.
+
+Do not double-count slippage by doing both:
+
+```text
+execution_price = open * (1 +/- slippage_bps / 10000)
+trade_cost += trade_value * slippage_bps / 10000
+```
+
+The simulator should execute at the open price and subtract the slippage cash
+cost from PnL.
+
+## Relationship To The Simulator
+
+`ReferenceSimulator.fill()` clips desired trades through constraints first, then
+passes the actual `traded_shares` to the cost model. The per-name result is
+stored in the fills parquet as `trade_cost`.
+
+`ReferenceSimulator.run()` sums per-name `trade_cost` into the daily PnL row's
+`cost` column and subtracts that total from daily PnL:
+
+```text
+pnl = overnight + intraday - cost_total
+```
+
+## What This Model Does Not Cover
+
+The current model intentionally does not model:
+
+- Minimum commissions.
+- Buy/sell asymmetric fees.
+- Sell-side stamp duty.
+- Exchange handling fees.
+- Regulatory fees.
+- Transfer fees.
+- Date-aware fee schedule changes.
+- Nonlinear price impact.
+- Auction liquidity / queue effects.
+- Partial fills caused by open auction depth.
+
+These omissions are deliberate. The current model is the default reference
+model, not a detailed brokerage fee simulator.
+
+## Future Extension
+
+The simulator is structured around a cost model abstraction:
+
+```python
+class CostModel:
+    def compute(
+        self,
+        traded_shares,
+        execution_price,
+        side,
+        date,
+        metadata,
+    ):
+        ...
+```
+
+The current implementation is `SimpleProportionalCostModel`.
+
+A future `AShareDetailedCostModel` can add:
+
+- Commission, optionally subject to minimum commission.
+- Sell-side stamp duty.
+- Transfer fee.
+- Exchange handling fee.
+- Regulatory fee.
+- Date-aware fee rates.
+- Separate buy-side and sell-side rates.
+- Optional nonlinear slippage / market-impact model.
+
+Any future model must preserve the same high-level simulator contract: costs
+are computed from realized trades after constraints, and slippage must not be
+counted both through execution-price adjustment and cash cost.

docs/reversal_5d_all_universe_pipeline_report.md

@@ -0,0 +1,509 @@
+# Tutorial: Testing a 5-Day Reversal Alpha
+
+This document is a teaching walkthrough for someone who is new to this research
+framework and only lightly familiar with quant research. We will use one
+concrete experiment, a 5-day reversal alpha on the full downloaded Chinese
+A-share universe, to learn how the framework defines an alpha, stores it, tests
+it, turns it into a portfolio, and explains the gap between a research result
+and simulated trading PnL.
+
+This generated version was refreshed at 2026-06-12T22:52:56.
+The important point is not the timestamp; it is the research method.
+
+## The Research Question
+
+A quant research project starts with a hypothesis:
+
+> If a stock fell a lot over the last few trading days, it may rebound soon; if
+> it rose a lot, it may cool off soon.
+
+This is called **short-horizon reversal**. It is a simple idea: recent losers
+are candidates to buy, and recent winners are candidates to sell or underweight.
+In this repo, the tested version looks back 5 trading days.
+
+The central research question is:
+
+> Does this 5-day reversal rule create useful portfolio returns after the
+> framework applies realistic storage, portfolio construction, execution
+> constraints, and trading costs?
+
+The answer from this run is nuanced:
+
+- The naive built-in version is positive under the tradable
+  next-open-to-next-open research convention (**41.40%**),
+  but its stored weights still show that raw z-score weighting is too sensitive
+  to A-share outliers.
+- A rank-weighted version on a liquid, non-ST, tradable universe has a positive
+  costless research result: **209.58%**
+  at Sharpe **1.44**.
+- The daily-traded implementation is still not tradable after costs because
+  turnover is too high.
+
+That is a normal research outcome. Good research is not just asking "did the
+backtest go up?" It is asking **which layer explains the result**: signal,
+weighting, universe, construction, execution, or cost.
+
+## How This Framework Defines An Alpha
+
+In many quant textbooks, an alpha is described as a **prediction** of future
+returns. This framework uses a stricter and more practical convention:
+
+> An alpha is a signed cross-sectional position weight.
+
+That sentence is the key to the whole repo.
+
+- **Signed** means positive values are long exposure and negative values are
+  short exposure.
+- **Cross-sectional** means the alpha compares stocks to other stocks on the
+  same date.
+- **Position weight** means the output is already an instruction about what the
+  portfolio wants to own. It is not merely a score to correlate with future
+  returns.
+
+The stored alpha file always has this schema:
+
+| column | meaning |
+| --- | --- |
+| `symbol_id` | Stock identifier such as `sh600000` or `sz000001`. |
+| `date` | The signal date. The alpha is formed using information known by this date's close. |
+| `alpha_name` | A label for this particular run, such as `reversal_5d_all`. |
+| `weight` | Signed desired exposure. Positive means long; negative means short. |
+
+Because the framework treats alphas as position weights, it evaluates them with
+portfolio metrics: return, Sharpe, turnover, drawdown, and hit rate. It does
+**not** use IC/IR, because IC/IR would treat the alpha as a return predictor.
+
+## The Pipeline In One Picture
+
+Every phase reads parquet files and writes parquet files. That makes the system
+easy to inspect and rerun one layer at a time.
+
+```text
+daily bars
+   -> alpha weights
+   -> combined weights
+   -> portfolio targets and integer positions
+   -> simulated fills and PnL
+   -> evaluation metrics
+```
+
+For this experiment, the important phases are:
+
+| phase | command family | what it teaches you |
+| --- | --- | --- |
+| Data | `cli.py data download` | What market data is available. |
+| Alpha compute | `cli.py alpha compute` | How a raw research idea becomes stored weights. |
+| Alpha eval | `cli.py alpha eval` | How close-formed weights perform over the tradable next-open-to-next-open interval. |
+| Combo | `cli.py combo combine` | How one or more alphas become one combined book. |
+| Portfolio build | `cli.py portfolio build` | How weights become target values and integer shares. |
+| Portfolio simulate | `cli.py portfolio simulate` | How the integer book trades at next open with constraints and costs. |
+| Portfolio eval | `cli.py portfolio eval` | How the continuous target portfolio behaves over the same costless open-to-open research interval. |
+
+In a real research workflow, you should learn to pause after every phase and
+inspect the parquet output. Most mistakes are easier to find at the interface
+between two phases than at the final PnL line.
+
+## Step 1: Define The Raw Reversal Signal
+
+The built-in 5-day reversal alpha is implemented as:
+
+```python
+signal = -close.pct_change(5, fill_method=None)
+```
+
+For stock `i` on date `t`, this is approximately:
+
+```text
+signal[i, t] = -(close[i, t] / close[i, t-5] - 1)
+```
+
+So:
+
+- If a stock rose by 10% over the last 5 trading days, the raw signal is `-10%`.
+  It becomes a candidate short or underweight.
+- If a stock fell by 10% over the last 5 trading days, the raw signal is `+10%`.
+  It becomes a candidate long or overweight.
+
+Notice the timing. The signal uses prices through date `t`. It must not use the
+return from `t` to `t+1`, because that is the future. The costless alpha
+evaluator tests the weight formed on date `t` over the tradable interval from
+`open[t+1]` to `open[t+2]`; the later execution simulator is the separate layer
+that trades the constructed integer book at the next open.
+
+The code lives in `pipeline/alpha/library/reversal.py`:
+
+```python
+class ReversalAlpha(BaseAlpha):
+    name = "reversal"
+
+    def __init__(self, lookback: int = 5):
+        self.lookback = lookback
+
+    def signal(self, close: pd.DataFrame) -> pd.DataFrame:
+        return -close.pct_change(self.lookback, fill_method=None)
+```
+
+The alpha class only defines the raw signal. The base class then turns that
+signal into weights.
+
+## Step 2: Turn A Signal Into Cross-Sectional Weights
+
+By default, `BaseAlpha.to_weights()` does a cross-sectional z-score each date:
+
+```text
+weight[i, t] = (signal[i, t] - mean_signal[t]) / std_signal[t]
+```
+
+This means the framework asks:
+
+> On this date, which stocks have stronger reversal scores than the rest of the
+> market, and by how much?
+
+That is useful, but it has a weakness. If a few stocks have extreme trailing
+returns because they are newly listed, suspended, illiquid, or limit-constrained,
+z-scoring can put a very large amount of relative exposure into exactly those
+names.
+
+That is visible in the naive full-universe run. Stored weights reached about
+`-52` standard deviations. The
+result is positive under the open-to-open convention, but it is much weaker and
+less robust than the rank-weighted versions:
+
+| run | weighting | research cumulative return | research Sharpe | research turnover/year |
+| --- | --- | --- | --- | --- |
+| naive z-score, full universe | z-score | 41.40% | 0.4514 | 160x |
+
+The lesson is not "reversal is solved." The lesson is:
+
+> The same raw signal can become a fragile portfolio if the weighting method
+> reacts badly to outliers.
+
+## Step 3: Make The Weighting More Robust
+
+The repo also has a rank-weighted version, `reversal_rank`. It uses the same raw
+5-day reversal signal, but converts the cross-section to ranks instead of
+z-scores:
+
+```python
+ranks = signal.rank(axis=1)
+weights = ranks.subtract(ranks.mean(axis=1), axis=0)
+```
+
+Rank weighting keeps the ordering of stocks but removes the importance of the
+exact outlier magnitude. A stock can be "the worst recent loser" or "the best
+recent winner," but it cannot become dozens of standard deviations important
+just because its raw percentage move is unusual.
+
+The full-universe rank version was much less pathological, but still not a
+clean signal:
+
+| run | weighting | research cumulative return | research Sharpe | research turnover/year |
+| --- | --- | --- | --- | --- |
+| rank, full universe | rank | 73.44% | 0.8860 | 143x |
+
+That tells us the weighting fix helped, but the universe still contains many
+names that are poor candidates for a daily reversal strategy.
+
+## Step 4: Define The Investable Universe
+
+An alpha should be tested on stocks that could plausibly be traded. The liquid
+run applies a per-date mask before weights are created. A stock must be:
+
+- seasoned, with at least 60 observed closes;
+- currently tradable, using `tradestatus == 1`;
+- not ST, using `isST == 0`;
+- inside the top 1000 names by trailing 20-day average traded amount.
+
+This mask is applied to the signal, not to the price history used to compute the
+5-day return. That distinction matters. We still compute `pct_change(5)` on the
+full contiguous price history, then decide which names are eligible to hold on
+each signal date.
+
+The liquid rank result is the cleanest research result:
+
+| run | weighting | universe | research cumulative return | research Sharpe | hit rate |
+| --- | --- | --- | --- | --- | --- |
+| rank, liquid subset | rank | top 1000 liquid, tradable, non-ST | 209.58% | 1.4422 | 55.68% |
+
+This is the first point where a researcher can say:
+
+> There appears to be a real 5-day reversal effect in a cleaner A-share
+> universe, before trading costs.
+
+That last phrase, **before trading costs**, is essential.
+
+![Research equity](assets/reversal_5d_research_equity.png)
+
+When reading this chart, focus on the shape and relative behavior:
+
+- The naive z-score line shows why outlier-sensitive weighting is fragile.
+- The rank full-universe line shows that robust weighting helps, but the full
+  universe still contains noisy and hard-to-trade names.
+- The liquid rank line shows the signal-level edge before execution costs.
+
+## Step 5: Check That The Alpha File Is Sane
+
+Before trusting any metric, inspect the stored alpha artifact. The run checked:
+
+- The columns match `ALPHA_COLUMNS`.
+- There are no null weights.
+- There are no non-finite weights.
+- There are no duplicate `(symbol_id, date)` rows.
+- The daily cross-sectional mean is approximately zero.
+- A one-alpha combo is an exact identity transform.
+
+| run | schema ok | null w | non-finite w | dup keys | max \|daily mean\| | weight range | combo identity Δ |
+| --- | --- | --- | --- | --- | --- | --- | --- |
+| naive z-score (full) | True | 0 | 0 | 0 | 3.32e-16 | [-52.2, 19.2] | 0.00e+00 |
+| rank (full) | True | 0 | 0 | 0 | 0.00e+00 | [-2603.0, 2603.0] | 0.00e+00 |
+| rank (liquid subset) | True | 0 | 0 | 0 | 0.00e+00 | [-498.5, 498.5] | 0.00e+00 |
+
+The rank ranges look numerically large because rank weights scale with the
+number of names. That is fine: later evaluation divides by gross exposure, and
+portfolio construction normalizes by `sum(abs(weight))`. The important
+difference is that rank weights are bounded by cross-sectional rank, not by the
+raw size of an abnormal stock move.
+
+![Weight distributions](assets/reversal_5d_weight_distributions.png)
+
+This is a good habit: when a backtest looks strange, plot the weights before
+debugging the PnL. A broken or concentrated weight distribution often explains
+the result.
+
+## Step 6: Understand The Alpha Evaluation Formula
+
+The costless alpha evaluator now asks:
+
+> If we compute alpha weights after close on date `t`, trade them at `open[t+1]`,
+> and hold them until `open[t+2]`, what return would we earn before costs?
+
+This is still a **research-layer approximation**, not the trading simulator. At
+this stage the framework has only an alpha weight file. It has not yet rounded
+shares, checked limits, clipped trades, or paid costs. The purpose is to answer
+a clean signal question: "Do these close-formed weights line up with returns
+over the interval we could actually own after next-open execution?"
+
+The daily research return is:
+
+```text
+R[t] = sum_i(weight[i, t] * (open[i, t+2] / open[i, t+1] - 1)) / sum_i(abs(weight[i, t]))
+```
+
+This has three important consequences:
+
+- The alpha is normalized by its gross exposure, so the scale of raw weights
+  does not by itself create a higher return.
+- The new signal does not receive credit for the overnight gap from `close[t]`
+  to `open[t+1]`, because it cannot be traded until `open[t+1]`.
+- The final two signal dates are dropped from performance metrics because they
+  do not have a complete next-open-to-next-open holding interval.
+
+Turnover is still measured from the weights:
+
+```text
+turnover[t] = sum_i(abs(weight[i, t] - weight[i, t-1])) / sum_i(abs(weight[i, t-1]))
+```
+
+The annualized turnover numbers are a warning. Even a positive signal can be
+hard to monetize if it asks the portfolio to trade too much every day.
+
+## Step 7: Build A Portfolio From The Alpha
+
+The alpha file is still an abstract research book. `portfolio build` turns it
+into target exposures and integer shares.
+
+The main normalization is:
+
+```text
+target_weight[i, t] = weight[i, t] / sum_i(abs(weight[i, t]))
+target_value[i, t] = booksize * target_weight[i, t]
+target_shares[i, t] = target_value[i, t] / construction_price[i, t]
+```
+
+Then the framework creates an integer A-share book using lot rules and repair
+logic. This is where a research portfolio starts to become a tradable portfolio.
+
+The continuous target portfolio matched the stored alpha almost exactly:
+
+| run | target_value identity max\|Δ\| | alpha→target max\|Δ\| | research corr(alpha,portfolio) | mean integer gross | mean L1 tracking |
+| --- | --- | --- | --- | --- | --- |
+| naive z-score (full) | 0.0000 | 0.00e+00 | 1.000000 | 9,138,331 | 2,542,655 |
+| rank (full) | 0.0000 | 0.00e+00 | 1.000000 | 8,984,098 | 2,678,278 |
+| rank (liquid subset) | 0.0000 | 0.00e+00 | 1.000000 | 9,810,256 | 862,303 |
+
+The integer book is not exact because small target positions can be rounded
+away. The liquid subset has lower tracking error because it spreads the book
+over fewer and more tradable names.
+
+![Portfolio tracking](assets/reversal_5d_portfolio_tracking.png)
+
+When you research a new alpha, ask two separate questions:
+
+- Does the continuous target portfolio match the alpha? It should.
+- Does the integer tradable portfolio still resemble the target? It may not,
+  especially for small books or very broad universes.
+
+## Step 8: Simulate Execution And Costs
+
+Research returns are not the same as tradable PnL. The simulator executes the
+integer `position_shares` at the next available open and applies constraints:
+
+- suspension;
+- price limit;
+- volume cap;
+- proportional trading cost.
+
+The cost model is:
+
+```text
+cost = abs(traded_shares * open) * (cost_bps + slippage_bps) / 10000
+```
+
+For this run, cost is 5 bps commission plus 5 bps slippage. Slippage is treated
+as cash cost, not as an additional execution price adjustment.
+
+The execution results explain the final research conclusion:
+
+| run | corr(alpha, exec net) | PnL before cost | total cost | net PnL | mean daily turnover |
+| --- | --- | --- | --- | --- | --- |
+| naive z-score (full) | 0.8956 | 1,838,974 | 13,032,720 | -11,193,746 | 0.5711 |
+| rank (full) | 0.9126 | 5,052,067 | 11,713,451 | -6,661,383 | 0.5133 |
+| rank (liquid subset) | 0.8884 | 11,017,842 | 12,733,803 | -1,715,960 | 0.5715 |
+
+For the liquid rank run, simulated PnL before cost is about
+11,017,842, but total cost is about
+12,733,803. That is why the final net PnL is
+weak or negative.
+
+This is not a contradiction. It is exactly what a research pipeline should show:
+
+> The signal can exist in the costless layer, but the daily implementation can
+> still trade too much to keep the edge.
+
+![Execution vs research](assets/reversal_5d_exec_vs_research.png)
+
+## Step 9: Read The Headline Metrics Like A Researcher
+
+The complete summary is:
+
+| run | weighting | research cum | research Sharpe | research turn/yr | exec before cost | exec net | exec net Sharpe |
+| --- | --- | --- | --- | --- | --- | --- | --- |
+| naive z-score (full) | z-score | 41.40% | 0.4514 | 160× | 18.39% | -111.94% | -1.4508 |
+| rank (full) | rank | 73.44% | 0.8860 | 143× | 50.52% | -66.61% | -1.1839 |
+| rank (liquid subset) | rank | 209.58% | 1.4422 | 148× | 110.18% | -17.16% | -0.2226 |
+
+*Research = costless, no-look-ahead weights over the next-open-to-next-open
+holding interval. Execution = next-open fills on the discretized integer book
+under suspension / price-limit / volume-cap constraints, 5 bps commission + 5
+bps slippage.*
+
+Here is the interpretation:
+
+- **Naive z-score full universe**: positive under open-to-open research, but a
+  less reliable test of the reversal idea because the weighting scheme lets
+  outliers dominate parts of the book.
+- **Rank full universe**: a better test of the same idea, but still noisy
+  because the universe includes too many problematic names.
+- **Rank liquid subset**: the best signal-level test; it finds the cleanest
+  costless reversal effect.
+- **Execution net**: daily rebalancing remains heavily constrained by cost.
+
+A beginner might look only at the final net PnL and say "the alpha failed." A
+researcher should be more precise:
+
+> The raw 5-day reversal idea has signal value in a liquid universe, but the
+> current daily trading rule has too much turnover for the assumed cost model.
+
+## Step 10: Reproduce The Experiment
+
+These commands reproduce the important artifacts, assuming the full daily-bar
+dataset already exists at `data/daily_bars/all`.
+
+```bash
+# Naive z-score baseline: built-in reversal alpha, full universe.
+uv run python cli.py alpha compute --data-path data/daily_bars/all \
+    --alpha-name reversal_5d_all --alpha-type reversal --lookback 5 \
+    --output-dir alphas
+
+# Rank-weighted full and liquid runs.
+bash scripts/run_reversal_rank_e2e.sh
+
+# Regenerate figures, diagnostics, and this tutorial report.
+uv run python scripts/generate_reversal_5d_report.py
+```
+
+If you are learning the framework, do not run the whole pipeline blindly. Run
+one phase, inspect the output parquet, then continue.
+
+## How To Research Your Own Alpha
+
+Use this checklist for a new idea.
+
+1. State the hypothesis in plain language.
+   Example: "Stocks with poor 5-day returns may rebound over the next day."
+
+2. Write the raw signal.
+   Implement `signal(close) -> wide DataFrame` in an alpha class. Higher values
+   should mean stronger long preference.
+
+3. Choose the weighting method.
+   The default z-score is useful, but it can be fragile. Consider rank weights,
+   caps, neutralization, or liquidity-aware filters if outliers dominate.
+
+4. Define the investable universe before trusting results.
+   Make sure the strategy is not depending on suspended, ST, newly listed, or
+   illiquid names.
+
+5. Evaluate the alpha as a portfolio, not as a prediction.
+   Check cumulative return, Sharpe, drawdown, hit rate, and turnover over the
+   next-open-to-next-open holding interval. Do not add IC/IR unless the
+   framework's alpha convention changes.
+
+6. Build the portfolio and inspect tracking.
+   Confirm that target weights match the alpha, then check whether integer
+   shares still track the target book.
+
+7. Simulate execution with costs.
+   The final research question is not only "is there a signal?" It is "is there
+   enough signal left after realistic trading?"
+
+8. Diagnose the failure layer.
+   If results are bad, identify whether the problem is the raw signal, weighting,
+   universe, construction, execution constraints, turnover, or cost.
+
+For this 5-day reversal study, the diagnosis is clear: **the signal-level result
+is most promising after robust weighting and a liquid universe filter, but the
+current implementation needs turnover control before it can be considered
+tradable.**
+
+## Next Research Directions
+
+The natural next experiments are:
+
+- Add turnover control: no-trade bands, slower rebalancing, or weight smoothing.
+- Sweep the lookback window: compare 3-day, 5-day, 10-day, and 20-day reversal.
+- Sweep liquidity filters: top 500, top 1000, top 1500 by traded amount.
+- Add position caps so no single name can dominate after normalization.
+- Compare rank weighting with volatility-scaled reversal.
+
+The most important habit is to keep the layers separate. A good alpha research
+workflow does not stop at a single performance number; it explains how the idea
+travels from hypothesis, to signal, to weights, to portfolio, to executable PnL.
+
+## Appendix: Phase Timings From This Rerun
+
+| phase | rank full (s) | rank liquid (s) |
+| --- | --- | --- |
+| alpha compute | 94.1 | 107.8 |
+| alpha eval | 93.3 | 96.9 |
+| combo combine | 21.6 | 21.7 |
+| portfolio build | 537.6 | 236.7 |
+| portfolio eval | 95.1 | 88.3 |
+| portfolio simulate | 139.6 | 139.1 |
+| total | 981.3 | 690.5 |
+
+![Phase timings](assets/reversal_5d_phase_timings.png)
+
+`portfolio build` usually dominates because it iterates per signal date and
+repairs a multi-thousand-name integer book under lot rules. The liquid run is
+faster because it carries fewer non-zero names per date.

pipeline/alpha/base.py

@@ -5,7 +5,7 @@ position weights. Subclasses implement :meth:`signal` — the raw, unnormalized
 score. The base class turns a signal into cross-sectionally z-scored weights
 via :meth:`to_weights` (override it for a different normalization).
 """
-from abc import ABC, abstractmethod
+from abc import ABC
 
 import numpy as np
 import pandas as pd
@@ -15,15 +15,14 @@ class BaseAlpha(ABC):
     """A position-weight alpha over a cross-section of stocks.
 
     Concrete subclasses must set a unique class-level :attr:`name` (the registry
-    key) and implement :meth:`signal`. Construct subclasses with their own typed
+    key) and implement either :meth:`signal` or :meth:`signal_from_data`.
-    parameters (e.g. ``lookback``); the factory passes only the parameters a
+    Construct subclasses with their own typed parameters (e.g. ``lookback``);
-    given ``__init__`` accepts.
+    the factory passes only the parameters a given ``__init__`` accepts.
     """
 
     #: Unique registry key. Every concrete alpha must set this to a non-empty str.
     name: str = ""
 
-    @abstractmethod
     def signal(self, close: pd.DataFrame) -> pd.DataFrame:
         """Compute the raw signal.
 
@@ -34,6 +33,21 @@ class BaseAlpha(ABC):
             A wide DataFrame aligned to ``close`` where higher values indicate a
             stronger long. Use NaN where the signal is undefined.
         """
+        raise NotImplementedError(
+            f"{type(self).__name__} must implement signal() or signal_from_data()"
+        )
+
+    def signal_from_data(
+        self,
+        data: pd.DataFrame,
+        close: pd.DataFrame,
+    ) -> pd.DataFrame:
+        """Compute the raw signal from long daily data plus wide closes.
+
+        Feature-aware alphas can override this to pivot joined feature columns
+        from ``data``. The default preserves the existing close-only alpha API.
+        """
+        return self.signal(close)
 
     def to_weights(self, signal: pd.DataFrame) -> pd.DataFrame:
         """Cross-sectionally z-score a signal into signed position weights.

pipeline/alpha/cli.py

@@ -56,6 +56,10 @@ def list_(alpha_modules):
 @click.option("--output-dir", default="alphas", help="Directory to save alpha parquet")
 @click.option("--lookback", default=5, type=int, help="Lookback days")
 @click.option("--vol-window", default=20, type=int, help="Volatility window (reversal_vol only)")
+@click.option(
+    "--feature-path", "feature_paths", multiple=True,
+    help="Daily derived/feature parquet file or dataset to left-join on symbol_id,date (repeatable)",
+)
 @click.option(
     "--alpha-module", "alpha_modules", multiple=True,
     help="External module(s) to import so their alphas register (dotted path or .py file)",
@@ -64,8 +68,17 @@ def list_(alpha_modules):
     "--param", "extra_params", multiple=True,
     help="Extra alpha constructor param as name=value (repeatable)",
 )
+@click.option(
+    "--liquid-universe", is_flag=True, default=False,
+    help="Mask weights to a per-date investable universe (tradable, non-ST, "
+         "seasoned, top liquidity) before normalization",
+)
+@click.option(
+    "--universe-top-n", default=1000, type=int,
+    help="Most-liquid names kept per date when --liquid-universe is set",
+)
 def compute(data_path, alpha_name, alpha_type, output_dir, lookback, vol_window,
-            alpha_modules, extra_params):
+            feature_paths, alpha_modules, extra_params, liquid_universe, universe_top_n):
     """Compute one alpha from raw data and save as parquet."""
     for spec in alpha_modules:
         load_alpha_module(spec)
@@ -81,6 +94,8 @@ def compute(data_path, alpha_name, alpha_type, output_dir, lookback, vol_window,
     params = {"lookback": lookback, "vol_window": vol_window}
     params.update(_parse_params(extra_params))
 
+    universe = {"top_n": universe_top_n} if liquid_universe else None
+
     data = pd.read_parquet(data_path)
     click.echo(f"Loaded data: {len(data):,} rows from {data_path}")
 
@@ -88,6 +103,8 @@ def compute(data_path, alpha_name, alpha_type, output_dir, lookback, vol_window,
         data=data,
         alpha_name=alpha_name,
         alpha_type=alpha_type,
+        universe=universe,
+        feature_paths=feature_paths,
         **params,
     )
 
@@ -143,7 +160,8 @@ def reversal_vol(data_path, output_dir, lookback, vol_window):
 @alpha.command("eval")
 @click.option("--alpha-path", required=True, help="Path to alpha parquet file")
 @click.option("--data-path", required=True, help="Path to data parquet (for price data)")
-def eval_(alpha_path, data_path):
+@click.option("--report-dir", default="reports", help="Directory to save JSON report")
+def eval_(alpha_path, data_path, report_dir):
     """Evaluate an alpha's performance (return, Sharpe, turnover).
 
     Alphas are interpreted as position WEIGHTS, not return predictors.
@@ -166,9 +184,9 @@ def eval_(alpha_path, data_path):
     click.echo("=" * 50)
 
     # Also dump JSON
-    os.makedirs("reports", exist_ok=True)
+    os.makedirs(report_dir, exist_ok=True)
     alpha_name = alpha_df["alpha_name"].iloc[0]
-    json_path = f"reports/{alpha_name}_eval.json"
+    json_path = os.path.join(report_dir, f"{alpha_name}_eval.json")
     with open(json_path, "w") as f:
         json.dump(metrics, f, indent=2)
     click.echo(f"\nReport saved: {json_path}")

pipeline/alpha/compute.py

@@ -7,12 +7,19 @@ through :mod:`pipeline.alpha.registry`.
 """
 
 import logging
+from pathlib import Path
+from typing import Iterable
 
 import numpy as np
 import pandas as pd
 
 from pipeline.alpha.registry import get_alpha
 from pipeline.common.schema import ALPHA_COLUMNS
+from pipeline.derived.compute import (
+    DERIVED_KEY_COLUMNS,
+    read_derived_frames,
+    validate_derived_frame,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -25,15 +32,120 @@ def _pivot_close(df: pd.DataFrame) -> pd.DataFrame:
     return pivot.sort_index()
 
 
-def _daily_returns(close: pd.DataFrame) -> pd.DataFrame:
+def _pivot_open(df: pd.DataFrame) -> pd.DataFrame:
-    """Compute daily returns from wide close DataFrame."""
+    """Pivot data to wide format: date index, columns = symbol_id, values = open."""
-    return close.pct_change()
+    pivot = df.pivot_table(
+        index="date", columns="symbol_id", values="open", aggfunc="first"
+    )
+    return pivot.sort_index()
+
+
+def join_feature_frames(
+    data: pd.DataFrame,
+    feature_frames: Iterable[pd.DataFrame],
+) -> pd.DataFrame:
+    """Left-join validated daily derived/feature frames onto long daily data."""
+    out = data.copy()
+    out["date"] = pd.to_datetime(out["date"])
+    existing = set(out.columns)
+    joined_cols: list[str] = []
+
+    for frame in feature_frames:
+        features = validate_derived_frame(frame)
+        feature_cols = [col for col in features.columns if col not in DERIVED_KEY_COLUMNS]
+        overlap = sorted(existing.intersection(feature_cols))
+        if overlap:
+            raise ValueError(
+                f"Feature columns conflict with existing daily data columns: {overlap}"
+            )
+        out = out.merge(
+            features,
+            on=DERIVED_KEY_COLUMNS,
+            how="left",
+            validate="many_to_one",
+        )
+        existing.update(feature_cols)
+        joined_cols.extend(feature_cols)
+
+    if joined_cols:
+        logger.info("Joined feature columns into daily data: %s", joined_cols)
+    return out
+
+
+def _forward_open_to_open_returns(open_: pd.DataFrame) -> pd.DataFrame:
+    """Return earned by a close-formed signal after next-open execution.
+
+    A weight formed after close on date t can first be traded at open[t+1].
+    With daily retargeting it is then held until open[t+2], so the signal-date
+    forward return is open[t+2] / open[t+1] - 1.
+    """
+    return open_.shift(-2).divide(open_.shift(-1)) - 1.0
+
+
+def investable_universe_mask(
+    data: pd.DataFrame,
+    template: pd.DataFrame,
+    *,
+    top_n: int = 1000,
+    min_history: int = 60,
+    require_tradable: bool = True,
+    exclude_st: bool = True,
+) -> pd.DataFrame:
+    """Build a per-date investable-universe mask aligned to ``template``.
+
+    A ``(date, symbol_id)`` cell is ``True`` when the name is, on that date,
+    seasoned (at least ``min_history`` prior closes), currently tradable
+    (``tradestatus == 1``), not flagged ST (``isST == 0``), and inside the
+    ``top_n`` most liquid names by trailing 20-day mean ``amount``. The mask is
+    applied to the *signal* (computed on full contiguous prices), so it
+    restricts only what is *held*, never the price history used to form the
+    signal — that keeps ``pct_change`` correct and look-ahead free.
+
+    Args:
+        data: Long DataFrame with at least ``symbol_id``, ``date``, ``close``,
+            ``amount``, ``isST``, ``tradestatus``.
+        template: Wide signal (date index × ``symbol_id`` columns) to align to.
+        top_n: Keep this many most-liquid names per date.
+        min_history: Minimum number of observed closes before a name is eligible.
+        require_tradable: Require ``tradestatus == 1`` on the date.
+        exclude_st: Drop names flagged ``isST == 1``.
+
+    Returns:
+        Boolean wide DataFrame aligned to ``template``.
+    """
+    def _wide(col: str) -> pd.DataFrame:
+        return (
+            data.pivot_table(index="date", columns="symbol_id", values=col, aggfunc="first")
+            .sort_index()
+            .reindex(index=template.index, columns=template.columns)
+        )
+
+    close = _wide("close")
+    mask = close.notna()
+
+    seasoned = close.notna().cumsum() >= min_history
+    mask &= seasoned
+
+    if exclude_st and "isST" in data.columns:
+        mask &= _wide("isST").fillna(1) == 0
+    if require_tradable and "tradestatus" in data.columns:
+        mask &= _wide("tradestatus").fillna(0) == 1
+
+    amount = _wide("amount")
+    amt_ma = amount.rolling(20, min_periods=10).mean()
+    liquid_rank = amt_ma.rank(axis=1, ascending=False)
+    mask &= liquid_rank <= top_n
+
+    return mask.fillna(False)
 
 
 def compute_alpha(
     data: pd.DataFrame,
     alpha_name: str,
     alpha_type: str,
+    universe: dict | None = None,
+    feature_paths: Iterable[str | Path] | None = None,
+    feature_frames: Iterable[pd.DataFrame] | None = None,
     **params,
 ) -> pd.DataFrame:
     """Compute alpha weights from raw data.
@@ -42,6 +154,15 @@ def compute_alpha(
         data: DataFrame with DATA_COLUMNS.
         alpha_name: Label stored in the ``alpha_name`` output column.
         alpha_type: Registry key of the alpha class (e.g. ``reversal``).
+        universe: Optional investable-universe filter. When given, the alpha's
+            raw signal is masked to the investable set (see
+            :func:`investable_universe_mask`) *before* it is turned into
+            weights, so unheld names get weight 0. Keys are forwarded as keyword
+            arguments to :func:`investable_universe_mask`.
+        feature_paths: Optional parquet files/datasets keyed by ``symbol_id``
+            and ``date``. Their numeric feature columns are left-joined onto
+            ``data`` before alpha logic runs.
+        feature_frames: Optional in-memory feature frames with the same schema.
         **params: Constructor parameters for the alpha (e.g. ``lookback``,
             ``vol_window``). Only the params the alpha's ``__init__`` accepts are
             used; extras are ignored.
@@ -52,9 +173,22 @@ def compute_alpha(
     Raises:
         KeyError: If ``alpha_type`` is not registered.
     """
+    feature_inputs: list[pd.DataFrame] = []
+    if feature_paths:
+        feature_inputs.extend(read_derived_frames(feature_paths))
+    if feature_frames:
+        feature_inputs.extend(feature_frames)
+    if feature_inputs:
+        data = join_feature_frames(data, feature_inputs)
+
     alpha = get_alpha(alpha_type, **params)
     close = _pivot_close(data)
-    weights = alpha.weights(close)
+    signal = alpha.signal_from_data(data, close)
+    if universe is None:
+        weights = alpha.to_weights(signal)
+    else:
+        mask = investable_universe_mask(data, signal, **universe)
+        weights = alpha.to_weights(signal.where(mask))
 
     # Melt to long format
     weights_melted = weights.reset_index().melt(
@@ -82,8 +216,11 @@ def evaluate_alpha(alpha_df: pd.DataFrame, data_df: pd.DataFrame) -> dict:
 
     Computes return, annualized Sharpe, annualized turnover, max drawdown.
 
-    Alpha is interpreted as POSITION WEIGHTS, not predictions.
+    Alpha is interpreted as POSITION WEIGHTS, not predictions. A close-formed
-    Return on date t = sum(weight[s,t] * realized_return[s,t]) / sum(abs(weight[s,t]))
+    weight on date t is assumed tradable at open[t+1] and held until open[t+2].
+    Return on signal date t = sum(weight[s,t] * open_to_open_return[s,t]) /
+    sum(abs(weight[s,t])). This matches the execution convention without
+    crediting the new signal for the overnight gap before it can be traded.
 
     Args:
         alpha_df: DataFrame with ALPHA_COLUMNS.
@@ -93,31 +230,50 @@ def evaluate_alpha(alpha_df: pd.DataFrame, data_df: pd.DataFrame) -> dict:
         Dict with metrics: cumulative_return, sharpe_annual, turnover_annual,
         max_drawdown, hit_rate, n_dates.
     """
-    close = _pivot_close(data_df)
+    open_ = _pivot_open(data_df)
-    returns = _daily_returns(close)
+    fwd_returns_all = _forward_open_to_open_returns(open_)
 
     # Pivot alpha weights to wide format
     weights = alpha_df.pivot_table(
         index="date", columns="symbol_id", values="weight", aggfunc="first"
     ).sort_index()
 
-    # Align dates
+    # Align weights to signal dates that exist on the market calendar. Compute
-    common_dates = weights.index.intersection(returns.index)
+    # forward open-to-open returns on the full market calendar first, so sparse
+    # signal grids still earn the next available open-to-open interval instead
+    # of the next signal date.
+    common_dates = weights.index.intersection(open_.index)
     weights = weights.loc[common_dates]
-    returns = returns.loc[common_dates]
+    fwd_returns = fwd_returns_all.reindex(common_dates)
 
-    if len(common_dates) < 2:
+    if len(common_dates) < 1:
         return {
             "cumulative_return": 0.0,
             "sharpe_annual": 0.0,
             "turnover_annual": 0.0,
             "max_drawdown": 0.0,
             "hit_rate": 0.0,
-            "n_dates": len(common_dates),
+            "n_dates": 0,
         }
 
-    # Daily portfolio return = sum(w * r) / sum(|w|)  — normalized by gross exposure
+    # Daily portfolio return = sum(w_t * r_open[t+1→t+2]) / sum(|w_t|).
-    daily_returns = (weights * returns).sum(axis=1) / weights.abs().sum(axis=1)
+    # The final two signal dates have no complete next-open holding interval
+    # and are dropped below.
+    gross = weights.abs().sum(axis=1)
+    daily_returns = (
+        (weights * fwd_returns).sum(axis=1, min_count=1)
+        / gross.replace(0.0, np.nan)
+    )
+    daily_returns = daily_returns.dropna()
+    if len(daily_returns) < 2:
+        return {
+            "cumulative_return": 0.0,
+            "sharpe_annual": 0.0,
+            "turnover_annual": 0.0,
+            "max_drawdown": 0.0,
+            "hit_rate": 0.0,
+            "n_dates": int(len(daily_returns)),
+        }
 
     # Cumulative return
     cumulative_return = float((1.0 + daily_returns).prod() - 1.0)
@@ -130,7 +286,7 @@ def evaluate_alpha(alpha_df: pd.DataFrame, data_df: pd.DataFrame) -> dict:
     # Annualized turnover: avg daily turnover * 252
     # Daily turnover = sum(|w_t - w_{t-1}|) / sum(|w_{t-1}|)
     weight_change = weights.diff().abs().sum(axis=1)
-    gross_exposure = weights.abs().sum(axis=1).shift(1)
+    gross_exposure = gross.shift(1)
     daily_turnover = weight_change / gross_exposure
     turnover_annual = float(daily_turnover.mean() * 252)
 
@@ -149,5 +305,5 @@ def evaluate_alpha(alpha_df: pd.DataFrame, data_df: pd.DataFrame) -> dict:
         "turnover_annual": turnover_annual,
         "max_drawdown": max_drawdown,
         "hit_rate": hit_rate,
-        "n_dates": len(common_dates),
+        "n_dates": int(len(daily_returns)),
     }

pipeline/alpha/library/__init__.py

@@ -4,4 +4,9 @@ Importing this package imports each alpha module, which registers the alpha via
 the ``@register_alpha`` decorator. Add a new built-in by dropping a module here
 and importing it below.
 """
-from pipeline.alpha.library import momentum, reversal, reversal_vol  # noqa: F401
+from pipeline.alpha.library import (  # noqa: F401
+    momentum,
+    reversal,
+    reversal_rank,
+    reversal_vol,
+)

pipeline/alpha/library/momentum.py

@@ -15,4 +15,4 @@ class MomentumAlpha(BaseAlpha):
         self.lookback = lookback
 
     def signal(self, close: pd.DataFrame) -> pd.DataFrame:
-        return close.pct_change(self.lookback)
+        return close.pct_change(self.lookback, fill_method=None)

pipeline/alpha/library/reversal.py

@@ -15,4 +15,4 @@ class ReversalAlpha(BaseAlpha):
         self.lookback = lookback
 
     def signal(self, close: pd.DataFrame) -> pd.DataFrame:
-        return -close.pct_change(self.lookback)
+        return -close.pct_change(self.lookback, fill_method=None)

pipeline/alpha/library/reversal_rank.py

@@ -0,0 +1,33 @@
+"""Outlier-robust short-horizon reversal alpha."""
+import pandas as pd
+
+from pipeline.alpha.base import BaseAlpha
+from pipeline.alpha.registry import register_alpha
+
+
+@register_alpha
+class ReversalRankAlpha(BaseAlpha):
+    """Reversal weighted by cross-sectional rank instead of z-score.
+
+    The signal is the same trailing-return reversal as :class:`ReversalAlpha`,
+    but :meth:`to_weights` converts it with a cross-sectional rank that is then
+    demeaned. Rank weighting is bounded and monotone, so it does not dump the
+    book into a handful of extreme movers the way raw z-scoring does — the
+    failure mode that makes plain ``reversal`` collapse on the A-share universe,
+    where newly listed / post-suspension / limit-up names produce huge
+    ``pct_change`` outliers.
+    """
+
+    name = "reversal_rank"
+
+    def __init__(self, lookback: int = 5):
+        self.lookback = lookback
+
+    def signal(self, close: pd.DataFrame) -> pd.DataFrame:
+        return -close.pct_change(self.lookback, fill_method=None)
+
+    def to_weights(self, signal: pd.DataFrame) -> pd.DataFrame:
+        signal = signal.dropna(how="all")
+        ranks = signal.rank(axis=1)
+        weights = ranks.subtract(ranks.mean(axis=1), axis=0)
+        return weights.fillna(0.0)

pipeline/alpha/library/reversal_vol.py

@@ -21,6 +21,6 @@ class ReversalVolAlpha(BaseAlpha):
         self.vol_window = vol_window
 
     def signal(self, close: pd.DataFrame) -> pd.DataFrame:
-        reversal = -close.pct_change(self.lookback)
+        reversal = -close.pct_change(self.lookback, fill_method=None)
-        vol = close.pct_change().rolling(self.vol_window).std()
+        vol = close.pct_change(fill_method=None).rolling(self.vol_window).std()
         return reversal / vol

pipeline/combo/cli.py

@@ -26,8 +26,8 @@ def combo():
 def combine(alpha_paths, combo_name, method, output_dir):
     """Combine multiple alphas and save as parquet."""
     paths = [p.strip() for p in alpha_paths.split(",") if p.strip()]
-    if len(paths) < 2:
+    if len(paths) < 1:
-        click.echo("Error: --alpha-paths requires at least 2 comma-separated paths", err=True)
+        click.echo("Error: --alpha-paths requires at least 1 path", err=True)
         return
 
     result = combine_alphas(

pipeline/common/schema.py

@@ -26,6 +26,31 @@ DATA_COLUMNS: Final[list[str]] = [
     "pcfNcfTTM",     # float64: P/CF (net cash flow, TTM)
 ]
 
+# Required columns for raw intraday minute bar parquet files.
+MINUTE_BAR_COLUMNS: Final[list[str]] = [
+    "symbol_id",     # str: internal code like 'sh600000'
+    "symbol_name",   # str: stock name like '浦发银行'
+    "datetime",      # datetime64: intraday bar timestamp
+    "date",          # date component, aligned with daily DATA_COLUMNS date
+    "time",          # str: HH:MM:SS bar time
+    "frequency",     # str: e.g. '5m'
+    "open",          # float64
+    "high",          # float64
+    "low",           # float64
+    "close",         # float64
+    "volume",        # float64 (shares)
+    "amount",        # float64 (turnover in yuan, raw/unadjusted)
+    "vwap",          # float64: amount / volume
+    "adjustflag",    # str: baostock adjustment flag; '3' for raw/unadjusted
+]
+
+# Required key columns for daily derived-data parquet files. Value columns are
+# user/plugin-defined and must be numeric.
+DERIVED_KEY_COLUMNS: Final[list[str]] = [
+    "symbol_id",     # str
+    "date",          # date: normalized daily timestamp
+]
+
 # Required columns for alpha parquet files.
 # Alphas are position WEIGHTS: positive=long, negative=short.
 ALPHA_COLUMNS: Final[list[str]] = [

pipeline/data/cli.py

@@ -3,7 +3,7 @@
 import click
 from datetime import date
 
-from pipeline.data.downloader import download_universe
+from pipeline.data.downloader import download_minute_universe, download_universe
 
 
 @click.group(name="data")
@@ -42,3 +42,37 @@ def download(universe, start_date, end_date, output_dir, symbols, chunk_size, ad
         f"{stats['n_rows']:,} bars, {stats['date_min']} → {stats['date_max']}"
     )
     click.echo(f"Dataset: {stats['dataset_path']}")
+
+
+@data.command("download-minute")
+@click.option(
+    "--universe", default="csi500",
+    help="Which universe: hs300, csi500, all (~5000 A-shares), file path, or comma-separated symbols",
+)
+@click.option("--start-date", default="2017-01-01", help="Start date YYYY-MM-DD")
+@click.option("--end-date", default=str(date.today()), help="End date YYYY-MM-DD")
+@click.option("--output-dir", default="data/minute_bars", help="Root for the partitioned dataset")
+@click.option("--symbols", default=0, type=int, help="Max symbols (0=all)")
+@click.option("--chunk-size", default=100, type=int, help="Symbols per durability flush")
+@click.option("--frequency", default="5", help="Minute frequency: 5, 15, 30, or 60")
+def download_minute(universe, start_date, end_date, output_dir, symbols, chunk_size, frequency):
+    """Download raw Baostock minute bars into a partitioned parquet dataset.
+
+    Writes ``{output_dir}/{universe}/frequency=5m/month=YYYY-MM/*.pq`` for the
+    default 5-minute frequency.
+    """
+    stats = download_minute_universe(
+        universe=universe,
+        start_date=start_date,
+        end_date=end_date,
+        output_dir=output_dir,
+        max_symbols=symbols,
+        chunk_size=chunk_size,
+        frequency=frequency,
+    )
+    click.echo(
+        f"\nSummary: {stats['n_symbols']}/{stats['n_requested']} symbols, "
+        f"{stats['n_rows']:,} bars, {stats['date_min']} → {stats['date_max']}, "
+        f"frequency={stats['frequency']}"
+    )
+    click.echo(f"Dataset: {stats['dataset_path']}")

pipeline/data/downloader.py

@@ -11,9 +11,13 @@ import pyarrow.dataset as pads
 
 # Reuse existing downloader and universe modules
 sys.path.insert(0, str(Path(__file__).resolve().parents[2]))
-from data.downloader import download_daily_batch
+from data.downloader import (
+    _normalize_minute_frequency,
+    download_daily_batch,
+    download_minute_batch,
+)
 from data.universe import get_all_stocks, get_hs300_stocks, get_zz500_stocks
-from pipeline.common.schema import DATA_COLUMNS
+from pipeline.common.schema import DATA_COLUMNS, MINUTE_BAR_COLUMNS
 
 logger = logging.getLogger(__name__)
 
@@ -89,6 +93,25 @@ def _write_month_partitions(df: pd.DataFrame, base_dir: Path, basename_prefix: s
     )
 
 
+def _write_minute_partitions(df: pd.DataFrame, base_dir: Path, basename_prefix: str) -> None:
+    """Append rows to a Hive-partitioned minute dataset.
+
+    Layout: ``frequency=5m/month=YYYY-MM/*.pq``.
+    """
+    out = df.copy()
+    out["month"] = pd.to_datetime(out["date"]).dt.strftime("%Y-%m")
+    table = pa.Table.from_pandas(out, preserve_index=False)
+    pads.write_dataset(
+        table,
+        str(base_dir),
+        format="parquet",
+        partitioning=["frequency", "month"],
+        partitioning_flavor="hive",
+        basename_template=f"{basename_prefix}-{{i}}.pq",
+        existing_data_behavior="overwrite_or_ignore",
+    )
+
+
 def download_universe(
     universe: str = "csi500",
     start_date: str = "2017-01-01",
@@ -177,3 +200,104 @@ def download_universe(
         "date_min": None if date_min is None else str(pd.Timestamp(date_min).date()),
         "date_max": None if date_max is None else str(pd.Timestamp(date_max).date()),
     }
+
+
+def download_minute_universe(
+    universe: str = "csi500",
+    start_date: str = "2017-01-01",
+    end_date: str = "2026-12-31",
+    output_dir: str = "data/minute_bars",
+    max_symbols: int = 0,
+    chunk_size: int = 100,
+    frequency: str | int = 5,
+) -> dict:
+    """Download raw minute bars into a frequency/month-partitioned dataset.
+
+    Args:
+        universe: ``hs300``, ``csi500``, ``all``/``full``, a file path, or a
+            comma-separated symbol list.
+        start_date, end_date: ``YYYY-MM-DD`` bounds.
+        output_dir: Root under which ``{universe}/frequency=5m/month=YYYY-MM``
+            is written.
+        max_symbols: Cap on symbols (0 = all).
+        chunk_size: Symbols per durability flush.
+        frequency: Minute interval. ``5``/``"5"``/``"5m"`` are 5-minute bars.
+
+    Returns:
+        Stats dict with dataset path, row count, symbol count, date range, and
+        frequency label.
+    """
+    _, frequency_label = _normalize_minute_frequency(frequency)
+    constituents = _resolve_universe(universe, max_symbols)
+    symbols = constituents["symbol_id"].tolist()
+    names = dict(zip(constituents["symbol_id"], constituents["symbol_name"]))
+    n_requested = len(symbols)
+    logger.info(
+        "Minute universe %s: %d symbols, %s → %s, frequency=%s",
+        universe,
+        n_requested,
+        start_date,
+        end_date,
+        frequency,
+    )
+
+    base_dir = Path(output_dir) / universe
+    target_frequency_dir = base_dir / f"frequency={frequency_label}"
+    if target_frequency_dir.exists():
+        shutil.rmtree(target_frequency_dir)
+    base_dir.mkdir(parents=True, exist_ok=True)
+
+    buffer: list[pd.DataFrame] = []
+    chunk_idx = 0
+    succeeded = 0
+    n_rows = 0
+    date_min = None
+    date_max = None
+
+    def flush() -> None:
+        nonlocal buffer, chunk_idx, n_rows, date_min, date_max
+        if not buffer:
+            return
+        chunk = pd.concat(buffer, ignore_index=True)
+        _write_minute_partitions(chunk, base_dir, basename_prefix=f"chunk{chunk_idx:04d}")
+        n_rows += len(chunk)
+        cmin, cmax = chunk["date"].min(), chunk["date"].max()
+        date_min = cmin if date_min is None else min(date_min, cmin)
+        date_max = cmax if date_max is None else max(date_max, cmax)
+        logger.info(
+            "Flushed minute chunk %d: %d rows (%d symbols done)",
+            chunk_idx,
+            len(chunk),
+            succeeded,
+        )
+        buffer = []
+        chunk_idx += 1
+
+    for i, (symbol, df) in enumerate(
+        download_minute_batch(symbols, start_date, end_date, frequency=frequency), start=1
+    ):
+        if df is None:
+            logger.warning("  %s: no minute data", symbol)
+        else:
+            df["symbol_id"] = symbol
+            df["symbol_name"] = names.get(symbol, symbol)
+            buffer.append(df[MINUTE_BAR_COLUMNS])
+            succeeded += 1
+        if len(buffer) >= chunk_size:
+            flush()
+        if i % 100 == 0:
+            logger.info("Minute progress: %d/%d symbols", i, n_requested)
+    flush()
+
+    if succeeded == 0:
+        raise RuntimeError("No minute data downloaded for any symbol")
+
+    return {
+        "dataset_path": str(base_dir),
+        "frequency": frequency_label,
+        "n_symbols": succeeded,
+        "n_requested": n_requested,
+        "n_rows": n_rows,
+        "date_min": None if date_min is None else str(pd.Timestamp(date_min).date()),
+        "date_max": None if date_max is None else str(pd.Timestamp(date_max).date()),
+    }

pipeline/derived/__init__.py

@@ -0,0 +1,2 @@
+"""Daily derived-data plugin package."""
+

pipeline/derived/base.py

@@ -0,0 +1,38 @@
+"""Base class for daily derived-data plugins."""
+
+from abc import ABC, abstractmethod
+
+import pandas as pd
+
+
+class BaseDerivedData(ABC):
+    """Compute daily, symbol-keyed numeric derived data.
+
+    Derived-data plugins may use daily bars, minute bars, or both as inputs, but
+    they must always return daily rows keyed by ``symbol_id,date``.
+    """
+
+    #: Unique registry key. Every concrete derived-data plugin must set this.
+    name: str = ""
+
+    @abstractmethod
+    def compute(
+        self,
+        daily: pd.DataFrame | None = None,
+        minute: pd.DataFrame | None = None,
+    ) -> pd.DataFrame:
+        """Compute daily derived data.
+
+        Args:
+            daily: Optional daily market data.
+            minute: Optional raw minute bars.
+
+        Returns:
+            DataFrame with ``symbol_id``, ``date``, and one or more numeric
+            derived-data columns.
+        """
+
+    def __repr__(self) -> str:
+        params = ", ".join(f"{k}={v!r}" for k, v in vars(self).items())
+        return f"{type(self).__name__}({params})"
+

pipeline/derived/cli.py

@@ -0,0 +1,145 @@
+"""CLI for daily derived-data ingestion and computation."""
+
+import click
+import pandas as pd
+
+from pipeline.derived.compute import (
+    DERIVED_KEY_COLUMNS,
+    compute_derived,
+    read_derived_frame,
+    write_derived_frame,
+)
+from pipeline.derived.registry import (
+    available_derived,
+    load_derived_module,
+)
+
+
+@click.group(name="derived")
+def derived():
+    """Ingest, compute, and validate daily derived data."""
+
+
+def _coerce(value: str):
+    """Best-effort coercion of a CLI string to int, then float, else str."""
+    for cast in (int, float):
+        try:
+            return cast(value)
+        except ValueError:
+            continue
+    return value
+
+
+def _parse_params(pairs: tuple[str, ...]) -> dict:
+    """Parse repeated ``name=value`` options into a params dict."""
+    params: dict = {}
+    for pair in pairs:
+        if "=" not in pair:
+            raise click.BadParameter(f"--param must be name=value, got '{pair}'")
+        key, value = pair.split("=", 1)
+        params[key.strip()] = _coerce(value.strip())
+    return params
+
+
+def _read_optional_parquet(path: str | None) -> pd.DataFrame | None:
+    return None if path is None else pd.read_parquet(path)
+
+
+def _summarize(result: pd.DataFrame) -> str:
+    value_cols = [col for col in result.columns if col not in DERIVED_KEY_COLUMNS]
+    return f"{len(result):,} rows, {len(value_cols)} columns"
+
+
+@derived.command("list")
+@click.option(
+    "--derived-module", "derived_modules", multiple=True,
+    help="External module(s) to import first (dotted path or .py file)",
+)
+def list_(derived_modules):
+    """List the registered derived-data plugin types."""
+    for spec in derived_modules:
+        load_derived_module(spec)
+    for name in available_derived():
+        click.echo(name)
+
+
+@derived.command("validate")
+@click.option("--input-path", required=True, help="CSV/parquet file or parquet dataset to validate")
+def validate(input_path):
+    """Validate a daily derived-data file without writing output."""
+    try:
+        result = read_derived_frame(input_path)
+    except Exception as exc:
+        raise click.ClickException(str(exc)) from exc
+    click.echo(f"Valid derived data: {input_path}  ({_summarize(result)})")
+
+
+@derived.command("ingest")
+@click.option("--input-path", required=True, help="CSV/parquet file to ingest")
+@click.option("--derived-name", required=True, help="Name for this derived-data output file")
+@click.option("--output-dir", default="derived", help="Directory to save derived parquet")
+def ingest(input_path, derived_name, output_dir):
+    """Ingest a user-provided daily derived-data CSV/parquet file."""
+    try:
+        result = read_derived_frame(input_path)
+        out_path = write_derived_frame(result, derived_name, output_dir=output_dir)
+    except Exception as exc:
+        raise click.ClickException(str(exc)) from exc
+    click.echo(f"Saved derived data: {out_path}  ({_summarize(result)})")
+
+
+@derived.command("compute")
+@click.option("--daily-path", default=None, help="Optional daily data parquet/dataset")
+@click.option("--minute-path", default=None, help="Optional minute parquet/dataset")
+@click.option("--derived-type", required=True, help="Registry key of the derived-data plugin")
+@click.option("--derived-name", required=True, help="Name for this derived-data output file")
+@click.option("--output-dir", default="derived", help="Directory to save derived parquet")
+@click.option(
+    "--derived-module", "derived_modules", multiple=True,
+    help="External module(s) to import so their derived-data plugins register",
+)
+@click.option(
+    "--param", "extra_params", multiple=True,
+    help="Extra derived-data constructor param as name=value (repeatable)",
+)
+def compute(
+    daily_path,
+    minute_path,
+    derived_type,
+    derived_name,
+    output_dir,
+    derived_modules,
+    extra_params,
+):
+    """Compute one daily derived-data file from daily and/or minute inputs."""
+    for spec in derived_modules:
+        load_derived_module(spec)
+
+    options = available_derived()
+    if derived_type not in options:
+        raise click.BadParameter(
+            f"Unknown derived-type '{derived_type}'. Available: {options}. "
+            f"Use --derived-module to register an external derived-data plugin.",
+            param_hint="--derived-type",
+        )
+    if daily_path is None and minute_path is None:
+        raise click.UsageError("At least one of --daily-path or --minute-path is required")
+
+    daily = _read_optional_parquet(daily_path)
+    if daily_path:
+        click.echo(f"Loaded daily data: {len(daily):,} rows from {daily_path}")
+    minute = _read_optional_parquet(minute_path)
+    if minute_path:
+        click.echo(f"Loaded minute bars: {len(minute):,} rows from {minute_path}")
+
+    try:
+        result = compute_derived(
+            derived_type=derived_type,
+            daily=daily,
+            minute=minute,
+            **_parse_params(extra_params),
+        )
+        out_path = write_derived_frame(result, derived_name, output_dir=output_dir)
+    except Exception as exc:
+        raise click.ClickException(str(exc)) from exc
+    click.echo(f"Saved derived data: {out_path}  ({_summarize(result)})")

pipeline/derived/compute.py

@@ -0,0 +1,115 @@
+"""Derived-data computation and validation."""
+
+import csv
+import logging
+from pathlib import Path
+from typing import Iterable
+
+import pandas as pd
+from pandas.api.types import is_bool_dtype, is_numeric_dtype
+
+from pipeline.common.schema import DERIVED_KEY_COLUMNS
+from pipeline.derived.registry import get_derived
+
+logger = logging.getLogger(__name__)
+
+
+def validate_derived_frame(derived: pd.DataFrame) -> pd.DataFrame:
+    """Validate and normalize a daily derived-data frame.
+
+    A valid derived frame is keyed by unique ``symbol_id,date`` rows and has at
+    least one numeric value column beyond those keys. Dates are normalized to
+    daily timestamps before duplicate-key checks.
+    """
+    duplicated = derived.columns[derived.columns.duplicated()].tolist()
+    if duplicated:
+        raise ValueError(f"Derived data has duplicate columns: {duplicated}")
+
+    missing = [col for col in DERIVED_KEY_COLUMNS if col not in derived.columns]
+    if missing:
+        raise ValueError(f"Derived data missing required columns: {missing}")
+
+    out = derived.copy()
+    out["date"] = pd.to_datetime(out["date"]).dt.normalize()
+
+    if out.duplicated(DERIVED_KEY_COLUMNS).any():
+        raise ValueError("Derived data has duplicate symbol_id,date rows")
+
+    value_cols = [col for col in out.columns if col not in DERIVED_KEY_COLUMNS]
+    if not value_cols:
+        raise ValueError("Derived data must include at least one value column")
+
+    non_numeric = [
+        col
+        for col in value_cols
+        if is_bool_dtype(out[col]) or not is_numeric_dtype(out[col])
+    ]
+    if non_numeric:
+        raise ValueError(f"Derived data value columns must be numeric: {non_numeric}")
+
+    out = out[DERIVED_KEY_COLUMNS + value_cols].copy()
+    return out.sort_values(DERIVED_KEY_COLUMNS).reset_index(drop=True)
+
+
+def compute_derived(
+    derived_type: str,
+    daily: pd.DataFrame | None = None,
+    minute: pd.DataFrame | None = None,
+    **params,
+) -> pd.DataFrame:
+    """Compute one registered derived-data plugin."""
+    if daily is None and minute is None:
+        raise ValueError("Derived data computation requires --daily-path or --minute-path")
+
+    derived = get_derived(derived_type, **params)
+    result = validate_derived_frame(derived.compute(daily=daily, minute=minute))
+    value_cols = [col for col in result.columns if col not in DERIVED_KEY_COLUMNS]
+    logger.info(
+        "Derived data '%s' (%r): %d symbols × %d dates, columns=%s",
+        derived_type,
+        derived,
+        result["symbol_id"].nunique(),
+        result["date"].nunique(),
+        value_cols,
+    )
+    return result
+
+
+def read_derived_frame(path: str | Path) -> pd.DataFrame:
+    """Read and validate one derived CSV/parquet file or parquet dataset."""
+    path = Path(path)
+    if path.suffix.lower() == ".csv":
+        return validate_derived_frame(_read_csv_with_duplicate_header_check(path))
+    return validate_derived_frame(pd.read_parquet(path))
+
+
+def read_derived_frames(derived_paths: Iterable[str | Path]) -> list[pd.DataFrame]:
+    """Read and validate derived-data files."""
+    return [read_derived_frame(path) for path in derived_paths]
+
+
+def write_derived_frame(
+    derived: pd.DataFrame,
+    derived_name: str,
+    output_dir: str | Path = "derived",
+) -> Path:
+    """Validate and write derived data to ``{output_dir}/{derived_name}.pq``."""
+    result = validate_derived_frame(derived)
+    output_dir = Path(output_dir)
+    output_dir.mkdir(parents=True, exist_ok=True)
+    out_path = output_dir / f"{derived_name}.pq"
+    result.to_parquet(out_path, index=False)
+    return out_path
+
+
+def _read_csv_with_duplicate_header_check(path: Path) -> pd.DataFrame:
+    with path.open(newline="") as fh:
+        reader = csv.reader(fh)
+        try:
+            header = next(reader)
+        except StopIteration as exc:
+            raise ValueError("CSV input is empty") from exc
+    duplicated = sorted({col for col in header if header.count(col) > 1})
+    if duplicated:
+        raise ValueError(f"Derived data has duplicate columns: {duplicated}")
+    return pd.read_csv(path)

pipeline/derived/library/__init__.py

@@ -0,0 +1,4 @@
+"""Built-in derived-data library."""
+
+from pipeline.derived.library import minute_daily_summary  # noqa: F401
+

pipeline/derived/library/minute_daily_summary.py

@@ -0,0 +1,88 @@
+"""Daily summary data derived from raw minute bars."""
+
+import numpy as np
+import pandas as pd
+
+from pipeline.derived.base import BaseDerivedData
+from pipeline.derived.registry import register_derived
+
+
+@register_derived
+class MinuteDailySummaryDerived(BaseDerivedData):
+    """Aggregate intraday bars into daily summary columns."""
+
+    name = "minute_daily_summary"
+
+    def compute(
+        self,
+        daily: pd.DataFrame | None = None,
+        minute: pd.DataFrame | None = None,
+    ) -> pd.DataFrame:
+        if minute is None:
+            raise ValueError("minute_daily_summary requires minute input")
+
+        minute = minute.copy()
+        minute["date"] = pd.to_datetime(minute["date"]).dt.normalize()
+        sort_cols = ["symbol_id", "date"]
+        if "datetime" in minute.columns:
+            minute["datetime"] = pd.to_datetime(minute["datetime"])
+            sort_cols.append("datetime")
+        elif "time" in minute.columns:
+            sort_cols.append("time")
+        minute = minute.sort_values(sort_cols)
+
+        grouped = minute.groupby(["symbol_id", "date"], sort=True)
+        summary = grouped.agg(
+            minute_bar_count=("close", "count"),
+            first_open=("open", "first"),
+            last_close=("close", "last"),
+            high=("high", "max"),
+            low=("low", "min"),
+            volume_sum=("volume", "sum"),
+            amount_sum=("amount", "sum"),
+        )
+        summary["minute_intraday_return"] = (
+            summary["last_close"] / summary["first_open"] - 1.0
+        )
+        summary["minute_intraday_range"] = summary["high"] / summary["low"] - 1.0
+        summary["minute_vwap"] = (
+            summary["amount_sum"] / summary["volume_sum"].where(summary["volume_sum"] > 0)
+        )
+        summary = summary.reset_index()
+
+        if daily is not None:
+            daily_keys = daily[["symbol_id", "date"]].copy()
+            daily_keys["date"] = pd.to_datetime(daily_keys["date"]).dt.normalize()
+            daily_keys = daily_keys.drop_duplicates(["symbol_id", "date"])
+            result = daily_keys.merge(summary, on=["symbol_id", "date"], how="left")
+            if "close" in daily.columns:
+                daily_close = daily[["symbol_id", "date", "close"]].copy()
+                daily_close["date"] = pd.to_datetime(daily_close["date"]).dt.normalize()
+                daily_close = daily_close.drop_duplicates(["symbol_id", "date"])
+                result = result.merge(
+                    daily_close.rename(columns={"close": "daily_close"}),
+                    on=["symbol_id", "date"],
+                    how="left",
+                )
+                reference_close = result["daily_close"].fillna(result["last_close"])
+            else:
+                reference_close = result["last_close"]
+        else:
+            result = summary
+            reference_close = result["last_close"]
+
+        result["minute_vwap_deviation"] = (
+            result["minute_vwap"] / reference_close.replace(0.0, np.nan) - 1.0
+        )
+        return result[
+            [
+                "symbol_id",
+                "date",
+                "minute_bar_count",
+                "minute_intraday_return",
+                "minute_intraday_range",
+                "minute_vwap",
+                "minute_vwap_deviation",
+            ]
+        ]
+

pipeline/derived/registry.py

@@ -0,0 +1,80 @@
+"""Registry and factory for daily derived-data plugins."""
+
+import importlib
+import importlib.util
+import inspect
+from pathlib import Path
+from typing import Optional, Type
+
+from pipeline.derived.base import BaseDerivedData
+
+_REGISTRY: dict[str, Type[BaseDerivedData]] = {}
+_builtins_loaded = False
+
+
+def register_derived(cls: Type[BaseDerivedData]) -> Type[BaseDerivedData]:
+    """Class decorator that registers derived data under ``BaseDerivedData.name``."""
+    if not (isinstance(cls, type) and issubclass(cls, BaseDerivedData)):
+        raise TypeError(f"{cls!r} is not a BaseDerivedData subclass")
+    key = getattr(cls, "name", "")
+    if not key:
+        raise ValueError(f"{cls.__name__} must set a non-empty class attribute `name`")
+    existing = _REGISTRY.get(key)
+    if existing is not None and existing is not cls:
+        raise ValueError(
+            f"Derived data name '{key}' already registered by {existing.__name__}"
+        )
+    _REGISTRY[key] = cls
+    return cls
+
+
+def available_derived() -> list[str]:
+    """Sorted names of all registered derived-data plugins."""
+    _ensure_builtins()
+    return sorted(_REGISTRY)
+
+
+def get_derived(name: str, **params) -> BaseDerivedData:
+    """Instantiate a registered derived-data plugin by name.
+
+    Only parameters accepted by the plugin class's ``__init__`` are forwarded.
+    """
+    _ensure_builtins()
+    if name not in _REGISTRY:
+        raise KeyError(f"Unknown derived data '{name}'. Available: {sorted(_REGISTRY)}")
+    cls = _REGISTRY[name]
+    accepted = _accepted_params(cls)
+    kwargs = params if accepted is None else {k: v for k, v in params.items() if k in accepted}
+    return cls(**kwargs)
+
+
+def load_derived_module(spec: str) -> None:
+    """Import an external module so its ``@register_derived`` classes register."""
+    looks_like_file = spec.endswith(".py") or Path(spec).expanduser().exists()
+    if looks_like_file:
+        path = Path(spec).expanduser().resolve()
+        if not path.exists():
+            raise FileNotFoundError(f"Derived data module not found: {path}")
+        module_spec = importlib.util.spec_from_file_location(path.stem, path)
+        if module_spec is None or module_spec.loader is None:
+            raise ImportError(f"Cannot load derived data module from {path}")
+        module = importlib.util.module_from_spec(module_spec)
+        module_spec.loader.exec_module(module)
+    else:
+        importlib.import_module(spec)
+
+
+def _accepted_params(cls: Type[BaseDerivedData]) -> Optional[set[str]]:
+    """Param names ``cls.__init__`` accepts, or None if it takes ``**kwargs``."""
+    sig = inspect.signature(cls.__init__)
+    if any(p.kind is p.VAR_KEYWORD for p in sig.parameters.values()):
+        return None
+    return {name for name in sig.parameters if name != "self"}
+
+
+def _ensure_builtins() -> None:
+    global _builtins_loaded
+    if not _builtins_loaded:
+        import pipeline.derived.library  # noqa: F401
+        _builtins_loaded = True
+

pipeline/features/__init__.py

@@ -0,0 +1 @@
+"""Daily feature plugin package."""

pipeline/features/base.py

@@ -0,0 +1,8 @@
+"""Compatibility alias for daily feature plugins.
+
+The canonical plugin API is ``pipeline.derived``. ``BaseFeature`` remains as an
+alias so existing external feature modules continue to register unchanged.
+"""
+
+from pipeline.derived.base import BaseDerivedData as BaseFeature
+

pipeline/features/cli.py

@@ -0,0 +1,108 @@
+"""CLI for daily feature computation."""
+
+import os
+
+import click
+import pandas as pd
+
+from pipeline.features.compute import compute_feature
+from pipeline.features.registry import available_features, load_feature_module
+
+
+@click.group(name="feature")
+def feature():
+    """Compute daily feature parquet files from minute bars."""
+
+
+def _coerce(value: str):
+    """Best-effort coercion of a CLI string to int, then float, else str."""
+    for cast in (int, float):
+        try:
+            return cast(value)
+        except ValueError:
+            continue
+    return value
+
+
+def _parse_params(pairs: tuple[str, ...]) -> dict:
+    """Parse repeated ``name=value`` options into a params dict."""
+    params: dict = {}
+    for pair in pairs:
+        if "=" not in pair:
+            raise click.BadParameter(f"--param must be name=value, got '{pair}'")
+        key, value = pair.split("=", 1)
+        params[key.strip()] = _coerce(value.strip())
+    return params
+
+
+@feature.command("list")
+@click.option(
+    "--feature-module", "feature_modules", multiple=True,
+    help="External module(s) to import first (dotted path or .py file)",
+)
+def list_(feature_modules):
+    """List the registered feature types."""
+    for spec in feature_modules:
+        load_feature_module(spec)
+    for name in available_features():
+        click.echo(name)
+
+
+@feature.command("compute")
+@click.option("--minute-path", required=True, help="Path to minute parquet dataset/file")
+@click.option("--daily-path", default=None, help="Optional daily data parquet for alignment")
+@click.option("--feature-type", required=True, help="Registry key of the feature class")
+@click.option("--feature-name", required=True, help="Name for this feature run/output file")
+@click.option("--output-dir", default="features", help="Directory to save feature parquet")
+@click.option(
+    "--feature-module", "feature_modules", multiple=True,
+    help="External module(s) to import so their features register (dotted path or .py file)",
+)
+@click.option(
+    "--param", "extra_params", multiple=True,
+    help="Extra feature constructor param as name=value (repeatable)",
+)
+def compute(
+    minute_path,
+    daily_path,
+    feature_type,
+    feature_name,
+    output_dir,
+    feature_modules,
+    extra_params,
+):
+    """Compute one daily feature file from raw minute bars."""
+    for spec in feature_modules:
+        load_feature_module(spec)
+
+    options = available_features()
+    if feature_type not in options:
+        raise click.BadParameter(
+            f"Unknown feature-type '{feature_type}'. Available: {options}. "
+            f"Use --feature-module to register an external feature.",
+            param_hint="--feature-type",
+        )
+
+    minute = pd.read_parquet(minute_path)
+    click.echo(f"Loaded minute bars: {len(minute):,} rows from {minute_path}")
+
+    daily = None
+    if daily_path:
+        daily = pd.read_parquet(daily_path)
+        click.echo(f"Loaded daily data: {len(daily):,} rows from {daily_path}")
+
+    result = compute_feature(
+        minute=minute,
+        daily=daily,
+        feature_type=feature_type,
+        **_parse_params(extra_params),
+    )
+
+    os.makedirs(output_dir, exist_ok=True)
+    out_path = f"{output_dir}/{feature_name}.pq"
+    result.to_parquet(out_path, index=False)
+    feature_cols = [col for col in result.columns if col not in ("symbol_id", "date")]
+    click.echo(
+        f"Saved feature: {out_path}  ({len(result):,} rows, "
+        f"{len(feature_cols)} columns)"
+    )

pipeline/features/compute.py

@@ -0,0 +1,41 @@
+"""Compatibility wrappers for daily feature computation and validation."""
+
+from pathlib import Path
+from typing import Iterable
+
+import pandas as pd
+
+from pipeline.derived.compute import (
+    DERIVED_KEY_COLUMNS,
+    compute_derived,
+    read_derived_frames,
+    validate_derived_frame,
+)
+
+FEATURE_KEY_COLUMNS = DERIVED_KEY_COLUMNS
+
+
+def validate_feature_frame(features: pd.DataFrame) -> pd.DataFrame:
+    """Validate and normalize a legacy daily feature frame."""
+    return validate_derived_frame(features)
+
+
+def compute_feature(
+    minute: pd.DataFrame,
+    feature_type: str,
+    daily: pd.DataFrame | None = None,
+    **params,
+) -> pd.DataFrame:
+    """Compute one registered feature through the derived-data registry."""
+    return compute_derived(
+        derived_type=feature_type,
+        daily=daily,
+        minute=minute,
+        **params,
+    )
+
+
+def read_feature_frames(feature_paths: Iterable[str | Path]) -> list[pd.DataFrame]:
+    """Read and validate feature/derived-data parquet files."""
+    return read_derived_frames(feature_paths)
+

pipeline/features/library/__init__.py

@@ -0,0 +1,3 @@
+"""Built-in feature library."""
+
+from pipeline.features.library import minute_daily_summary  # noqa: F401

pipeline/features/library/minute_daily_summary.py

@@ -0,0 +1,16 @@
+"""Compatibility wrapper for the built-in minute daily summary plugin."""
+
+import pandas as pd
+
+from pipeline.derived.library.minute_daily_summary import MinuteDailySummaryDerived
+
+
+class MinuteDailySummaryFeature(MinuteDailySummaryDerived):
+    """Legacy minute-first wrapper around the derived-data implementation."""
+
+    def compute(
+        self,
+        minute: pd.DataFrame,
+        daily: pd.DataFrame | None = None,
+    ) -> pd.DataFrame:
+        return super().compute(daily=daily, minute=minute)

pipeline/features/registry.py

@@ -0,0 +1,36 @@
+"""Compatibility registry wrappers for daily feature plugins."""
+
+from typing import Type
+
+from pipeline.derived.base import BaseDerivedData
+from pipeline.derived.registry import (
+    available_derived,
+    get_derived,
+    load_derived_module,
+    register_derived,
+)
+from pipeline.features.base import BaseFeature
+
+
+def register_feature(cls: Type[BaseFeature]) -> Type[BaseFeature]:
+    """Register a legacy feature plugin in the derived-data registry."""
+    return register_derived(cls)
+
+
+def available_features() -> list[str]:
+    """Sorted names of all registered feature/derived-data plugins."""
+    return available_derived()
+
+
+def get_feature(name: str, **params) -> BaseDerivedData:
+    """Instantiate a registered feature/derived-data plugin by name."""
+    if name == "minute_daily_summary":
+        from pipeline.features.library.minute_daily_summary import MinuteDailySummaryFeature
+
+        return MinuteDailySummaryFeature(**params)
+    return get_derived(name, **params)
+
+
+def load_feature_module(spec: str) -> None:
+    """Import an external module so its ``@register_feature`` classes register."""
+    load_derived_module(spec)

pipeline/portfolio/construct.py

@@ -7,9 +7,9 @@ across dates (positions are stateful, unlike alphas/combos), discretizing and
 repairing each day's target into a tradable integer book.
 
 Return-convention note: weights here are *target allocations*. The research
-evaluation in :mod:`pipeline.portfolio.research` marks them close-to-close on the
+evaluation in :mod:`pipeline.portfolio.research` marks them from next open to
-*next* period (no look-ahead); the execution simulator marks the actually-filled
+the following open (no look-ahead); the execution simulator marks the
-book at the next open. See those modules for details.
+actually-filled book at the next open. See those modules for details.
 """
 
 from __future__ import annotations

pipeline/portfolio/costs.py

@@ -0,0 +1,50 @@
+"""Trading cost models for portfolio execution simulation."""
+
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+from dataclasses import dataclass
+from typing import Mapping
+
+import numpy as np
+
+
+class CostModel(ABC):
+    """Interface for per-name execution cost models."""
+
+    @abstractmethod
+    def compute(
+        self,
+        traded_shares: np.ndarray,
+        execution_price: np.ndarray,
+        side: np.ndarray,
+        date,
+        metadata: Mapping[str, object] | None = None,
+    ) -> np.ndarray:
+        """Return per-name trading cost in yuan."""
+
+
+@dataclass(frozen=True)
+class SimpleProportionalCostModel(CostModel):
+    """Simplified open-execution proportional cost model.
+
+    Slippage is represented as an additional cash cost. The execution price is
+    not adjusted by slippage, which avoids double-counting.
+    """
+
+    cost_bps: float = 0.0
+    slippage_bps: float = 0.0
+
+    def compute(
+        self,
+        traded_shares: np.ndarray,
+        execution_price: np.ndarray,
+        side: np.ndarray,
+        date,
+        metadata: Mapping[str, object] | None = None,
+    ) -> np.ndarray:
+        shares = np.asarray(traded_shares, dtype=np.float64)
+        price = np.asarray(execution_price, dtype=np.float64)
+        open_price = np.where(np.isfinite(price), price, 0.0)
+        trade_value = np.abs(shares * open_price)
+        return trade_value * (self.cost_bps + self.slippage_bps) / 1e4

pipeline/portfolio/research.py

@@ -6,9 +6,10 @@ trading constraints. Metrics are return / Sharpe / turnover / max-drawdown /
 convention that an alpha is a position weight, not a return predictor.
 
 Return convention (documented): the target weight formed from information at
-date ``t`` earns the *next* period's close-to-close return, i.e. weights are
+date ``t`` is assumed tradable at ``open[t+1]`` and held until ``open[t+2]``.
-shifted one day relative to realized returns, so there is no look-ahead:
+This is a costless approximation of the next-open execution path: no lots,
-``R_t = sum_i w_{i,t} · r_{i,t+1}`` normalized by gross exposure.
+constraints, or costs, but no credit for an overnight gap that the new signal
+could not have owned.
 """
 
 from __future__ import annotations
@@ -26,39 +27,46 @@ def evaluate_portfolio(positions_df: pd.DataFrame, data_df: pd.DataFrame) -> dic
     Args:
         positions_df: POSITION_COLUMNS (uses ``target_weight``; zero-gross
             construction carry dates remain flat in this research view).
-        data_df: DATA_COLUMNS (uses ``close`` for returns).
+        data_df: DATA_COLUMNS (uses ``open`` for returns).
 
     Returns:
         Dict with ``cumulative_return, sharpe_annual, turnover_annual,
         max_drawdown, fitness, hit_rate, n_dates``. No IC key.
     """
-    close = data_df.pivot_table(
+    open_ = data_df.pivot_table(
-        index="date", columns="symbol_id", values="close", aggfunc="first"
+        index="date", columns="symbol_id", values="open", aggfunc="first"
     ).sort_index()
-    returns = close.pct_change()
+    fwd = open_.shift(-2).divide(open_.shift(-1)) - 1.0
 
     weights = positions_df.pivot_table(
         index="date", columns="symbol_id", values="target_weight", aggfunc="first"
     ).sort_index()
 
-    common = weights.index.intersection(returns.index)
+    common = weights.index.intersection(open_.index)
     weights = weights.loc[common]
-    returns = returns.loc[common]
+    # Compute forward returns on the full market calendar before selecting
+    # signal dates. This preserves the next available open-to-open holding
+    # interval when the signal grid is sparser than the data grid.
+    fwd = fwd.reindex(common)
 
     empty = {
         "cumulative_return": 0.0, "sharpe_annual": 0.0, "turnover_annual": 0.0,
         "max_drawdown": 0.0, "fitness": 0.0, "hit_rate": 0.0,
         "n_dates": len(common),
     }
-    if len(common) < 3:
+    if len(common) < 1:
+        empty["n_dates"] = 0
         return empty
 
     gross = weights.abs().sum(axis=1)
-    # Weights at t earn the return from t to t+1: shift returns back by one.
+    # Weights at t earn the costless tradable interval open[t+1] -> open[t+2].
-    fwd = returns.shift(-1)
+    daily = (
-    daily = (weights * fwd).sum(axis=1) / gross.replace(0.0, np.nan)
+        (weights * fwd).sum(axis=1, min_count=1)
+        / gross.replace(0.0, np.nan)
+    )
     daily = daily.dropna()
     if len(daily) < 2:
+        empty["n_dates"] = int(len(daily))
         return empty
 
     cumulative_return = float((1.0 + daily).prod() - 1.0)

pipeline/portfolio/simulator.py

@@ -4,7 +4,8 @@ Execution model (documented convention): a position book targeted from
 information available on date ``t`` is executed at ``open[t+1]``. Trades that
 violate a :class:`~pipeline.portfolio.constraints.TradeConstraint` (suspension,
 price limit, volume cap, …) are clipped; a fully blocked buy leaves the position
-at its previous level. Realized PnL marks the *actually filled* book.
+at its previous level. Realized PnL marks the *actually filled* book. Trading
+cost defaults to a simplified open-execution proportional cash-cost model.
 
 The simulator is an ABC + a :class:`ReferenceSimulator`; constraints compose by
 intersecting their per-name signed delta bounds.
@@ -21,6 +22,7 @@ import pandas as pd
 
 from pipeline.common.schema import FILL_COLUMNS, PNL_COLUMNS
 from pipeline.portfolio.constraints import TradeConstraint
+from pipeline.portfolio.costs import CostModel, SimpleProportionalCostModel
 from pipeline.portfolio.market_rules import MarketRule, compute_limit_status
 
 logger = logging.getLogger(__name__)
@@ -65,10 +67,12 @@ class ExecutionSimulator(ABC):
     """Abstract execution layer. Subclasses define how a target gets filled."""
 
     def __init__(self, constraints: list[TradeConstraint] | None = None,
-                 cost_bps: float = 0.0, slippage_bps: float = 0.0):
+                 cost_bps: float = 0.0, slippage_bps: float = 0.0,
+                 cost_model: CostModel | None = None):
         self.constraints = constraints or []
-        self.cost_bps = cost_bps
+        self.cost_model = cost_model or SimpleProportionalCostModel(
-        self.slippage_bps = slippage_bps
+            cost_bps=cost_bps, slippage_bps=slippage_bps
+        )
 
     @abstractmethod
     def fill(self, ctx: TradeContext) -> FillResult:
@@ -104,9 +108,17 @@ class ReferenceSimulator(ExecutionSimulator):
         blocked = (traded != desired).astype(np.int64)
 
         realized = prev + traded
-        open_px = np.where(np.isfinite(ctx.slice.price), ctx.slice.price, 0.0)
+        cost = self.cost_model.compute(
-        trade_value = np.abs(traded.astype(np.float64) * open_px)
+            traded_shares=traded,
-        cost = trade_value * (self.cost_bps + self.slippage_bps) / 1e4
+            execution_price=ctx.slice.price,
+            side=np.sign(traded),
+            date=ctx.slice.date,
+            metadata={
+                "symbol_ids": ctx.slice.symbol_ids,
+                "booksize": ctx.booksize,
+                "market_slice": ctx.slice,
+            },
+        )
         return FillResult(realized, traded, cost, blocked)
 
     def run(
@@ -154,13 +166,14 @@ class ReferenceSimulator(ExecutionSimulator):
         st = wide(data_df, "isST") if "isST" in data_df.columns else opn * 0.0
 
         symbols = sorted(set(tgt.columns) | set(opn.columns))
+        data_index = close.index
         tgt = tgt.reindex(columns=symbols)
-        opn = opn.reindex(columns=symbols)
+        opn = opn.reindex(index=data_index, columns=symbols)
         close = close.reindex(columns=symbols)
-        preclose = preclose.reindex(columns=symbols)
+        preclose = preclose.reindex(index=data_index, columns=symbols)
-        amount = amount.reindex(columns=symbols)
+        amount = amount.reindex(index=data_index, columns=symbols)
-        tstat = tstat.reindex(columns=symbols)
+        tstat = tstat.reindex(index=data_index, columns=symbols)
-        st = st.reindex(columns=symbols)
+        st = st.reindex(index=data_index, columns=symbols)
 
         sym_arr = np.asarray(symbols, dtype=object)
         n = len(symbols)

plugins/__init__.py

@@ -0,0 +1,2 @@
+"""Optional plugin packages for the research pipeline."""
+

plugins/joinquant/README.md

@@ -0,0 +1,103 @@
+# JoinQuant Comparison Plugin
+
+This plugin exports frozen targets from the internal A-share research pipeline,
+drives a standalone JoinQuant wrapper strategy, ingests JoinQuant output files,
+and reconciles them against the internal reference simulator.
+
+The plugin validates system mechanics, not alpha quality:
+
+- date alignment
+- symbol mapping
+- target position generation
+- open execution timing
+- lot rounding and filled shares
+- position carry
+- trading cost and PnL accounting
+- blocked trades from suspension and price limits
+
+## Commands
+
+```bash
+uv run python cli.py joinquant prepare-smoke \
+  --out-dir /tmp/chinese-equity-quant-realdata
+
+uv sync --extra joinquant-browser
+uv run playwright install chromium
+
+uv run python cli.py joinquant browser-login \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json
+
+uv run python cli.py joinquant write-browser-config \
+  --out-path /tmp/chinese-equity-quant-realdata/joinquant_browser_config.json \
+  --strategy-url "https://www.joinquant.com/..." \
+  --flow backtest
+
+uv run python cli.py joinquant run-browser-backtest \
+  --manifest-path /tmp/chinese-equity-quant-realdata/joinquant_smoke_manifest.json \
+  --config-path /tmp/chinese-equity-quant-realdata/joinquant_browser_config.json \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json
+
+uv run python cli.py joinquant write-browser-config \
+  --out-path /tmp/chinese-equity-quant-realdata/joinquant_sim_config.json \
+  --strategy-url "https://www.joinquant.com/<模拟盘 page>" \
+  --flow sim-trade
+
+uv run python cli.py joinquant run-browser-sim \
+  --manifest-path /tmp/chinese-equity-quant-realdata/joinquant_smoke_manifest.json \
+  --config-path /tmp/chinese-equity-quant-realdata/joinquant_sim_config.json \
+  --storage-state ~/.config/chinese-equity-quant/joinquant_storage_state.json
+
+uv run python cli.py joinquant export-targets \
+  --positions-path portfolio/run1.pq \
+  --portfolio-name run1 \
+  --mode target_shares \
+  --execution-calendar-path data/daily_bars/csi500 \
+  --start-date 2026-07-01 \
+  --end-date 2026-07-31 \
+  --out-dir plugins_output/joinquant/targets
+
+uv run python cli.py joinquant write-wrapper \
+  --portfolio-name run1 \
+  --mode target_shares \
+  --out-path plugins_output/joinquant/wrapper_strategy_run1.py
+
+uv run python cli.py joinquant ingest \
+  --portfolio-name run1 \
+  --fills-csv path/to/jq_fills.csv \
+  --positions-csv path/to/jq_positions.csv \
+  --pnl-csv path/to/jq_pnl.csv \
+  --out-dir plugins_output/joinquant/ingested
+
+uv run python cli.py joinquant reconcile \
+  --portfolio-name run1 \
+  --targets-dir plugins_output/joinquant/targets/run1 \
+  --our-fills-path fills/run1.pq \
+  --our-positions-path portfolio/run1.pq \
+  --our-pnl-path pnl/run1.pq \
+  --jq-fills-path plugins_output/joinquant/ingested/run1/fills.pq \
+  --jq-positions-path plugins_output/joinquant/ingested/run1/positions.pq \
+  --jq-pnl-path plugins_output/joinquant/ingested/run1/pnl.pq \
+  --out-dir plugins_output/joinquant/reconcile
+```
+
+`target_shares` is the default and uses the built integer `position_shares`
+from `portfolio build`, matching what the internal simulator executes.
+For strict simulator-vs-JoinQuant comparison, pass `--execution-calendar-path`
+so position dates are shifted to the next session open, matching the internal
+simulator's next-open convention.
+
+`prepare-smoke` automates the local side of the first sanity check: tiny real
+data download, one-stock long-only position file, internal simulation, aligned
+target export, wrapper generation, and a manifest with expected JoinQuant CSV
+export paths.
+
+`run-browser-backtest` automates the remote JoinQuant web run through
+Playwright. It reuses a saved browser login state, executes the configured UI
+actions, downloads JoinQuant CSVs when configured, and runs ingest/reconcile
+automatically once all three CSVs are present.
+
+`run-browser-sim` is the forward-test / 模拟盘 equivalent. Use a `--flow
+sim-trade` config to upload the frozen next-session target file, save the
+strategy, and start or restart the JoinQuant simulated-trading job. After close,
+run it again with download actions or use `ingest` / `reconcile` directly on
+exported CSVs.

plugins/joinquant/__init__.py

@@ -0,0 +1,10 @@
+"""JoinQuant comparison plugin.
+
+This package keeps JoinQuant-specific export, ingest, and reconciliation code
+outside the core portfolio modules.
+"""
+
+from plugins.joinquant.symbols import from_joinquant_symbol, to_joinquant_symbol
+
+__all__ = ["from_joinquant_symbol", "to_joinquant_symbol"]
+

plugins/joinquant/browser.py

@@ -0,0 +1,710 @@
+"""Browser automation for JoinQuant cloud backtest and simulated-trading runs.
+
+JoinQuant's public ``jqdatasdk`` is a data API; cloud strategy upload/run/export
+is exposed through the web application. This module keeps that automation
+optional and configurable so UI selector drift does not affect the core test
+suite.
+"""
+
+from __future__ import annotations
+
+import json
+import re
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+from plugins.joinquant.ingest import ingest_joinquant_outputs
+from plugins.joinquant.reconcile import reconcile_joinquant
+
+
+DEFAULT_LOGIN_URL = "https://www.joinquant.com/user/login/index"
+
+
+def _require_playwright():
+    try:
+        from playwright.sync_api import TimeoutError as PlaywrightTimeoutError
+        from playwright.sync_api import sync_playwright
+    except ImportError as exc:
+        raise RuntimeError(
+            "Playwright is required for JoinQuant browser automation. Install it "
+            "in this uv environment, then install Chromium:\n"
+            "  uv sync --extra joinquant-browser\n"
+            "  uv run playwright install chromium"
+        ) from exc
+    return sync_playwright, PlaywrightTimeoutError
+
+
+def _backtest_actions() -> list[dict[str, Any]]:
+    return [
+        {"type": "goto", "url": "{strategy_url}"},
+        {
+            "type": "click",
+            "selector": "text=不再提示",
+            "timeout_ms": 5000,
+            "force": True,
+            "optional": True,
+            "description": "Dismiss JoinQuant first-run editor guide.",
+        },
+        {
+            "type": "click",
+            "selector": "text=确定",
+            "timeout_ms": 5000,
+            "force": True,
+            "optional": True,
+            "description": "Close JoinQuant first-run editor guide.",
+        },
+        {
+            "type": "click",
+            "selector": "text=跳过",
+            "timeout_ms": 5000,
+            "force": True,
+            "optional": True,
+            "description": "Skip JoinQuant first-run editor guide.",
+        },
+        {
+            "type": "evaluate",
+            "script": (
+                "() => {"
+                "document.querySelectorAll("
+                "'.introjs-overlay,.introjs-helperLayer,.introjs-tooltipReferenceLayer,"
+                ".introjs-tooltip,.introjs-disableInteraction'"
+                ").forEach((node) => node.remove());"
+                "document.body.classList.remove('introjs-noscroll');"
+                "}"
+            ),
+            "optional": True,
+            "description": "Remove any remaining JoinQuant intro overlay.",
+        },
+        {
+            "type": "set_ace_editor_file",
+            "selector": ".ace_editor",
+            "path": "{wrapper_path}",
+            "description": "Set generated wrapper strategy in the Ace code editor.",
+        },
+        {
+            "type": "set_input_files",
+            "selector": "input[type=file]",
+            "paths": "{target_csvs}",
+            "description": "Upload all aligned daily target CSV files.",
+            "timeout_ms": 5000,
+            "optional": True,
+        },
+        {
+            "type": "evaluate",
+            "script": (
+                "(arg) => {"
+                "const setValue = (selector, value) => {"
+                "const el = document.querySelector(selector);"
+                "if (!el) return false;"
+                "el.value = value;"
+                "el.setAttribute('value', value);"
+                "el.dispatchEvent(new Event('input', {bubbles: true}));"
+                "el.dispatchEvent(new Event('change', {bubbles: true}));"
+                "return true;"
+                "};"
+                "return {"
+                "start: setValue('#startTime', arg.start),"
+                "end: setValue('#endTime', arg.end),"
+                "capital: setValue('#daily_backtest_capital_base_box', arg.capital)"
+                "};"
+                "}"
+            ),
+            "arg": {
+                "start": "{backtest_start_date}",
+                "end": "{backtest_end_date}",
+                "capital": "{booksize}",
+            },
+            "description": "Set JoinQuant backtest dates and starting capital.",
+            "optional": True,
+        },
+        {
+            "type": "click",
+            "selector": "#algo-save-button",
+            "timeout_ms": 5000,
+            "description": "Save the JoinQuant strategy code.",
+            "optional": True,
+        },
+        {
+            "type": "click",
+            "selector": ".bootstrap-dialog .btn-primary, .modal.in button:has-text(\"确定\")",
+            "timeout_ms": 5000,
+            "force": True,
+            "description": "Confirm any JoinQuant save dialog.",
+            "optional": True,
+        },
+        {"type": "wait_for_timeout", "timeout_ms": 2000},
+        {
+            "type": "click",
+            "selector": "#daily-new-backtest-button",
+            "description": "Start the JoinQuant backtest.",
+        },
+        {
+            "type": "wait_for_selector",
+            "selector": "text=/回测完成|运行完成|Backtest complete|Finished/i",
+            "timeout_ms": 600_000,
+            "optional": True,
+        },
+        {
+            "type": "download",
+            "selector": "text=/导出成交|下载成交|fills|trades/i",
+            "save_as": "{expected_joinquant_csvs.fills}",
+            "timeout_ms": 15000,
+            "optional": True,
+        },
+        {
+            "type": "download",
+            "selector": "text=/导出持仓|下载持仓|positions/i",
+            "save_as": "{expected_joinquant_csvs.positions}",
+            "timeout_ms": 15000,
+            "optional": True,
+        },
+        {
+            "type": "download",
+            "selector": "text=/导出收益|下载收益|pnl|收益/i",
+            "save_as": "{expected_joinquant_csvs.pnl}",
+            "timeout_ms": 15000,
+            "optional": True,
+        },
+        {"type": "screenshot", "path": "{run_artifact_dir}/final.png"},
+    ]
+
+
+def _sim_trade_actions() -> list[dict[str, Any]]:
+    return [
+        {"type": "goto", "url": "{strategy_url}"},
+        {
+            "type": "paste_text_file",
+            "selector": "textarea, .ace_text-input, .cm-content, .CodeMirror textarea",
+            "path": "{wrapper_path}",
+            "description": "Paste generated wrapper strategy into the simulated-trading strategy editor.",
+        },
+        {
+            "type": "set_input_files",
+            "selector": "input[type=file]",
+            "paths": "{target_csvs}",
+            "description": "Upload frozen target CSV files for 模拟盘.",
+            "optional": True,
+        },
+        {
+            "type": "click",
+            "selector": "text=/保存|Save/i",
+            "description": "Save the strategy code and uploaded files.",
+            "optional": True,
+        },
+        {
+            "type": "wait_for_selector",
+            "selector": "text=/保存成功|已保存|Saved/i",
+            "timeout_ms": 120_000,
+            "optional": True,
+        },
+        {
+            "type": "click",
+            "selector": "text=/模拟盘|模拟交易|启动模拟|运行模拟|启动|重启|Run Sim|Start|Restart/i",
+            "description": "Start or restart the JoinQuant simulated-trading job.",
+        },
+        {
+            "type": "wait_for_selector",
+            "selector": "text=/运行中|已启动|模拟交易运行|Started|Running/i",
+            "timeout_ms": 180_000,
+            "optional": True,
+        },
+        {"type": "screenshot", "path": "{run_artifact_dir}/sim_trade_final.png"},
+    ]
+
+
+def default_browser_config(strategy_url: str = "", *, flow: str = "backtest") -> dict[str, Any]:
+    """Return a selector/action template for JoinQuant browser automation."""
+    if flow not in {"backtest", "sim-trade", "sim_trade"}:
+        raise ValueError("flow must be 'backtest' or 'sim-trade'")
+    normalized_flow = "sim-trade" if flow == "sim_trade" else flow
+    actions = _sim_trade_actions() if normalized_flow == "sim-trade" else _backtest_actions()
+    return {
+        "flow": normalized_flow,
+        "strategy_url": strategy_url,
+        "login_url": DEFAULT_LOGIN_URL,
+        "headless": False,
+        "timeout_ms": 120_000,
+        "notes": [
+            "Fill strategy_url and selectors after inspecting your JoinQuant strategy page.",
+            "Use `joinquant browser-snapshot` to capture HTML/screenshots for selector tuning.",
+            "The action list is declarative and runs in order.",
+        ],
+        "actions": actions,
+    }
+
+
+def write_browser_config_template(
+    path: str | Path,
+    *,
+    strategy_url: str = "",
+    flow: str = "backtest",
+) -> Path:
+    """Write a JSON config template for browser automation."""
+    out_path = Path(path)
+    out_path.parent.mkdir(parents=True, exist_ok=True)
+    out_path.write_text(
+        json.dumps(
+            default_browser_config(strategy_url, flow=flow),
+            indent=2,
+            ensure_ascii=False,
+        ) + "\n",
+        encoding="utf-8",
+    )
+    return out_path
+
+
+def load_json(path: str | Path) -> dict[str, Any]:
+    """Load a JSON object from disk."""
+    data = json.loads(Path(path).read_text(encoding="utf-8"))
+    if not isinstance(data, dict):
+        raise ValueError(f"Expected JSON object in {path}")
+    return data
+
+
+def load_env_file(path: str | Path) -> dict[str, str]:
+    """Load simple KEY=VALUE env files without requiring shell-safe quoting."""
+    values: dict[str, str] = {}
+    for raw in Path(path).expanduser().read_text(encoding="utf-8").splitlines():
+        line = raw.strip()
+        if not line or line.startswith("#") or "=" not in line:
+            continue
+        key, value = line.split("=", 1)
+        clean = value.strip()
+        if len(clean) >= 2 and clean[0] == clean[-1] and clean[0] in {"'", '"'}:
+            clean = clean[1:-1]
+        else:
+            clean = clean.strip("'\"")
+        values[key.strip()] = clean
+    return values
+
+
+def _get_dotted(data: dict[str, Any], dotted: str) -> Any:
+    current: Any = data
+    for part in dotted.split("."):
+        if isinstance(current, dict) and part in current:
+            current = current[part]
+        else:
+            raise KeyError(dotted)
+    return current
+
+
+def _manifest_context(manifest: dict[str, Any], artifact_dir: Path, config: dict[str, Any]) -> dict[str, Any]:
+    targets_dir = Path(str(manifest["targets_dir"]))
+    target_csvs = [str(path) for path in sorted(targets_dir.glob("*.csv"))]
+    if not target_csvs:
+        raise ValueError(f"No target CSV files found under {targets_dir}")
+
+    target_dates = [
+        datetime.strptime(Path(path).stem, "%Y%m%d").strftime("%Y-%m-%d")
+        for path in target_csvs
+    ]
+    context = dict(manifest)
+    context.update({
+        "strategy_url": config.get("strategy_url", ""),
+        "target_csvs": target_csvs,
+        "target_csvs_csv": ",".join(target_csvs),
+        "backtest_start_date": config.get("backtest_start_date") or min(target_dates),
+        "backtest_end_date": config.get("backtest_end_date") or max(target_dates),
+        "run_artifact_dir": str(artifact_dir),
+    })
+    return context
+
+
+_TOKEN_RE = re.compile(r"^\{([A-Za-z_][A-Za-z0-9_]*(?:\.[A-Za-z_][A-Za-z0-9_]*)*)\}$")
+_PARTIAL_TOKEN_RE = re.compile(r"\{([A-Za-z_][A-Za-z0-9_]*(?:\.[A-Za-z_][A-Za-z0-9_]*)*)\}")
+
+
+def resolve_template(value: Any, context: dict[str, Any]) -> Any:
+    """Resolve ``{tokens}`` in config values against the manifest context."""
+    if isinstance(value, str):
+        whole = _TOKEN_RE.match(value)
+        if whole:
+            return _get_dotted(context, whole.group(1))
+
+        def replace(match: re.Match[str]) -> str:
+            resolved = _get_dotted(context, match.group(1))
+            return str(resolved)
+
+        return _PARTIAL_TOKEN_RE.sub(replace, value)
+    if isinstance(value, list):
+        return [resolve_template(item, context) for item in value]
+    if isinstance(value, dict):
+        return {key: resolve_template(val, context) for key, val in value.items()}
+    return value
+
+
+def save_login_state(
+    *,
+    storage_state: str | Path,
+    login_url: str = DEFAULT_LOGIN_URL,
+    headless: bool = False,
+    wait_seconds: int = 0,
+) -> Path:
+    """Open a browser for manual login and save the authenticated state."""
+    if headless and wait_seconds <= 0:
+        raise ValueError("headless browser-login requires --wait-seconds > 0")
+    sync_playwright, _ = _require_playwright()
+    state_path = Path(storage_state).expanduser()
+    state_path.parent.mkdir(parents=True, exist_ok=True)
+    with sync_playwright() as pw:
+        browser = pw.chromium.launch(headless=headless)
+        context = browser.new_context()
+        page = context.new_page()
+        page.goto(login_url, wait_until="domcontentloaded")
+        if wait_seconds > 0:
+            page.wait_for_timeout(wait_seconds * 1000)
+        else:
+            input("Log in to JoinQuant in the opened browser, then press Enter here to save state...")
+        context.storage_state(path=str(state_path))
+        browser.close()
+    state_path.chmod(0o600)
+    return state_path
+
+
+def save_login_state_from_env(
+    *,
+    env_path: str | Path,
+    storage_state: str | Path,
+    login_url: str = DEFAULT_LOGIN_URL,
+    headless: bool = True,
+    out_dir: str | Path | None = None,
+    timeout_ms: int = 120_000,
+) -> dict[str, Any]:
+    """Try to log in with env-file credentials and save browser state.
+
+    This handles the normal password-login form. If JoinQuant presents CAPTCHA,
+    slide verification, SMS verification, or 2FA, the report will mark the run
+    as not logged in and save a screenshot for manual diagnosis.
+    """
+    env = load_env_file(env_path)
+    username = env.get("JOINQUANT_USERNAME")
+    password = env.get("JOINQUANT_PASSWORD")
+    if not username or not password:
+        raise ValueError("JOINQUANT_USERNAME and JOINQUANT_PASSWORD are required")
+
+    sync_playwright, _ = _require_playwright()
+    state_path = Path(storage_state).expanduser()
+    state_path.parent.mkdir(parents=True, exist_ok=True)
+    artifact_dir = Path(out_dir or state_path.parent / "joinquant_login_artifacts")
+    artifact_dir.mkdir(parents=True, exist_ok=True)
+    screenshot_path = artifact_dir / "login_after_submit.png"
+    html_path = artifact_dir / "login_after_submit.html"
+
+    with sync_playwright() as pw:
+        browser = pw.chromium.launch(headless=headless)
+        context = browser.new_context()
+        page = context.new_page()
+        page.set_default_timeout(timeout_ms)
+        page.goto(login_url, wait_until="networkidle", timeout=timeout_ms)
+        page.locator('input[name="username"], input.pwd-phone').first.fill(username)
+        page.locator('input[name="pwd"], input.jq-login__password').first.fill(password)
+        checkbox = page.locator("#agreementBox, input.agreement-box").first
+        if checkbox.count():
+            checkbox.check(force=True)
+        page.locator("button.btnPwdSubmit, button.login-submit").first.click()
+        page.wait_for_timeout(5000)
+
+        html = page.content()
+        html_path.write_text(html, encoding="utf-8")
+        page.screenshot(path=str(screenshot_path), full_page=True)
+        login_inputs = page.locator('input[name="username"], input[name="pwd"]').count()
+        current_url = page.url
+        logged_in = login_inputs == 0 and "/user/login" not in current_url
+        if logged_in:
+            context.storage_state(path=str(state_path))
+            state_path.chmod(0o600)
+        browser.close()
+
+    report = {
+        "created_at": datetime.now(timezone.utc).isoformat(),
+        "logged_in": bool(logged_in),
+        "storage_state": str(state_path) if logged_in else "",
+        "artifact_dir": str(artifact_dir),
+        "screenshot": str(screenshot_path),
+        "html": str(html_path),
+        "current_url": current_url,
+        "notes": (
+            "Logged in and saved browser state."
+            if logged_in
+            else "Login did not complete. CAPTCHA/SMS/2FA or invalid credentials may be required."
+        ),
+    }
+    report_path = artifact_dir / "login_report.json"
+    report["report_path"] = str(report_path)
+    report_path.write_text(json.dumps(report, indent=2, ensure_ascii=False) + "\n", encoding="utf-8")
+    return report
+
+
+def browser_snapshot(
+    *,
+    url: str,
+    storage_state: str | Path,
+    out_dir: str | Path,
+    headless: bool = True,
+    timeout_ms: int = 60_000,
+) -> dict[str, Path]:
+    """Save a logged-in page screenshot and HTML for selector discovery."""
+    sync_playwright, _ = _require_playwright()
+    root = Path(out_dir)
+    root.mkdir(parents=True, exist_ok=True)
+    html_path = root / "page.html"
+    screenshot_path = root / "page.png"
+
+    with sync_playwright() as pw:
+        browser = pw.chromium.launch(headless=headless)
+        context = browser.new_context(storage_state=str(Path(storage_state).expanduser()))
+        page = context.new_page()
+        page.set_default_timeout(timeout_ms)
+        page.goto(url, wait_until="networkidle")
+        html_path.write_text(page.content(), encoding="utf-8")
+        page.screenshot(path=str(screenshot_path), full_page=True)
+        browser.close()
+    return {"html": html_path, "screenshot": screenshot_path}
+
+
+def _locator(page: Any, selector: str):
+    return page.locator(selector).first
+
+
+def _action_fail(action: dict[str, Any], exc: Exception) -> None:
+    if action.get("optional", False):
+        return
+    raise exc
+
+
+def _run_action(page: Any, action: dict[str, Any], context: dict[str, Any], timeout_default: int) -> dict[str, Any]:
+    resolved = resolve_template(action, context)
+    kind = resolved["type"]
+    timeout_ms = int(resolved.get("timeout_ms", timeout_default))
+    record: dict[str, Any] = {"type": kind, "status": "ok"}
+    try:
+        if kind == "goto":
+            page.goto(str(resolved["url"]), wait_until=resolved.get("wait_until", "domcontentloaded"), timeout=timeout_ms)
+        elif kind == "click":
+            _locator(page, str(resolved["selector"])).click(
+                timeout=timeout_ms,
+                force=bool(resolved.get("force", False)),
+            )
+        elif kind == "fill":
+            _locator(page, str(resolved["selector"])).fill(str(resolved.get("text", "")), timeout=timeout_ms)
+        elif kind == "press":
+            _locator(page, str(resolved["selector"])).press(str(resolved["key"]), timeout=timeout_ms)
+        elif kind == "set_input_files":
+            paths = resolved.get("paths", [])
+            if isinstance(paths, str):
+                paths = [path for path in paths.split(",") if path]
+            _locator(page, str(resolved["selector"])).set_input_files(paths, timeout=timeout_ms)
+            record["n_files"] = len(paths)
+        elif kind == "paste_text_file":
+            text = Path(str(resolved["path"])).read_text(encoding="utf-8")
+            loc = _locator(page, str(resolved["selector"]))
+            loc.click(timeout=timeout_ms)
+            page.keyboard.press("Control+A")
+            page.keyboard.insert_text(text)
+            record["n_chars"] = len(text)
+        elif kind == "set_ace_editor_file":
+            text = Path(str(resolved["path"])).read_text(encoding="utf-8")
+            selector = str(resolved.get("selector", ".ace_editor"))
+            page.wait_for_function(
+                "(selector) => Boolean(window.ace && document.querySelector(selector))",
+                arg=selector,
+                timeout=timeout_ms,
+            )
+            page.evaluate(
+                """(arg) => {
+                    const node = document.querySelector(arg.selector);
+                    if (!node || !window.ace) {
+                        throw new Error(`Ace editor not found for ${arg.selector}`);
+                    }
+                    const editor = window.ace.edit(node);
+                    editor.setValue(arg.text, -1);
+                    editor.clearSelection();
+                    editor.focus();
+                    return editor.getValue().length;
+                }""",
+                {"selector": selector, "text": text},
+            )
+            record["n_chars"] = len(text)
+        elif kind == "evaluate":
+            page.evaluate(str(resolved["script"]), resolved.get("arg"))
+        elif kind == "wait_for_selector":
+            page.wait_for_selector(str(resolved["selector"]), timeout=timeout_ms)
+        elif kind == "wait_for_timeout":
+            page.wait_for_timeout(int(resolved.get("timeout_ms", 1000)))
+        elif kind == "download":
+            save_as = Path(str(resolved["save_as"]))
+            save_as.parent.mkdir(parents=True, exist_ok=True)
+            with page.expect_download(timeout=timeout_ms) as download_info:
+                _locator(page, str(resolved["selector"])).click(timeout=timeout_ms)
+            download = download_info.value
+            download.save_as(str(save_as))
+            record["path"] = str(save_as)
+        elif kind == "screenshot":
+            path = Path(str(resolved["path"]))
+            path.parent.mkdir(parents=True, exist_ok=True)
+            page.screenshot(path=str(path), full_page=bool(resolved.get("full_page", True)))
+            record["path"] = str(path)
+        else:
+            raise ValueError(f"Unsupported browser action type: {kind}")
+    except Exception as exc:  # pragma: no cover - exercised only with Playwright.
+        record["status"] = "skipped" if resolved.get("optional", False) else "failed"
+        record["error"] = str(exc)
+        _action_fail(resolved, exc)
+    return record
+
+
+def run_browser_flow(
+    *,
+    manifest_path: str | Path,
+    config_path: str | Path,
+    storage_state: str | Path,
+    out_dir: str | Path | None = None,
+    headless: bool | None = None,
+    auto_reconcile: bool = True,
+    flow_name: str | None = None,
+) -> dict[str, Any]:
+    """Run configured browser automation for a JoinQuant web workflow."""
+    sync_playwright, _ = _require_playwright()
+    manifest = load_json(manifest_path)
+    config = load_json(config_path)
+    flow = flow_name or str(config.get("flow") or "browser")
+    artifact_dir = Path(out_dir or Path(str(manifest["joinquant_export_dir"])).parent / f"browser_{flow}")
+    artifact_dir.mkdir(parents=True, exist_ok=True)
+    context_vars = _manifest_context(manifest, artifact_dir, config)
+
+    timeout_ms = int(config.get("timeout_ms", 120_000))
+    actions = config.get("actions") or []
+    if not actions:
+        raise ValueError("Browser config contains no actions")
+
+    records: list[dict[str, Any]] = []
+    failure: Exception | None = None
+    failure_artifacts: dict[str, str] = {}
+    with sync_playwright() as pw:
+        browser = pw.chromium.launch(
+            headless=bool(config.get("headless", False) if headless is None else headless)
+        )
+        context = browser.new_context(
+            storage_state=str(Path(storage_state).expanduser()),
+            accept_downloads=True,
+        )
+        page = context.new_page()
+        page.set_default_timeout(timeout_ms)
+        try:
+            for action in actions:
+                records.append(_run_action(page, action, context_vars, timeout_ms))
+        except Exception as exc:  # pragma: no cover - requires live browser UI.
+            failure = exc
+            screenshot_path = artifact_dir / "failure.png"
+            html_path = artifact_dir / "failure.html"
+            try:
+                page.screenshot(path=str(screenshot_path), full_page=True)
+                html_path.write_text(page.content(), encoding="utf-8")
+                failure_artifacts = {
+                    "screenshot": str(screenshot_path),
+                    "html": str(html_path),
+                }
+            except Exception as artifact_exc:
+                failure_artifacts = {"artifact_error": str(artifact_exc)}
+            records.append({
+                "type": "browser_flow",
+                "status": "failed",
+                "error": str(exc),
+                **failure_artifacts,
+            })
+        browser.close()
+
+    expected = manifest.get("expected_joinquant_csvs", {})
+    downloaded = {
+        key: str(path)
+        for key, value in expected.items()
+        if (path := Path(str(value))).exists()
+    }
+    reconcile_paths: dict[str, str] = {}
+    if auto_reconcile and {"fills", "positions", "pnl"}.issubset(downloaded):
+        ingested = ingest_joinquant_outputs(
+            portfolio_name=str(manifest["portfolio_name"]),
+            fills_csv=downloaded["fills"],
+            positions_csv=downloaded["positions"],
+            pnl_csv=downloaded["pnl"],
+            out_dir=Path(str(manifest["joinquant_export_dir"])).parent / "ingested",
+        )
+        reconciled = reconcile_joinquant(
+            portfolio_name=str(manifest["portfolio_name"]),
+            targets_dir=str(manifest["targets_dir"]),
+            our_fills_path=str(manifest["fills_path"]),
+            our_positions_path=str(manifest["positions_path"]),
+            our_pnl_path=str(manifest["pnl_path"]),
+            jq_fills_path=str(ingested["fills"]),
+            jq_positions_path=str(ingested["positions"]),
+            jq_pnl_path=str(ingested["pnl"]),
+            out_dir=Path(str(manifest["joinquant_export_dir"])).parent / "reconcile",
+        )
+        reconcile_paths = {key: str(value) for key, value in reconciled.items()}
+
+    report = {
+        "created_at": datetime.now(timezone.utc).isoformat(),
+        "manifest_path": str(manifest_path),
+        "config_path": str(config_path),
+        "flow": flow,
+        "status": "failed" if failure else "ok",
+        "storage_state": str(storage_state),
+        "artifact_dir": str(artifact_dir),
+        "actions": records,
+        "downloaded": downloaded,
+        "reconcile_paths": reconcile_paths,
+    }
+    report_path = artifact_dir / "browser_run_report.json"
+    report["report_path"] = str(report_path)
+    report_path.write_text(
+        json.dumps(report, indent=2, ensure_ascii=False) + "\n",
+        encoding="utf-8",
+    )
+    if failure is not None:
+        raise RuntimeError(
+            f"Browser flow failed; report saved to {report_path}: {failure}"
+        ) from failure
+    return report
+
+
+def run_browser_backtest(
+    *,
+    manifest_path: str | Path,
+    config_path: str | Path,
+    storage_state: str | Path,
+    out_dir: str | Path | None = None,
+    headless: bool | None = None,
+    auto_reconcile: bool = True,
+) -> dict[str, Any]:
+    """Run configured browser automation for a JoinQuant backtest."""
+    return run_browser_flow(
+        manifest_path=manifest_path,
+        config_path=config_path,
+        storage_state=storage_state,
+        out_dir=out_dir,
+        headless=headless,
+        auto_reconcile=auto_reconcile,
+        flow_name="backtest",
+    )
+
+
+def run_browser_sim_trade(
+    *,
+    manifest_path: str | Path,
+    config_path: str | Path,
+    storage_state: str | Path,
+    out_dir: str | Path | None = None,
+    headless: bool | None = None,
+    auto_reconcile: bool = True,
+) -> dict[str, Any]:
+    """Run configured browser automation for JoinQuant 模拟盘."""
+    return run_browser_flow(
+        manifest_path=manifest_path,
+        config_path=config_path,
+        storage_state=storage_state,
+        out_dir=out_dir,
+        headless=headless,
+        auto_reconcile=auto_reconcile,
+        flow_name="sim-trade",
+    )

plugins/joinquant/cli.py

@@ -0,0 +1,426 @@
+"""CLI commands for the JoinQuant comparison plugin."""
+
+from __future__ import annotations
+
+import click
+
+from plugins.joinquant.browser import (
+    browser_snapshot,
+    load_env_file,
+    run_browser_backtest,
+    run_browser_sim_trade,
+    save_login_state,
+    save_login_state_from_env,
+    write_browser_config_template,
+)
+from plugins.joinquant.export_targets import export_targets
+from plugins.joinquant.ingest import ingest_joinquant_outputs
+from plugins.joinquant.reconcile import reconcile_joinquant
+from plugins.joinquant.smoke import prepare_smoke_test
+from plugins.joinquant.wrapper_strategy import write_wrapper_strategy
+
+
+@click.group(name="joinquant")
+def joinquant():
+    """Compare internal portfolio simulation with JoinQuant output."""
+
+
+@joinquant.command("export-targets")
+@click.option("--positions-path", required=True, help="Portfolio positions parquet from `portfolio build`")
+@click.option("--portfolio-name", required=True, help="Portfolio run to export")
+@click.option(
+    "--mode",
+    "mode",
+    type=click.Choice(["target_shares", "target_value"]),
+    default="target_shares",
+    show_default=True,
+    help="JoinQuant target order mode",
+)
+@click.option("--start-date", default=None, help="Inclusive YYYY-MM-DD start date")
+@click.option("--end-date", default=None, help="Inclusive YYYY-MM-DD end date")
+@click.option(
+    "--execution-calendar-path",
+    default=None,
+    help="Daily data parquet/dataset used to shift position dates to next execution session",
+)
+@click.option("--out-dir", default="plugins_output/joinquant/targets", show_default=True)
+@click.option("--force", is_flag=True, help="Overwrite frozen target/snapshot files")
+def export_targets_cmd(
+    positions_path,
+    portfolio_name,
+    mode,
+    start_date,
+    end_date,
+    execution_calendar_path,
+    out_dir,
+    force,
+):
+    """Export frozen daily target files for JoinQuant."""
+    snapshots = export_targets(
+        positions_path=positions_path,
+        portfolio_name=portfolio_name,
+        mode=mode,
+        start_date=start_date,
+        end_date=end_date,
+        execution_calendar_path=execution_calendar_path,
+        out_dir=out_dir,
+        force=force,
+    )
+    click.echo(f"Exported JoinQuant targets: {len(snapshots)} day(s)")
+    for snapshot in snapshots:
+        click.echo(
+            f"  {snapshot['date']}: {snapshot['n_symbols']} symbols, "
+            f"sha256={str(snapshot['file_sha256'])[:12]}"
+        )
+
+
+@joinquant.command("ingest")
+@click.option("--portfolio-name", required=True, help="Portfolio run name")
+@click.option("--fills-csv", required=True, help="JoinQuant fills CSV")
+@click.option("--positions-csv", required=True, help="JoinQuant positions CSV")
+@click.option("--pnl-csv", required=True, help="JoinQuant daily PnL CSV")
+@click.option("--out-dir", default="plugins_output/joinquant/ingested", show_default=True)
+def ingest_cmd(portfolio_name, fills_csv, positions_csv, pnl_csv, out_dir):
+    """Normalize JoinQuant CSV exports to parquet."""
+    paths = ingest_joinquant_outputs(
+        portfolio_name=portfolio_name,
+        fills_csv=fills_csv,
+        positions_csv=positions_csv,
+        pnl_csv=pnl_csv,
+        out_dir=out_dir,
+    )
+    click.echo(f"Saved JoinQuant fills:     {paths['fills']}")
+    click.echo(f"Saved JoinQuant positions: {paths['positions']}")
+    click.echo(f"Saved JoinQuant pnl:       {paths['pnl']}")
+
+
+@joinquant.command("reconcile")
+@click.option("--portfolio-name", required=True, help="Portfolio run name")
+@click.option("--targets-dir", required=True, help="Directory containing exported daily target files")
+@click.option("--our-fills-path", required=True, help="Internal simulator fills parquet")
+@click.option("--our-positions-path", required=True, help="Internal portfolio positions parquet")
+@click.option("--our-pnl-path", required=True, help="Internal simulator PnL parquet")
+@click.option("--jq-fills-path", required=True, help="Normalized JoinQuant fills parquet")
+@click.option("--jq-positions-path", required=True, help="Normalized JoinQuant positions parquet")
+@click.option("--jq-pnl-path", required=True, help="Normalized JoinQuant PnL parquet")
+@click.option("--out-dir", default="plugins_output/joinquant/reconcile", show_default=True)
+@click.option("--share-tolerance", default=0.0, show_default=True, type=float)
+@click.option("--price-rel-tolerance", default=1e-4, show_default=True, type=float)
+@click.option("--pnl-tolerance", default=1.0, show_default=True, type=float)
+@click.option("--booksize", default=None, type=float, help="Booksize for value tolerance inference")
+def reconcile_cmd(
+    portfolio_name,
+    targets_dir,
+    our_fills_path,
+    our_positions_path,
+    our_pnl_path,
+    jq_fills_path,
+    jq_positions_path,
+    jq_pnl_path,
+    out_dir,
+    share_tolerance,
+    price_rel_tolerance,
+    pnl_tolerance,
+    booksize,
+):
+    """Write per-symbol and daily JoinQuant reconciliation reports."""
+    paths = reconcile_joinquant(
+        portfolio_name=portfolio_name,
+        targets_dir=targets_dir,
+        our_fills_path=our_fills_path,
+        our_positions_path=our_positions_path,
+        our_pnl_path=our_pnl_path,
+        jq_fills_path=jq_fills_path,
+        jq_positions_path=jq_positions_path,
+        jq_pnl_path=jq_pnl_path,
+        out_dir=out_dir,
+        share_tolerance=share_tolerance,
+        price_rel_tolerance=price_rel_tolerance,
+        pnl_tolerance=pnl_tolerance,
+        booksize=booksize,
+    )
+    click.echo(f"Saved reconciliation parquet: {paths['daily_reconcile']}")
+    click.echo(f"Saved reconciliation summary: {paths['summary_md']}")
+    click.echo(f"Saved reconciliation CSV:     {paths['summary_csv']}")
+
+
+@joinquant.command("write-wrapper")
+@click.option("--portfolio-name", required=True, help="Portfolio run name")
+@click.option(
+    "--mode",
+    "mode",
+    type=click.Choice(["target_shares", "target_value"]),
+    default="target_shares",
+    show_default=True,
+)
+@click.option("--out-path", required=True, help="Path for generated standalone strategy")
+@click.option("--allow-short", is_flag=True, help="Do not clip negative targets in the generated wrapper")
+@click.option(
+    "--embed-targets-dir",
+    default=None,
+    help="Embed CSV target files from this directory into the strategy source",
+)
+def write_wrapper_cmd(portfolio_name, mode, out_path, allow_short, embed_targets_dir):
+    """Generate a standalone JoinQuant wrapper strategy."""
+    path = write_wrapper_strategy(
+        portfolio_name=portfolio_name,
+        mode=mode,
+        out_path=out_path,
+        allow_short=allow_short,
+        embedded_targets_dir=embed_targets_dir,
+    )
+    click.echo(f"Saved JoinQuant wrapper strategy: {path}")
+
+
+@joinquant.command("prepare-smoke")
+@click.option("--out-dir", required=True, help="Root directory for generated smoke-test artifacts")
+@click.option(
+    "--universe",
+    default="sh600000,sz000001,sh600519,sz002594,sz300750",
+    show_default=True,
+    help="Universe for the tiny real-data download",
+)
+@click.option("--trade-symbol", default="sh600000", show_default=True)
+@click.option("--start-date", default="2024-01-02", show_default=True)
+@click.option("--end-date", default="2024-01-12", show_default=True)
+@click.option("--portfolio-name", default="jq_smoke_one_stock_long", show_default=True)
+@click.option("--shares", default=1000, show_default=True, type=int)
+@click.option("--booksize", default=1_000_000.0, show_default=True, type=float)
+@click.option("--max-signal-dates", default=3, show_default=True, type=int)
+@click.option("--cost-bps", default=5.0, show_default=True, type=float)
+@click.option("--slippage-bps", default=5.0, show_default=True, type=float)
+@click.option("--volume-frac", default=0.02, show_default=True, type=float)
+@click.option("--force", is_flag=True, help="Overwrite existing frozen target files")
+def prepare_smoke_cmd(
+    out_dir,
+    universe,
+    trade_symbol,
+    start_date,
+    end_date,
+    portfolio_name,
+    shares,
+    booksize,
+    max_signal_dates,
+    cost_bps,
+    slippage_bps,
+    volume_frac,
+    force,
+):
+    """Prepare a one-command local real-data JoinQuant smoke test."""
+    manifest = prepare_smoke_test(
+        out_dir=out_dir,
+        universe=universe,
+        trade_symbol=trade_symbol,
+        start_date=start_date,
+        end_date=end_date,
+        portfolio_name=portfolio_name,
+        shares=shares,
+        booksize=booksize,
+        max_signal_dates=max_signal_dates,
+        cost_bps=cost_bps,
+        slippage_bps=slippage_bps,
+        volume_frac=volume_frac,
+        force=force,
+    )
+    click.echo(f"Prepared JoinQuant smoke manifest: {manifest['manifest_path']}")
+    click.echo(f"Wrapper: {manifest['wrapper_path']}")
+    click.echo(f"Targets: {manifest['targets_dir']}")
+    click.echo(f"Expected JoinQuant exports: {manifest['joinquant_export_dir']}")
+    summary = manifest["local_summary"]
+    click.echo(
+        f"Local simulator: {summary['n_pnl_rows']} days, "
+        f"PnL={summary['total_pnl']:,.2f}, cost={summary['total_cost']:,.2f}, "
+        f"blocked={summary['blocked_trades']}"
+    )
+
+
+@joinquant.command("browser-login")
+@click.option(
+    "--storage-state",
+    default="~/.config/chinese-equity-quant/joinquant_storage_state.json",
+    show_default=True,
+    help="Path where the authenticated browser state is stored",
+)
+@click.option(
+    "--login-url",
+    default="https://www.joinquant.com/user/login/index",
+    show_default=True,
+)
+@click.option("--headless", is_flag=True, help="Use a headless browser")
+@click.option(
+    "--wait-seconds",
+    default=0,
+    show_default=True,
+    type=int,
+    help="Seconds to wait before saving state; 0 prompts for Enter after login",
+)
+def browser_login_cmd(storage_state, login_url, headless, wait_seconds):
+    """Open JoinQuant login and save reusable browser session state."""
+    path = save_login_state(
+        storage_state=storage_state,
+        login_url=login_url,
+        headless=headless,
+        wait_seconds=wait_seconds,
+    )
+    click.echo(f"Saved JoinQuant browser state: {path}")
+
+
+@joinquant.command("browser-login-env")
+@click.option(
+    "--env-path",
+    default="~/.config/chinese-equity-quant/joinquant.env",
+    show_default=True,
+    help="Env file with JOINQUANT_USERNAME and JOINQUANT_PASSWORD",
+)
+@click.option(
+    "--storage-state",
+    default="~/.config/chinese-equity-quant/joinquant_storage_state.json",
+    show_default=True,
+    help="Path where the authenticated browser state is stored",
+)
+@click.option(
+    "--login-url",
+    default="https://www.joinquant.com/user/login/index",
+    show_default=True,
+)
+@click.option("--headed", is_flag=True, help="Run with a visible browser")
+@click.option("--out-dir", default=None, help="Login artifact directory")
+@click.option("--timeout-ms", default=120_000, show_default=True, type=int)
+def browser_login_env_cmd(env_path, storage_state, login_url, headed, out_dir, timeout_ms):
+    """Try credential-based JoinQuant login from an env file."""
+    env = load_env_file(env_path)
+    missing = [
+        key for key in ["JOINQUANT_USERNAME", "JOINQUANT_PASSWORD"]
+        if not env.get(key)
+    ]
+    if missing:
+        raise click.ClickException(f"Missing required env keys: {', '.join(missing)}")
+    report = save_login_state_from_env(
+        env_path=env_path,
+        storage_state=storage_state,
+        login_url=login_url,
+        headless=not headed,
+        out_dir=out_dir,
+        timeout_ms=timeout_ms,
+    )
+    click.echo(f"Saved login report: {report['report_path']}")
+    if report["logged_in"]:
+        click.echo(f"Saved JoinQuant browser state: {report['storage_state']}")
+    else:
+        raise click.ClickException(
+            "JoinQuant login did not complete. Check the saved screenshot/HTML; "
+            "CAPTCHA/SMS/2FA may require `joinquant browser-login` once."
+        )
+
+
+@joinquant.command("write-browser-config")
+@click.option("--out-path", required=True, help="Path for JSON browser automation config")
+@click.option("--strategy-url", default="", help="JoinQuant strategy edit/backtest/模拟盘 URL")
+@click.option(
+    "--flow",
+    type=click.Choice(["backtest", "sim-trade"]),
+    default="backtest",
+    show_default=True,
+    help="Browser automation template to write",
+)
+def write_browser_config_cmd(out_path, strategy_url, flow):
+    """Write a browser automation selector/action config template."""
+    path = write_browser_config_template(out_path, strategy_url=strategy_url, flow=flow)
+    click.echo(f"Saved JoinQuant browser config template: {path}")
+
+
+@joinquant.command("browser-snapshot")
+@click.option("--url", required=True, help="Logged-in JoinQuant page URL to inspect")
+@click.option(
+    "--storage-state",
+    default="~/.config/chinese-equity-quant/joinquant_storage_state.json",
+    show_default=True,
+)
+@click.option("--out-dir", required=True, help="Directory for page.html and page.png")
+@click.option("--headed", is_flag=True, help="Run with a visible browser")
+@click.option("--timeout-ms", default=60_000, show_default=True, type=int)
+def browser_snapshot_cmd(url, storage_state, out_dir, headed, timeout_ms):
+    """Save a page screenshot and HTML for selector discovery."""
+    paths = browser_snapshot(
+        url=url,
+        storage_state=storage_state,
+        out_dir=out_dir,
+        headless=not headed,
+        timeout_ms=timeout_ms,
+    )
+    click.echo(f"Saved HTML:       {paths['html']}")
+    click.echo(f"Saved screenshot: {paths['screenshot']}")
+
+
+@joinquant.command("run-browser-backtest")
+@click.option("--manifest-path", required=True, help="Manifest from `joinquant prepare-smoke`")
+@click.option("--config-path", required=True, help="JSON browser automation config")
+@click.option(
+    "--storage-state",
+    default="~/.config/chinese-equity-quant/joinquant_storage_state.json",
+    show_default=True,
+)
+@click.option("--out-dir", default=None, help="Browser-run artifact directory")
+@click.option("--headed", is_flag=True, help="Run with a visible browser")
+@click.option("--no-auto-reconcile", is_flag=True, help="Skip ingest/reconcile after downloads")
+def run_browser_backtest_cmd(
+    manifest_path,
+    config_path,
+    storage_state,
+    out_dir,
+    headed,
+    no_auto_reconcile,
+):
+    """Run JoinQuant browser automation from manifest and config."""
+    report = run_browser_backtest(
+        manifest_path=manifest_path,
+        config_path=config_path,
+        storage_state=storage_state,
+        out_dir=out_dir,
+        headless=not headed,
+        auto_reconcile=not no_auto_reconcile,
+    )
+    click.echo(f"Saved browser run report: {report['report_path']}")
+    if report["downloaded"]:
+        click.echo("Downloaded JoinQuant CSVs:")
+        for key, path in report["downloaded"].items():
+            click.echo(f"  {key}: {path}")
+    if report["reconcile_paths"]:
+        click.echo(f"Saved reconciliation summary: {report['reconcile_paths']['summary_md']}")
+
+
+@joinquant.command("run-browser-sim")
+@click.option("--manifest-path", required=True, help="Manifest from target preparation")
+@click.option("--config-path", required=True, help="JSON simulated-trading browser config")
+@click.option(
+    "--storage-state",
+    default="~/.config/chinese-equity-quant/joinquant_storage_state.json",
+    show_default=True,
+)
+@click.option("--out-dir", default=None, help="Browser-run artifact directory")
+@click.option("--headed", is_flag=True, help="Run with a visible browser")
+@click.option("--no-auto-reconcile", is_flag=True, help="Skip ingest/reconcile after downloads")
+def run_browser_sim_cmd(
+    manifest_path,
+    config_path,
+    storage_state,
+    out_dir,
+    headed,
+    no_auto_reconcile,
+):
+    """Run JoinQuant 模拟盘 browser automation from manifest and config."""
+    report = run_browser_sim_trade(
+        manifest_path=manifest_path,
+        config_path=config_path,
+        storage_state=storage_state,
+        out_dir=out_dir,
+        headless=not headed,
+        auto_reconcile=not no_auto_reconcile,
+    )
+    click.echo(f"Saved browser sim-trade report: {report['report_path']}")
+    if report["downloaded"]:
+        click.echo("Downloaded JoinQuant CSVs:")
+        for key, path in report["downloaded"].items():
+            click.echo(f"  {key}: {path}")
+    if report["reconcile_paths"]:
+        click.echo(f"Saved reconciliation summary: {report['reconcile_paths']['summary_md']}")

plugins/joinquant/export_targets.py

@@ -0,0 +1,248 @@
+"""Export portfolio positions as frozen JoinQuant target files."""
+
+from __future__ import annotations
+
+import hashlib
+import json
+import uuid
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Iterable, Literal
+
+import pandas as pd
+
+from pipeline.common.schema import POSITION_COLUMNS
+from plugins.joinquant.schema import JOINQUANT_TARGET_COLUMNS
+from plugins.joinquant.symbols import to_joinquant_symbol
+
+
+ExportMode = Literal["target_shares", "target_value"]
+
+
+def _date_text(value: object) -> str:
+    return pd.Timestamp(value).strftime("%Y-%m-%d")
+
+
+def _date_file_stem(date_text: str) -> str:
+    return pd.Timestamp(date_text).strftime("%Y%m%d")
+
+
+def _snapshot_root_for(targets_root: Path) -> Path:
+    if targets_root.name == "targets":
+        return targets_root.parent / "snapshots"
+    return targets_root / "snapshots"
+
+
+def _sha256_file(path: Path) -> str:
+    digest = hashlib.sha256()
+    with path.open("rb") as fh:
+        for chunk in iter(lambda: fh.read(1024 * 1024), b""):
+            digest.update(chunk)
+    return digest.hexdigest()
+
+
+def _check_position_columns(df: pd.DataFrame) -> None:
+    missing = [col for col in POSITION_COLUMNS if col not in df.columns]
+    if missing:
+        raise ValueError(f"Positions input missing required columns: {missing}")
+
+
+def _filter_dates(
+    df: pd.DataFrame,
+    start_date: str | None,
+    end_date: str | None,
+    *,
+    date_column: str = "date",
+) -> pd.DataFrame:
+    out = df.copy()
+    out[date_column] = pd.to_datetime(out[date_column]).dt.normalize()
+    if start_date:
+        out = out[out[date_column] >= pd.Timestamp(start_date).normalize()]
+    if end_date:
+        out = out[out[date_column] <= pd.Timestamp(end_date).normalize()]
+    return out
+
+
+def _read_execution_calendar(path: str | Path) -> pd.DatetimeIndex:
+    data = pd.read_parquet(path)
+    if "date" not in data.columns:
+        raise ValueError("execution calendar parquet must contain a 'date' column")
+    dates = pd.to_datetime(data["date"], errors="coerce").dropna().dt.normalize()
+    return pd.DatetimeIndex(sorted(dates.unique()))
+
+
+def _apply_execution_calendar(df: pd.DataFrame, calendar_path: str | Path) -> pd.DataFrame:
+    calendar = _read_execution_calendar(calendar_path)
+    if calendar.empty:
+        raise ValueError("execution calendar contains no dates")
+
+    source_dates = pd.to_datetime(df["date"]).dt.normalize()
+    positions = calendar.searchsorted(source_dates, side="right")
+    out = df.copy()
+    out["source_date"] = source_dates
+    out["export_date"] = pd.NaT
+    valid = positions < len(calendar)
+    out.loc[valid, "export_date"] = calendar.take(positions[valid])
+    return out[out["export_date"].notna()].copy()
+
+
+def build_target_frame(
+    positions: pd.DataFrame,
+    *,
+    portfolio_name: str | None = None,
+    mode: ExportMode = "target_shares",
+    start_date: str | None = None,
+    end_date: str | None = None,
+    snapshot_ids: dict[str, str] | None = None,
+    execution_calendar_path: str | Path | None = None,
+) -> pd.DataFrame:
+    """Build normalized JoinQuant target rows from portfolio positions.
+
+    ``target_shares`` is populated from ``position_shares`` because the core
+    simulator executes the discretized book, not continuous research shares.
+    """
+    if mode not in {"target_shares", "target_value"}:
+        raise ValueError("mode must be 'target_shares' or 'target_value'")
+    _check_position_columns(positions)
+
+    df = positions.copy()
+    df["date"] = pd.to_datetime(df["date"]).dt.normalize()
+    if execution_calendar_path is not None:
+        df = _apply_execution_calendar(df, execution_calendar_path)
+        df = df.drop(columns=["date"]).rename(columns={"export_date": "date"})
+        df["source_date"] = pd.to_datetime(df["source_date"]).dt.strftime("%Y-%m-%d")
+        df = _filter_dates(df, start_date, end_date)
+    else:
+        df = _filter_dates(df, start_date, end_date)
+    if portfolio_name is not None:
+        df = df[df["portfolio_name"].astype(str) == portfolio_name]
+    if df.empty:
+        return pd.DataFrame(columns=JOINQUANT_TARGET_COLUMNS)
+
+    out = pd.DataFrame({
+        "date": df["date"].map(_date_text),
+        "portfolio_name": df["portfolio_name"].astype(str),
+        "symbol_id": df["symbol_id"].astype(str),
+        "jq_symbol": df["symbol_id"].map(to_joinquant_symbol),
+        "target_shares": pd.to_numeric(df["position_shares"], errors="coerce").fillna(0).astype("int64"),
+        "target_value": pd.to_numeric(df["target_value"], errors="coerce").fillna(0.0),
+        "target_weight": pd.to_numeric(df["target_weight"], errors="coerce").fillna(0.0),
+        "export_mode": mode,
+        "snapshot_id": "",
+    })
+
+    if snapshot_ids:
+        out["snapshot_id"] = out["date"].map(snapshot_ids).fillna("")
+
+    return out[JOINQUANT_TARGET_COLUMNS].sort_values(
+        ["date", "portfolio_name", "symbol_id"]
+    ).reset_index(drop=True)
+
+
+def export_targets(
+    positions_path: str | Path,
+    *,
+    portfolio_name: str,
+    mode: ExportMode = "target_shares",
+    out_dir: str | Path = "plugins_output/joinquant/targets",
+    start_date: str | None = None,
+    end_date: str | None = None,
+    execution_calendar_path: str | Path | None = None,
+    force: bool = False,
+) -> list[dict[str, object]]:
+    """Export one daily CSV/parquet target file plus a snapshot JSON per date.
+
+    Args:
+        positions_path: Parquet file produced by ``portfolio build``.
+        portfolio_name: Portfolio run to export.
+        mode: ``target_shares`` or ``target_value``.
+        out_dir: Target root. Files are written to ``out_dir/portfolio_name``.
+            If the root is named ``targets``, snapshots are written to the
+            sibling ``snapshots`` directory.
+        start_date: Optional inclusive start date.
+        end_date: Optional inclusive end date.
+        execution_calendar_path: Optional daily-bar parquet dataset used to
+            shift each position date to the next available execution session.
+            This matches the internal simulator's next-open convention.
+        force: If false, existing target or snapshot files are treated as
+            frozen and cause ``FileExistsError``.
+
+    Returns:
+        Snapshot metadata dictionaries, one per exported date.
+    """
+    positions_path = Path(positions_path)
+    targets_root = Path(out_dir)
+    snapshot_root = _snapshot_root_for(targets_root)
+    targets_portfolio_dir = targets_root / portfolio_name
+    snapshots_portfolio_dir = snapshot_root / portfolio_name
+    targets_portfolio_dir.mkdir(parents=True, exist_ok=True)
+    snapshots_portfolio_dir.mkdir(parents=True, exist_ok=True)
+
+    positions = pd.read_parquet(positions_path)
+    filtered = positions.copy()
+    filtered["date"] = pd.to_datetime(filtered["date"]).dt.normalize()
+    if execution_calendar_path is not None:
+        filtered = _apply_execution_calendar(filtered, execution_calendar_path)
+        filtered = filtered.drop(columns=["date"]).rename(columns={"export_date": "date"})
+        filtered["source_date"] = pd.to_datetime(filtered["source_date"]).dt.strftime("%Y-%m-%d")
+        filtered = _filter_dates(filtered, start_date, end_date)
+    else:
+        filtered = _filter_dates(filtered, start_date, end_date)
+    filtered = filtered[filtered["portfolio_name"].astype(str) == portfolio_name]
+    if filtered.empty:
+        return []
+
+    date_texts = sorted(filtered["date"].map(_date_text).unique())
+    snapshot_ids = {
+        date_text: f"jq-{portfolio_name}-{date_text}-{uuid.uuid4().hex[:12]}"
+        for date_text in date_texts
+    }
+    targets = build_target_frame(
+        filtered,
+        portfolio_name=portfolio_name,
+        mode=mode,
+        snapshot_ids=snapshot_ids,
+        execution_calendar_path=None,
+    )
+
+    snapshots: list[dict[str, object]] = []
+    for date_text, daily in targets.groupby("date", sort=True):
+        stem = _date_file_stem(date_text)
+        csv_path = targets_portfolio_dir / f"{stem}.csv"
+        parquet_path = targets_portfolio_dir / f"{stem}.parquet"
+        snapshot_path = snapshots_portfolio_dir / f"{stem}.json"
+        existing: Iterable[Path] = (csv_path, parquet_path, snapshot_path)
+        if not force:
+            conflicts = [str(path) for path in existing if path.exists()]
+            if conflicts:
+                raise FileExistsError(
+                    "Frozen JoinQuant target already exists; use --force to overwrite: "
+                    + ", ".join(conflicts)
+                )
+
+        daily = daily[JOINQUANT_TARGET_COLUMNS].reset_index(drop=True)
+        daily.to_csv(csv_path, index=False)
+        daily.to_parquet(parquet_path, index=False)
+        file_hash = _sha256_file(csv_path)
+
+        snapshot = {
+            "snapshot_id": snapshot_ids[date_text],
+            "portfolio_name": portfolio_name,
+            "date": date_text,
+            "export_mode": mode,
+            "source_positions_path": str(positions_path),
+            "execution_calendar_path": str(execution_calendar_path) if execution_calendar_path else None,
+            "created_at": datetime.now(timezone.utc).isoformat(),
+            "n_symbols": int(len(daily)),
+            "file_sha256": file_hash,
+            "notes": "Frozen JoinQuant target file.",
+            "target_csv_path": str(csv_path),
+            "target_parquet_path": str(parquet_path),
+        }
+        snapshot_path.write_text(
+            json.dumps(snapshot, indent=2, ensure_ascii=False) + "\n",
+            encoding="utf-8",
+        )
+        snapshots.append(snapshot)
+
+    return snapshots

plugins/joinquant/ingest.py

@@ -0,0 +1,225 @@
+"""Normalize JoinQuant CSV exports into plugin parquet schemas."""
+
+from __future__ import annotations
+
+import re
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+from plugins.joinquant.schema import (
+    JOINQUANT_FILL_COLUMNS,
+    JOINQUANT_PNL_COLUMNS,
+    JOINQUANT_POSITION_COLUMNS,
+)
+from plugins.joinquant.symbols import normalize_symbol_pair, to_joinquant_symbol
+
+
+def _clean_name(name: object) -> str:
+    text = str(name).strip().lower()
+    text = re.sub(r"[\s\-.()/]+", "_", text)
+    return re.sub(r"[^0-9a-z_]+", "", text).strip("_")
+
+
+def _clean_columns(df: pd.DataFrame) -> pd.DataFrame:
+    out = df.copy()
+    out.columns = [_clean_name(col) for col in out.columns]
+    return out
+
+
+def _pick(df: pd.DataFrame, candidates: list[str]) -> str | None:
+    for candidate in candidates:
+        clean = _clean_name(candidate)
+        if clean in df.columns:
+            return clean
+    return None
+
+
+def _series_or_default(df: pd.DataFrame, candidates: list[str], default: object) -> pd.Series:
+    col = _pick(df, candidates)
+    if col is None:
+        return pd.Series([default] * len(df), index=df.index)
+    return df[col]
+
+
+def _date_series(df: pd.DataFrame) -> pd.Series:
+    values = _series_or_default(
+        df,
+        ["date", "trade_date", "datetime", "time", "created_at"],
+        pd.NaT,
+    )
+    parsed = pd.to_datetime(values, errors="coerce").dt.normalize()
+    return parsed.dt.strftime("%Y-%m-%d").fillna("")
+
+
+def _numeric(values: pd.Series, default: float = 0.0) -> pd.Series:
+    return pd.to_numeric(values, errors="coerce").replace([np.inf, -np.inf], np.nan).fillna(default)
+
+
+def _text(values: pd.Series, default: str = "") -> pd.Series:
+    return values.fillna(default).astype(str)
+
+
+def _portfolio_series(df: pd.DataFrame, portfolio_name: str) -> pd.Series:
+    return _text(_series_or_default(df, ["portfolio_name", "portfolio", "strategy"], portfolio_name), portfolio_name)
+
+
+def _symbol_frame(df: pd.DataFrame) -> pd.DataFrame:
+    internal_col = _pick(df, ["symbol_id", "internal_symbol"])
+    jq_col = _pick(df, ["jq_symbol", "security", "stock", "symbol", "code", "order_book_id"])
+
+    symbol_ids: list[str] = []
+    jq_symbols: list[str] = []
+    for idx in df.index:
+        internal = df.at[idx, internal_col] if internal_col else None
+        jq_value = df.at[idx, jq_col] if jq_col else None
+        value = internal if internal is not None and str(internal).strip() else jq_value
+        if value is None or not str(value).strip():
+            symbol_ids.append("")
+            jq_symbols.append("")
+            continue
+        symbol_id, jq_symbol = normalize_symbol_pair(value)
+        if jq_value is not None and str(jq_value).strip():
+            try:
+                _, jq_symbol = normalize_symbol_pair(jq_value)
+            except ValueError:
+                jq_symbol = to_joinquant_symbol(symbol_id)
+        symbol_ids.append(symbol_id)
+        jq_symbols.append(jq_symbol)
+
+    return pd.DataFrame({"symbol_id": symbol_ids, "jq_symbol": jq_symbols}, index=df.index)
+
+
+def _signed_shares(shares: pd.Series, side: pd.Series) -> pd.Series:
+    signed = _numeric(shares, 0.0)
+    side_text = side.fillna("").astype(str).str.lower()
+    sell = side_text.str.contains("sell|short|close|reduce|-", regex=True)
+    buy = side_text.str.contains("buy|long|open|add|\\+", regex=True)
+    signed = signed.abs()
+    signed = signed.mask(sell, -signed)
+    signed = signed.mask(~(sell | buy), _numeric(shares, 0.0))
+    return signed
+
+
+def normalize_fills_csv(path: str | Path, portfolio_name: str) -> pd.DataFrame:
+    """Read a JoinQuant fills CSV and return ``JOINQUANT_FILL_COLUMNS``."""
+    raw = _clean_columns(pd.read_csv(path))
+    symbols = _symbol_frame(raw)
+    side = _text(_series_or_default(raw, ["side", "action", "direction"], ""))
+    requested = _signed_shares(
+        _series_or_default(raw, ["requested_shares", "target_shares", "amount", "order_amount"], 0),
+        side,
+    )
+    filled = _signed_shares(
+        _series_or_default(raw, ["filled_shares", "filled", "filled_amount", "deal_amount", "traded_shares"], 0),
+        side,
+    )
+    price = _numeric(_series_or_default(raw, ["fill_price", "price", "avg_cost", "avg_price"], np.nan), np.nan)
+    trade_value = _numeric(
+        _series_or_default(raw, ["trade_value", "value", "filled_value", "turnover"], np.nan),
+        np.nan,
+    )
+    trade_value = trade_value.fillna((filled * price).abs()).fillna(0.0)
+
+    out = pd.DataFrame({
+        "date": _date_series(raw),
+        "portfolio_name": _portfolio_series(raw, portfolio_name),
+        "symbol_id": symbols["symbol_id"],
+        "jq_symbol": symbols["jq_symbol"],
+        "order_id": _text(_series_or_default(raw, ["order_id", "id"], "")),
+        "side": side,
+        "requested_shares": requested.astype(float),
+        "filled_shares": filled.astype(float),
+        "fill_price": price.astype(float),
+        "trade_value": trade_value.astype(float),
+        "trade_cost": _numeric(_series_or_default(raw, ["trade_cost", "cost", "commission", "fee"], 0.0), 0.0),
+        "blocked": _numeric(_series_or_default(raw, ["blocked", "is_blocked"], 0), 0).astype("int64"),
+        "raw_status": _text(_series_or_default(raw, ["raw_status", "status", "order_status"], "")),
+    })
+    return out[JOINQUANT_FILL_COLUMNS]
+
+
+def normalize_positions_csv(path: str | Path, portfolio_name: str) -> pd.DataFrame:
+    """Read a JoinQuant positions CSV and return ``JOINQUANT_POSITION_COLUMNS``."""
+    raw = _clean_columns(pd.read_csv(path))
+    symbols = _symbol_frame(raw)
+    out = pd.DataFrame({
+        "date": _date_series(raw),
+        "portfolio_name": _portfolio_series(raw, portfolio_name),
+        "symbol_id": symbols["symbol_id"],
+        "jq_symbol": symbols["jq_symbol"],
+        "position_shares": _numeric(
+            _series_or_default(raw, ["position_shares", "shares", "amount", "quantity", "total_amount"], 0),
+            0,
+        ),
+        "position_value": _numeric(
+            _series_or_default(raw, ["position_value", "market_value", "value"], 0.0),
+            0.0,
+        ),
+        "cash": _numeric(_series_or_default(raw, ["cash", "available_cash"], np.nan), np.nan),
+        "total_value": _numeric(
+            _series_or_default(raw, ["total_value", "portfolio_value", "total_asset"], np.nan),
+            np.nan,
+        ),
+    })
+    return out[JOINQUANT_POSITION_COLUMNS]
+
+
+def normalize_pnl_csv(path: str | Path, portfolio_name: str) -> pd.DataFrame:
+    """Read a JoinQuant daily PnL CSV and return ``JOINQUANT_PNL_COLUMNS``."""
+    raw = _clean_columns(pd.read_csv(path))
+    total_value = _numeric(
+        _series_or_default(raw, ["total_value", "portfolio_value", "total_asset"], np.nan),
+        np.nan,
+    )
+    pnl = _numeric(_series_or_default(raw, ["pnl", "daily_pnl", "profit", "returns_value"], np.nan), np.nan)
+    if pnl.isna().all() and total_value.notna().any():
+        pnl = total_value.diff().fillna(0.0)
+
+    out = pd.DataFrame({
+        "date": _date_series(raw),
+        "portfolio_name": _portfolio_series(raw, portfolio_name),
+        "gross_exposure": _numeric(
+            _series_or_default(raw, ["gross_exposure", "gross", "positions_value", "market_value"], np.nan),
+            np.nan,
+        ),
+        "net_exposure": _numeric(
+            _series_or_default(raw, ["net_exposure", "net"], np.nan),
+            np.nan,
+        ),
+        "cash": _numeric(_series_or_default(raw, ["cash", "available_cash"], np.nan), np.nan),
+        "total_value": total_value,
+        "pnl": pnl.fillna(0.0),
+        "cost": _numeric(_series_or_default(raw, ["cost", "trade_cost", "commission", "fee"], 0.0), 0.0),
+        "turnover": _numeric(_series_or_default(raw, ["turnover"], 0.0), 0.0),
+    })
+    return out[JOINQUANT_PNL_COLUMNS]
+
+
+def ingest_joinquant_outputs(
+    *,
+    portfolio_name: str,
+    fills_csv: str | Path,
+    positions_csv: str | Path,
+    pnl_csv: str | Path,
+    out_dir: str | Path = "plugins_output/joinquant/ingested",
+) -> dict[str, Path]:
+    """Normalize JoinQuant CSV exports and write parquet outputs."""
+    out_root = Path(out_dir) / portfolio_name
+    out_root.mkdir(parents=True, exist_ok=True)
+
+    fills = normalize_fills_csv(fills_csv, portfolio_name)
+    positions = normalize_positions_csv(positions_csv, portfolio_name)
+    pnl = normalize_pnl_csv(pnl_csv, portfolio_name)
+
+    paths = {
+        "fills": out_root / "fills.pq",
+        "positions": out_root / "positions.pq",
+        "pnl": out_root / "pnl.pq",
+    }
+    fills.to_parquet(paths["fills"], index=False)
+    positions.to_parquet(paths["positions"], index=False)
+    pnl.to_parquet(paths["pnl"], index=False)
+    return paths
+

plugins/joinquant/reconcile.py

@@ -0,0 +1,607 @@
+"""Reconcile internal simulator output against normalized JoinQuant output."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+from plugins.joinquant.schema import (
+    JOINQUANT_FILL_COLUMNS,
+    JOINQUANT_PNL_COLUMNS,
+    JOINQUANT_POSITION_COLUMNS,
+    JOINQUANT_TARGET_COLUMNS,
+    RECONCILE_COLUMNS,
+)
+from plugins.joinquant.symbols import to_joinquant_symbol
+
+
+def _date_text(value: object) -> str:
+    if pd.isna(value):
+        return ""
+    return pd.Timestamp(value).strftime("%Y-%m-%d")
+
+
+def _read_parquet(path: str | Path | None) -> pd.DataFrame:
+    if path is None:
+        return pd.DataFrame()
+    return pd.read_parquet(path)
+
+
+def _normalize_common(df: pd.DataFrame, portfolio_name: str) -> pd.DataFrame:
+    out = df.copy()
+    if "date" in out.columns:
+        out["date"] = out["date"].map(_date_text)
+    else:
+        out["date"] = ""
+    if "portfolio_name" not in out.columns:
+        out["portfolio_name"] = portfolio_name
+    out["portfolio_name"] = out["portfolio_name"].fillna(portfolio_name).astype(str)
+    if "symbol_id" in out.columns:
+        out["symbol_id"] = out["symbol_id"].fillna("").astype(str)
+    if "jq_symbol" not in out.columns and "symbol_id" in out.columns:
+        out["jq_symbol"] = out["symbol_id"].map(
+            lambda s: to_joinquant_symbol(s) if s else ""
+        )
+    elif "jq_symbol" in out.columns:
+        out["jq_symbol"] = out["jq_symbol"].fillna("").astype(str)
+    return out
+
+
+def _numeric(df: pd.DataFrame, column: str, default: float = 0.0) -> pd.Series:
+    if column not in df.columns:
+        return pd.Series([default] * len(df), index=df.index, dtype=float)
+    return pd.to_numeric(df[column], errors="coerce").replace([np.inf, -np.inf], np.nan)
+
+
+def _weighted_price(group: pd.DataFrame, price_col: str, shares_col: str) -> float:
+    prices = pd.to_numeric(group[price_col], errors="coerce")
+    shares = pd.to_numeric(group[shares_col], errors="coerce").abs()
+    valid = prices.notna() & shares.notna() & (shares > 0)
+    if not valid.any():
+        return np.nan
+    return float(np.average(prices[valid], weights=shares[valid]))
+
+
+def _load_targets(targets_dir: str | Path, portfolio_name: str) -> pd.DataFrame:
+    root = Path(targets_dir)
+    if not root.exists():
+        return pd.DataFrame(columns=JOINQUANT_TARGET_COLUMNS)
+
+    files_by_stem: dict[str, Path] = {}
+    for path in sorted(root.glob("*.csv")):
+        files_by_stem[path.stem] = path
+    for path in sorted(root.glob("*.parquet")):
+        files_by_stem[path.stem] = path
+
+    frames: list[pd.DataFrame] = []
+    for path in files_by_stem.values():
+        if path.suffix == ".parquet":
+            frame = pd.read_parquet(path)
+        else:
+            frame = pd.read_csv(path)
+        frames.append(frame)
+
+    if not frames:
+        return pd.DataFrame(columns=JOINQUANT_TARGET_COLUMNS)
+    targets = pd.concat(frames, ignore_index=True)
+    targets = _normalize_common(targets, portfolio_name)
+    targets = targets[targets["portfolio_name"].astype(str) == portfolio_name]
+    if "target_shares" not in targets.columns:
+        targets["target_shares"] = 0
+    return targets.reindex(columns=JOINQUANT_TARGET_COLUMNS)
+
+
+def _aggregate_targets(targets: pd.DataFrame, portfolio_name: str) -> pd.DataFrame:
+    if targets.empty:
+        return pd.DataFrame(columns=["date", "portfolio_name", "symbol_id", "jq_symbol", "target_shares"])
+    targets = _normalize_common(targets, portfolio_name)
+    targets["target_shares"] = _numeric(targets, "target_shares", 0.0)
+    grouped = (
+        targets.groupby(["date", "portfolio_name", "symbol_id"], as_index=False)
+        .agg(jq_symbol=("jq_symbol", "last"), target_shares=("target_shares", "last"))
+    )
+    return grouped
+
+
+def _aggregate_our_fills(fills: pd.DataFrame, portfolio_name: str) -> pd.DataFrame:
+    if fills.empty:
+        return pd.DataFrame(columns=[
+            "date", "portfolio_name", "symbol_id", "our_filled_shares",
+            "our_position_shares", "our_cost", "our_trade_price", "our_blocked",
+            "our_target_shares",
+        ])
+    fills = _normalize_common(fills, portfolio_name)
+    fills["traded_shares"] = _numeric(fills, "traded_shares", 0.0)
+    fills["realized_shares"] = _numeric(fills, "realized_shares", np.nan)
+    fills["trade_cost"] = _numeric(fills, "trade_cost", 0.0).fillna(0.0)
+    fills["target_shares"] = _numeric(fills, "target_shares", np.nan)
+    fills["blocked"] = _numeric(fills, "blocked", 0.0).fillna(0.0)
+    price_col = next(
+        (col for col in ["trade_price", "fill_price", "execution_price", "price"] if col in fills.columns),
+        None,
+    )
+
+    rows: list[dict[str, object]] = []
+    for key, group in fills.groupby(["date", "portfolio_name", "symbol_id"], sort=False):
+        row = {
+            "date": key[0],
+            "portfolio_name": key[1],
+            "symbol_id": key[2],
+            "our_filled_shares": float(group["traded_shares"].sum()),
+            "our_position_shares": float(group["realized_shares"].dropna().iloc[-1])
+            if group["realized_shares"].notna().any() else np.nan,
+            "our_cost": float(group["trade_cost"].sum()),
+            "our_trade_price": _weighted_price(group, price_col, "traded_shares")
+            if price_col else np.nan,
+            "our_blocked": int(group["blocked"].max()),
+            "our_target_shares": float(group["target_shares"].dropna().iloc[-1])
+            if group["target_shares"].notna().any() else np.nan,
+        }
+        rows.append(row)
+    return pd.DataFrame(rows)
+
+
+def _aggregate_our_positions(positions: pd.DataFrame, portfolio_name: str) -> pd.DataFrame:
+    if positions.empty:
+        return pd.DataFrame(columns=[
+            "date", "portfolio_name", "symbol_id", "jq_symbol",
+            "our_position_fallback", "our_position_price",
+        ])
+    positions = _normalize_common(positions, portfolio_name)
+    positions["position_shares"] = _numeric(positions, "position_shares", np.nan)
+    positions["price"] = _numeric(positions, "price", np.nan)
+    return (
+        positions.groupby(["date", "portfolio_name", "symbol_id"], as_index=False)
+        .agg(
+            jq_symbol=("jq_symbol", "last"),
+            our_position_fallback=("position_shares", "last"),
+            our_position_price=("price", "last"),
+        )
+    )
+
+
+def _aggregate_jq_fills(fills: pd.DataFrame, portfolio_name: str) -> pd.DataFrame:
+    if fills.empty:
+        return pd.DataFrame(columns=[
+            "date", "portfolio_name", "symbol_id", "jq_filled_shares",
+            "jq_trade_price", "jq_cost", "jq_blocked", "jq_requested_shares",
+            "raw_status",
+        ])
+    fills = _normalize_common(fills, portfolio_name)
+    for col in JOINQUANT_FILL_COLUMNS:
+        if col not in fills.columns:
+            fills[col] = np.nan
+    fills["filled_shares"] = _numeric(fills, "filled_shares", 0.0).fillna(0.0)
+    fills["requested_shares"] = _numeric(fills, "requested_shares", np.nan)
+    fills["fill_price"] = _numeric(fills, "fill_price", np.nan)
+    fills["trade_cost"] = _numeric(fills, "trade_cost", 0.0).fillna(0.0)
+    fills["blocked"] = _numeric(fills, "blocked", 0.0).fillna(0.0)
+    fills["raw_status"] = fills["raw_status"].fillna("").astype(str)
+
+    rows: list[dict[str, object]] = []
+    for key, group in fills.groupby(["date", "portfolio_name", "symbol_id"], sort=False):
+        row = {
+            "date": key[0],
+            "portfolio_name": key[1],
+            "symbol_id": key[2],
+            "jq_filled_shares": float(group["filled_shares"].sum()),
+            "jq_trade_price": _weighted_price(group, "fill_price", "filled_shares"),
+            "jq_cost": float(group["trade_cost"].sum()),
+            "jq_blocked": int(group["blocked"].max()),
+            "jq_requested_shares": float(group["requested_shares"].dropna().iloc[-1])
+            if group["requested_shares"].notna().any() else np.nan,
+            "raw_status": ";".join([s for s in group["raw_status"].astype(str) if s]),
+        }
+        rows.append(row)
+    return pd.DataFrame(rows)
+
+
+def _aggregate_jq_positions(positions: pd.DataFrame, portfolio_name: str) -> pd.DataFrame:
+    if positions.empty:
+        return pd.DataFrame(columns=[
+            "date", "portfolio_name", "symbol_id", "jq_symbol", "jq_position_shares",
+        ])
+    positions = _normalize_common(positions, portfolio_name)
+    for col in JOINQUANT_POSITION_COLUMNS:
+        if col not in positions.columns:
+            positions[col] = np.nan
+    positions["position_shares"] = _numeric(positions, "position_shares", np.nan)
+    return (
+        positions.groupby(["date", "portfolio_name", "symbol_id"], as_index=False)
+        .agg(jq_symbol=("jq_symbol", "last"), jq_position_shares=("position_shares", "last"))
+    )
+
+
+def _portfolio_frame(df: pd.DataFrame, portfolio_name: str, prefix: str) -> pd.DataFrame:
+    if df.empty:
+        return pd.DataFrame(columns=["date", "portfolio_name"])
+    out = _normalize_common(df, portfolio_name)
+    for col in JOINQUANT_PNL_COLUMNS:
+        if col not in out.columns:
+            out[col] = np.nan
+    keep = ["date", "portfolio_name", "gross_exposure", "net_exposure", "cash", "total_value", "pnl", "cost", "turnover"]
+    out = out[keep].copy()
+    for col in keep[2:]:
+        out[col] = pd.to_numeric(out[col], errors="coerce")
+    out = out.groupby(["date", "portfolio_name"], as_index=False).last()
+    return out.rename(columns={col: f"{prefix}_{col}" for col in keep[2:]})
+
+
+def _infer_booksize(targets: pd.DataFrame, our_pnl: pd.DataFrame, jq_pnl: pd.DataFrame) -> float:
+    candidates: list[float] = []
+    if "target_value" in targets.columns:
+        gross = (
+            pd.to_numeric(targets["target_value"], errors="coerce")
+            .abs()
+            .replace([np.inf, -np.inf], np.nan)
+            .dropna()
+            .sum()
+        )
+        if gross > 0:
+            candidates.append(float(gross))
+    for df in (our_pnl, jq_pnl):
+        if "gross_exposure" in df.columns:
+            val = pd.to_numeric(df["gross_exposure"], errors="coerce").max()
+            if pd.notna(val) and val > 0:
+                candidates.append(float(val))
+    return max(candidates) if candidates else 1.0
+
+
+def _status_reason(raw_status: object) -> str | None:
+    text = str(raw_status or "").lower()
+    if "suspend" in text or "halt" in text:
+        return "SUSPENSION"
+    if "limit_up" in text or "limit up" in text or "up_limit" in text:
+        return "LIMIT_UP_BLOCK"
+    if "limit_down" in text or "limit down" in text or "down_limit" in text:
+        return "LIMIT_DOWN_BLOCK"
+    if "volume" in text or "liquid" in text:
+        return "VOLUME_OR_LIQUIDITY"
+    if "cash" in text or "margin" in text:
+        return "CASH_CONSTRAINT"
+    return None
+
+
+def _classify_symbol_row(
+    row: pd.Series,
+    *,
+    share_tol: float,
+    price_rel_tol: float,
+    value_tol: float,
+    pnl_tol: float,
+) -> str:
+    target = row.get("target_shares", np.nan)
+    filled_diff = abs(row.get("filled_share_diff", 0.0))
+    position_diff = abs(row.get("position_share_diff", 0.0))
+    our_present = bool(row.get("_our_present", False))
+    jq_present = bool(row.get("_jq_present", False))
+
+    if pd.notna(target) and target < 0 and (filled_diff > share_tol or position_diff > share_tol or not jq_present):
+        return "SHORT_NOT_SUPPORTED"
+    if not jq_present and (our_present or pd.notna(target)):
+        return "MISSING_IN_JOINQUANT"
+    if not our_present and jq_present:
+        return "MISSING_IN_OUR_SYSTEM"
+
+    status_reason = _status_reason(row.get("raw_status", ""))
+    if (filled_diff > share_tol or position_diff > share_tol) and status_reason:
+        return status_reason
+    if filled_diff > share_tol or position_diff > share_tol:
+        return "UNKNOWN"
+
+    our_price = row.get("our_trade_price", np.nan)
+    jq_price = row.get("jq_trade_price", np.nan)
+    if pd.notna(our_price) and pd.notna(jq_price):
+        denom = max(abs(float(our_price)), abs(float(jq_price)), 1.0)
+        if abs(float(our_price) - float(jq_price)) > price_rel_tol * denom:
+            return "PRICE_MISMATCH"
+
+    if abs(row.get("cost_diff", 0.0)) > value_tol:
+        return "COST_MODEL"
+    if abs(row.get("pnl_diff", 0.0)) > pnl_tol:
+        return "UNKNOWN"
+    return "MATCH"
+
+
+def _classify_portfolio_row(row: pd.Series, value_tol: float, pnl_tol: float) -> str:
+    our_present = bool(row.get("_our_present", False))
+    jq_present = bool(row.get("_jq_present", False))
+    if not jq_present and our_present:
+        return "MISSING_IN_JOINQUANT"
+    if not our_present and jq_present:
+        return "MISSING_IN_OUR_SYSTEM"
+    if abs(row.get("cost_diff", 0.0)) > value_tol:
+        return "COST_MODEL"
+    if abs(row.get("pnl_diff", 0.0)) > pnl_tol:
+        return "UNKNOWN"
+    return "MATCH"
+
+
+def _build_symbol_reconcile(
+    *,
+    portfolio_name: str,
+    targets: pd.DataFrame,
+    our_fills: pd.DataFrame,
+    our_positions: pd.DataFrame,
+    our_pnl: pd.DataFrame,
+    jq_fills: pd.DataFrame,
+    jq_positions: pd.DataFrame,
+    jq_pnl: pd.DataFrame,
+    share_tol: float,
+    price_rel_tol: float,
+    value_tol: float,
+    pnl_tol: float,
+) -> pd.DataFrame:
+    target_agg = _aggregate_targets(targets, portfolio_name)
+    our_fill_agg = _aggregate_our_fills(our_fills, portfolio_name)
+    our_pos_agg = _aggregate_our_positions(our_positions, portfolio_name)
+    jq_fill_agg = _aggregate_jq_fills(jq_fills, portfolio_name)
+    jq_pos_agg = _aggregate_jq_positions(jq_positions, portfolio_name)
+
+    key_cols = ["date", "portfolio_name", "symbol_id"]
+    keys = []
+    for frame in [target_agg, our_fill_agg, our_pos_agg, jq_fill_agg, jq_pos_agg]:
+        if not frame.empty:
+            keys.append(frame[key_cols])
+    if not keys:
+        return pd.DataFrame(columns=RECONCILE_COLUMNS)
+
+    base = pd.concat(keys, ignore_index=True).drop_duplicates()
+    result = base.merge(target_agg, on=key_cols, how="left")
+    result = result.merge(our_fill_agg, on=key_cols, how="left")
+    result = result.merge(our_pos_agg, on=key_cols, how="left", suffixes=("", "_ourpos"))
+    result = result.merge(jq_fill_agg, on=key_cols, how="left")
+    result = result.merge(jq_pos_agg, on=key_cols, how="left", suffixes=("", "_jqpos"))
+
+    jq_symbol_cols = [col for col in result.columns if col.startswith("jq_symbol")]
+    jq_symbol_values = result[jq_symbol_cols].copy() if jq_symbol_cols else pd.DataFrame(index=result.index)
+    result["jq_symbol"] = ""
+    for col in jq_symbol_values.columns:
+        values = jq_symbol_values[col].fillna("").astype(str)
+        result["jq_symbol"] = result["jq_symbol"].mask(
+            result["jq_symbol"].eq("") & values.ne(""),
+            values,
+        )
+    result["jq_symbol"] = result.apply(
+        lambda row: row["jq_symbol"] or to_joinquant_symbol(row["symbol_id"]),
+        axis=1,
+    )
+
+    result["_our_present"] = (
+        result[["our_filled_shares", "our_position_shares", "our_position_fallback"]]
+        .notna()
+        .any(axis=1)
+    )
+    result["_jq_present"] = (
+        result[["jq_filled_shares", "jq_position_shares"]].notna().any(axis=1)
+    )
+
+    target_shares = pd.to_numeric(result["target_shares"], errors="coerce")
+    our_target = pd.to_numeric(result["our_target_shares"], errors="coerce")
+    jq_target = pd.to_numeric(result["jq_requested_shares"], errors="coerce")
+    result["target_shares"] = target_shares.where(target_shares.notna(), our_target)
+    result["target_shares"] = result["target_shares"].where(
+        result["target_shares"].notna(),
+        jq_target,
+    )
+
+    our_position = pd.to_numeric(result["our_position_shares"], errors="coerce")
+    our_position_fallback = pd.to_numeric(result["our_position_fallback"], errors="coerce")
+    result["our_position_shares"] = our_position.where(
+        our_position.notna(),
+        our_position_fallback,
+    )
+
+    for col in [
+        "target_shares", "our_filled_shares", "jq_filled_shares",
+        "our_position_shares", "jq_position_shares", "our_cost", "jq_cost",
+    ]:
+        result[col] = pd.to_numeric(result[col], errors="coerce").fillna(0.0)
+
+    result["filled_share_diff"] = result["our_filled_shares"] - result["jq_filled_shares"]
+    result["position_share_diff"] = result["our_position_shares"] - result["jq_position_shares"]
+    result["trade_price_diff"] = np.where(
+        result["our_trade_price"].notna() & result["jq_trade_price"].notna(),
+        result["our_trade_price"] - result["jq_trade_price"],
+        np.nan,
+    )
+    result["cost_diff"] = result["our_cost"] - result["jq_cost"]
+
+    our_daily = _portfolio_frame(our_pnl, portfolio_name, "our")
+    jq_daily = _portfolio_frame(jq_pnl, portfolio_name, "jq")
+    pnl_daily = our_daily.merge(jq_daily, on=["date", "portfolio_name"], how="outer")
+    if not pnl_daily.empty:
+        pnl_daily["our_pnl"] = pd.to_numeric(pnl_daily.get("our_pnl"), errors="coerce").fillna(0.0)
+        pnl_daily["jq_pnl"] = pd.to_numeric(pnl_daily.get("jq_pnl"), errors="coerce").fillna(0.0)
+        pnl_daily["pnl_diff"] = pnl_daily["our_pnl"] - pnl_daily["jq_pnl"]
+        result = result.merge(
+            pnl_daily[["date", "portfolio_name", "our_pnl", "jq_pnl", "pnl_diff"]],
+            on=["date", "portfolio_name"],
+            how="left",
+        )
+    else:
+        result["our_pnl"] = 0.0
+        result["jq_pnl"] = 0.0
+        result["pnl_diff"] = 0.0
+    for col in ["our_pnl", "jq_pnl", "pnl_diff"]:
+        result[col] = pd.to_numeric(result[col], errors="coerce").fillna(0.0)
+
+    result["raw_status"] = result.get("raw_status", "").fillna("")
+    result["diff_reason"] = result.apply(
+        _classify_symbol_row,
+        axis=1,
+        share_tol=share_tol,
+        price_rel_tol=price_rel_tol,
+        value_tol=value_tol,
+        pnl_tol=pnl_tol,
+    )
+
+    return result[RECONCILE_COLUMNS].sort_values(
+        ["date", "portfolio_name", "symbol_id"]
+    ).reset_index(drop=True)
+
+
+def _build_portfolio_summary(
+    *,
+    portfolio_name: str,
+    our_pnl: pd.DataFrame,
+    jq_pnl: pd.DataFrame,
+    value_tol: float,
+    pnl_tol: float,
+) -> pd.DataFrame:
+    our = _portfolio_frame(our_pnl, portfolio_name, "our")
+    jq = _portfolio_frame(jq_pnl, portfolio_name, "jq")
+    if our.empty and jq.empty:
+        return pd.DataFrame(columns=[
+            "date", "portfolio_name", "diff_reason",
+            "our_pnl", "jq_pnl", "pnl_diff",
+        ])
+    summary = our.merge(jq, on=["date", "portfolio_name"], how="outer")
+    summary["_our_present"] = summary.filter(regex=r"^our_").notna().any(axis=1)
+    summary["_jq_present"] = summary.filter(regex=r"^jq_").notna().any(axis=1)
+    metrics = ["gross_exposure", "net_exposure", "cash", "total_value", "pnl", "cost", "turnover"]
+    for metric in metrics:
+        our_col = f"our_{metric}"
+        jq_col = f"jq_{metric}"
+        if our_col not in summary.columns:
+            summary[our_col] = np.nan
+        if jq_col not in summary.columns:
+            summary[jq_col] = np.nan
+        summary[f"{metric}_diff"] = (
+            pd.to_numeric(summary[our_col], errors="coerce").fillna(0.0)
+            - pd.to_numeric(summary[jq_col], errors="coerce").fillna(0.0)
+        )
+    summary["our_cumulative_pnl"] = (
+        pd.to_numeric(summary["our_pnl"], errors="coerce").fillna(0.0).cumsum()
+    )
+    summary["jq_cumulative_pnl"] = (
+        pd.to_numeric(summary["jq_pnl"], errors="coerce").fillna(0.0).cumsum()
+    )
+    summary["cumulative_pnl_diff"] = summary["our_cumulative_pnl"] - summary["jq_cumulative_pnl"]
+    summary["diff_reason"] = summary.apply(
+        _classify_portfolio_row,
+        axis=1,
+        value_tol=value_tol,
+        pnl_tol=pnl_tol,
+    )
+    summary = summary.drop(columns=["_our_present", "_jq_present"])
+    return summary.sort_values(["date", "portfolio_name"]).reset_index(drop=True)
+
+
+def _write_summary_md(
+    path: Path,
+    *,
+    portfolio_name: str,
+    symbol_report: pd.DataFrame,
+    portfolio_summary: pd.DataFrame,
+) -> None:
+    symbol_counts = (
+        symbol_report["diff_reason"].value_counts().sort_index()
+        if not symbol_report.empty else pd.Series(dtype=int)
+    )
+    portfolio_counts = (
+        portfolio_summary["diff_reason"].value_counts().sort_index()
+        if not portfolio_summary.empty else pd.Series(dtype=int)
+    )
+    lines = [
+        "# JoinQuant Reconciliation Summary",
+        "",
+        f"Portfolio: `{portfolio_name}`",
+        "",
+        "## Per-symbol Difference Counts",
+        "",
+    ]
+    if symbol_counts.empty:
+        lines.append("No per-symbol rows were produced.")
+    else:
+        for reason, count in symbol_counts.items():
+            lines.append(f"- {reason}: {int(count)}")
+    lines.extend(["", "## Daily Portfolio Difference Counts", ""])
+    if portfolio_counts.empty:
+        lines.append("No daily portfolio rows were produced.")
+    else:
+        for reason, count in portfolio_counts.items():
+            lines.append(f"- {reason}: {int(count)}")
+
+    if not portfolio_summary.empty:
+        lines.extend(["", "## Daily Portfolio Preview", ""])
+        preview_cols = [
+            "date", "diff_reason", "our_pnl", "jq_pnl", "pnl_diff",
+            "our_cost", "jq_cost", "cost_diff",
+        ]
+        preview_cols = [col for col in preview_cols if col in portfolio_summary.columns]
+        lines.append(",".join(preview_cols))
+        for row in portfolio_summary[preview_cols].head(20).itertuples(index=False):
+            lines.append(",".join(str(value) for value in row))
+
+    path.write_text("\n".join(lines) + "\n", encoding="utf-8")
+
+
+def reconcile_joinquant(
+    *,
+    portfolio_name: str,
+    targets_dir: str | Path,
+    our_fills_path: str | Path,
+    our_positions_path: str | Path,
+    our_pnl_path: str | Path,
+    jq_fills_path: str | Path,
+    jq_positions_path: str | Path,
+    jq_pnl_path: str | Path,
+    out_dir: str | Path = "plugins_output/joinquant/reconcile",
+    share_tolerance: float = 0.0,
+    price_rel_tolerance: float = 1e-4,
+    pnl_tolerance: float = 1.0,
+    booksize: float | None = None,
+) -> dict[str, Path]:
+    """Reconcile JoinQuant output against internal simulator output."""
+    targets = _load_targets(targets_dir, portfolio_name)
+    our_fills = _read_parquet(our_fills_path)
+    our_positions = _read_parquet(our_positions_path)
+    our_pnl = _read_parquet(our_pnl_path)
+    jq_fills = _read_parquet(jq_fills_path)
+    jq_positions = _read_parquet(jq_positions_path)
+    jq_pnl = _read_parquet(jq_pnl_path)
+
+    inferred_booksize = booksize or _infer_booksize(targets, our_pnl, jq_pnl)
+    value_tol = max(1.0, 1e-6 * float(inferred_booksize))
+
+    symbol_report = _build_symbol_reconcile(
+        portfolio_name=portfolio_name,
+        targets=targets,
+        our_fills=our_fills,
+        our_positions=our_positions,
+        our_pnl=our_pnl,
+        jq_fills=jq_fills,
+        jq_positions=jq_positions,
+        jq_pnl=jq_pnl,
+        share_tol=share_tolerance,
+        price_rel_tol=price_rel_tolerance,
+        value_tol=value_tol,
+        pnl_tol=pnl_tolerance,
+    )
+    portfolio_summary = _build_portfolio_summary(
+        portfolio_name=portfolio_name,
+        our_pnl=our_pnl,
+        jq_pnl=jq_pnl,
+        value_tol=value_tol,
+        pnl_tol=pnl_tolerance,
+    )
+
+    root = Path(out_dir) / portfolio_name
+    root.mkdir(parents=True, exist_ok=True)
+    paths = {
+        "daily_reconcile": root / "daily_reconcile.pq",
+        "summary_csv": root / "summary.csv",
+        "summary_md": root / "summary.md",
+    }
+    symbol_report.to_parquet(paths["daily_reconcile"], index=False)
+    portfolio_summary.to_csv(paths["summary_csv"], index=False)
+    _write_summary_md(
+        paths["summary_md"],
+        portfolio_name=portfolio_name,
+        symbol_report=symbol_report,
+        portfolio_summary=portfolio_summary,
+    )
+    return paths

plugins/joinquant/schema.py

@@ -0,0 +1,99 @@
+"""Column contracts for the JoinQuant comparison plugin."""
+
+from typing import Final
+
+
+JOINQUANT_TARGET_COLUMNS: Final[list[str]] = [
+    "date",
+    "portfolio_name",
+    "symbol_id",
+    "jq_symbol",
+    "target_shares",
+    "target_value",
+    "target_weight",
+    "export_mode",
+    "snapshot_id",
+]
+
+JOINQUANT_FILL_COLUMNS: Final[list[str]] = [
+    "date",
+    "portfolio_name",
+    "symbol_id",
+    "jq_symbol",
+    "order_id",
+    "side",
+    "requested_shares",
+    "filled_shares",
+    "fill_price",
+    "trade_value",
+    "trade_cost",
+    "blocked",
+    "raw_status",
+]
+
+JOINQUANT_POSITION_COLUMNS: Final[list[str]] = [
+    "date",
+    "portfolio_name",
+    "symbol_id",
+    "jq_symbol",
+    "position_shares",
+    "position_value",
+    "cash",
+    "total_value",
+]
+
+JOINQUANT_PNL_COLUMNS: Final[list[str]] = [
+    "date",
+    "portfolio_name",
+    "gross_exposure",
+    "net_exposure",
+    "cash",
+    "total_value",
+    "pnl",
+    "cost",
+    "turnover",
+]
+
+RECONCILE_COLUMNS: Final[list[str]] = [
+    "date",
+    "portfolio_name",
+    "symbol_id",
+    "jq_symbol",
+    "target_shares",
+    "our_filled_shares",
+    "jq_filled_shares",
+    "filled_share_diff",
+    "our_position_shares",
+    "jq_position_shares",
+    "position_share_diff",
+    "our_trade_price",
+    "jq_trade_price",
+    "trade_price_diff",
+    "our_cost",
+    "jq_cost",
+    "cost_diff",
+    "our_pnl",
+    "jq_pnl",
+    "pnl_diff",
+    "diff_reason",
+]
+
+DIFF_REASONS: Final[list[str]] = [
+    "MATCH",
+    "SYMBOL_MAPPING",
+    "PRICE_MISMATCH",
+    "LOT_ROUNDING",
+    "SUSPENSION",
+    "LIMIT_UP_BLOCK",
+    "LIMIT_DOWN_BLOCK",
+    "VOLUME_OR_LIQUIDITY",
+    "COST_MODEL",
+    "CASH_CONSTRAINT",
+    "SHORT_NOT_SUPPORTED",
+    "CORPORATE_ACTION",
+    "JOINQUANT_INTERNAL_ROUNDING",
+    "MISSING_IN_OUR_SYSTEM",
+    "MISSING_IN_JOINQUANT",
+    "UNKNOWN",
+]
+

plugins/joinquant/smoke.py

@@ -0,0 +1,191 @@
+"""End-to-end local smoke preparation for JoinQuant comparison."""
+
+from __future__ import annotations
+
+import json
+from datetime import datetime, timezone
+from pathlib import Path
+
+import pandas as pd
+
+from pipeline.common.schema import POSITION_COLUMNS
+from pipeline.data.downloader import download_universe
+from pipeline.portfolio.constraints import get_constraint
+from pipeline.portfolio.simulator import ReferenceSimulator
+from plugins.joinquant.export_targets import export_targets
+from plugins.joinquant.wrapper_strategy import write_wrapper_strategy
+
+
+def build_fixed_share_positions(
+    data: pd.DataFrame,
+    *,
+    trade_symbol: str,
+    portfolio_name: str,
+    shares: int,
+    booksize: float,
+    max_signal_dates: int | None = None,
+) -> pd.DataFrame:
+    """Create a deterministic long-only fixed-share position book.
+
+    The final available data date is excluded because the internal simulator
+    executes each signal date at the next available open.
+    """
+    data = data.copy()
+    data["date"] = pd.to_datetime(data["date"]).dt.normalize()
+    symbol_data = (
+        data[data["symbol_id"].astype(str) == trade_symbol]
+        .sort_values("date")
+        .reset_index(drop=True)
+    )
+    if len(symbol_data) < 2:
+        raise ValueError(f"Need at least two daily bars for {trade_symbol}")
+
+    signal_data = symbol_data.iloc[:-1].copy()
+    if max_signal_dates is not None and max_signal_dates > 0:
+        signal_data = signal_data.tail(max_signal_dates)
+    if signal_data.empty:
+        raise ValueError("No signal dates available after excluding final data date")
+
+    rows: list[dict[str, object]] = []
+    for row in signal_data.itertuples(index=False):
+        price = float(row.close)
+        target_value = float(shares * price)
+        rows.append({
+            "symbol_id": trade_symbol,
+            "date": pd.Timestamp(row.date),
+            "portfolio_name": portfolio_name,
+            "target_weight": target_value / float(booksize),
+            "target_value": target_value,
+            "target_shares": float(shares),
+            "position_shares": int(shares),
+            "position_value": target_value,
+            "price": price,
+        })
+    return pd.DataFrame(rows, columns=POSITION_COLUMNS)
+
+
+def prepare_smoke_test(
+    *,
+    out_dir: str | Path,
+    universe: str = "sh600000,sz000001,sh600519,sz002594,sz300750",
+    trade_symbol: str = "sh600000",
+    start_date: str = "2024-01-02",
+    end_date: str = "2024-01-12",
+    portfolio_name: str = "jq_smoke_one_stock_long",
+    shares: int = 1000,
+    booksize: float = 1_000_000.0,
+    max_signal_dates: int = 3,
+    cost_bps: float = 5.0,
+    slippage_bps: float = 5.0,
+    volume_frac: float = 0.02,
+    force: bool = False,
+) -> dict[str, object]:
+    """Run the local side of a tiny real-data JoinQuant smoke test."""
+    root = Path(out_dir)
+    root.mkdir(parents=True, exist_ok=True)
+
+    stats = download_universe(
+        universe=universe,
+        start_date=start_date,
+        end_date=end_date,
+        output_dir=root / "daily_bars",
+        max_symbols=0,
+        chunk_size=100,
+        adjust="qfq",
+    )
+    data_path = Path(stats["dataset_path"])
+    data = pd.read_parquet(data_path)
+
+    positions = build_fixed_share_positions(
+        data,
+        trade_symbol=trade_symbol,
+        portfolio_name=portfolio_name,
+        shares=shares,
+        booksize=booksize,
+        max_signal_dates=max_signal_dates,
+    )
+    portfolio_dir = root / "portfolio"
+    portfolio_dir.mkdir(parents=True, exist_ok=True)
+    positions_path = portfolio_dir / f"{portfolio_name}.pq"
+    positions.to_parquet(positions_path, index=False)
+
+    constraints = [
+        get_constraint("suspension"),
+        get_constraint("price_limit"),
+        get_constraint("volume_cap", max_frac=volume_frac),
+    ]
+    sim = ReferenceSimulator(
+        constraints=constraints,
+        cost_bps=cost_bps,
+        slippage_bps=slippage_bps,
+    )
+    fills, pnl = sim.run(positions, data)
+    execution_dir = root / "execution"
+    fills_dir = execution_dir / "fills"
+    pnl_dir = execution_dir / "pnl"
+    fills_dir.mkdir(parents=True, exist_ok=True)
+    pnl_dir.mkdir(parents=True, exist_ok=True)
+    fills_path = fills_dir / f"{portfolio_name}.pq"
+    pnl_path = pnl_dir / f"{portfolio_name}.pq"
+    fills.to_parquet(fills_path, index=False)
+    pnl.to_parquet(pnl_path, index=False)
+
+    target_root = root / "plugins_output" / "joinquant" / "targets_aligned"
+    snapshots = export_targets(
+        positions_path=positions_path,
+        portfolio_name=portfolio_name,
+        mode="target_shares",
+        out_dir=target_root,
+        execution_calendar_path=data_path,
+        force=force,
+    )
+    wrapper_path = root / "plugins_output" / "joinquant" / f"wrapper_strategy_{portfolio_name}.py"
+    write_wrapper_strategy(
+        portfolio_name=portfolio_name,
+        mode="target_shares",
+        out_path=wrapper_path,
+    )
+
+    export_dir = root / "joinquant_exports"
+    export_dir.mkdir(parents=True, exist_ok=True)
+
+    manifest = {
+        "created_at": datetime.now(timezone.utc).isoformat(),
+        "portfolio_name": portfolio_name,
+        "universe": universe,
+        "trade_symbol": trade_symbol,
+        "start_date": start_date,
+        "end_date": end_date,
+        "shares": shares,
+        "booksize": booksize,
+        "data_path": str(data_path),
+        "positions_path": str(positions_path),
+        "fills_path": str(fills_path),
+        "pnl_path": str(pnl_path),
+        "targets_dir": str(target_root / portfolio_name),
+        "wrapper_path": str(wrapper_path),
+        "joinquant_export_dir": str(export_dir),
+        "expected_joinquant_csvs": {
+            "fills": str(export_dir / "jq_fills.csv"),
+            "positions": str(export_dir / "jq_positions.csv"),
+            "pnl": str(export_dir / "jq_pnl.csv"),
+        },
+        "target_snapshots": snapshots,
+        "local_summary": {
+            "n_data_rows": int(len(data)),
+            "n_position_rows": int(len(positions)),
+            "n_fill_rows": int(len(fills)),
+            "n_pnl_rows": int(len(pnl)),
+            "total_pnl": float(pnl["pnl"].sum()) if len(pnl) else 0.0,
+            "total_cost": float(pnl["cost"].sum()) if len(pnl) else 0.0,
+            "blocked_trades": int(fills["blocked"].sum()) if len(fills) else 0,
+        },
+    }
+    manifest_path = root / "joinquant_smoke_manifest.json"
+    manifest["manifest_path"] = str(manifest_path)
+    manifest_path.write_text(
+        json.dumps(manifest, indent=2, ensure_ascii=False) + "\n",
+        encoding="utf-8",
+    )
+    return manifest
+

plugins/joinquant/symbols.py

@@ -0,0 +1,94 @@
+"""Symbol conversion between internal A-share ids and JoinQuant ids."""
+
+from __future__ import annotations
+
+import re
+
+
+_INTERNAL_RE = re.compile(r"^(?P<prefix>sh|sz)(?P<code>\d{6})$", re.IGNORECASE)
+_JOINQUANT_RE = re.compile(
+    r"^(?P<code>\d{6})\.(?P<exchange>XSHG|XSHE)$", re.IGNORECASE
+)
+_BARE_RE = re.compile(r"^\d{6}$")
+
+
+def _validate_internal(prefix: str, code: str) -> None:
+    if prefix == "sh" and code.startswith("6"):
+        return
+    if prefix == "sz" and code.startswith(("0", "3")):
+        return
+    raise ValueError(f"Unsupported A-share symbol: {prefix}{code}")
+
+
+def _validate_joinquant(code: str, exchange: str) -> None:
+    if exchange == "XSHG" and code.startswith("6"):
+        return
+    if exchange == "XSHE" and code.startswith(("0", "3")):
+        return
+    raise ValueError(f"Unsupported JoinQuant A-share symbol: {code}.{exchange}")
+
+
+def to_joinquant_symbol(symbol_id: str) -> str:
+    """Convert an internal symbol like ``sh600000`` to ``600000.XSHG``.
+
+    Args:
+        symbol_id: Internal A-share id with ``sh`` or ``sz`` prefix.
+
+    Returns:
+        JoinQuant security id.
+
+    Raises:
+        ValueError: If the symbol is malformed or outside supported A-share
+            Shanghai/Shenzhen equity prefixes.
+    """
+    text = str(symbol_id).strip().lower()
+    match = _INTERNAL_RE.match(text)
+    if not match:
+        raise ValueError(f"Invalid internal symbol: {symbol_id!r}")
+
+    prefix = match.group("prefix").lower()
+    code = match.group("code")
+    _validate_internal(prefix, code)
+    exchange = "XSHG" if prefix == "sh" else "XSHE"
+    return f"{code}.{exchange}"
+
+
+def from_joinquant_symbol(jq_symbol: str) -> str:
+    """Convert a JoinQuant symbol like ``600000.XSHG`` to ``sh600000``."""
+    text = str(jq_symbol).strip().upper()
+    match = _JOINQUANT_RE.match(text)
+    if not match:
+        raise ValueError(f"Invalid JoinQuant symbol: {jq_symbol!r}")
+
+    code = match.group("code")
+    exchange = match.group("exchange").upper()
+    _validate_joinquant(code, exchange)
+    prefix = "sh" if exchange == "XSHG" else "sz"
+    return f"{prefix}{code}"
+
+
+def normalize_symbol_pair(value: object) -> tuple[str, str]:
+    """Return ``(symbol_id, jq_symbol)`` from any supported symbol spelling."""
+    text = str(value).strip()
+    if not text or text.lower() == "nan":
+        raise ValueError("Missing symbol")
+
+    if _INTERNAL_RE.match(text):
+        symbol_id = text.lower()
+        return symbol_id, to_joinquant_symbol(symbol_id)
+
+    if _JOINQUANT_RE.match(text):
+        jq_symbol = text.upper()
+        return from_joinquant_symbol(jq_symbol), jq_symbol
+
+    if _BARE_RE.match(text):
+        if text.startswith("6"):
+            symbol_id = f"sh{text}"
+        elif text.startswith(("0", "3")):
+            symbol_id = f"sz{text}"
+        else:
+            raise ValueError(f"Unsupported bare A-share code: {text}")
+        return symbol_id, to_joinquant_symbol(symbol_id)
+
+    raise ValueError(f"Unsupported symbol: {value!r}")
+

plugins/joinquant/wrapper_strategy.py

@@ -0,0 +1,271 @@
+"""Generate a standalone JoinQuant wrapper strategy.
+
+The module also defines default JoinQuant strategy hooks by executing the same
+template with ``run1`` / ``target_shares`` defaults. That means this file can be
+copied directly into JoinQuant for a quick smoke test, while the CLI can still
+write a configured standalone file for a real run.
+"""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from string import Template
+from typing import Literal
+
+
+WrapperMode = Literal["target_shares", "target_value"]
+
+
+_WRAPPER_TEMPLATE = Template(
+    r'''# Standalone JoinQuant target wrapper generated by chinese-equity-quant.
+# Copy this file and the exported daily CSV target files into JoinQuant.
+#
+# The file loader is isolated in _read_target_file(). The default implementation
+# uses JoinQuant's read_file API for uploaded files. If your JoinQuant runtime
+# allows HTTP or another storage backend, replace only that function.
+
+import csv
+import io
+import json
+
+
+PORTFOLIO_NAME = "${portfolio_name}"
+TARGET_MODE = "${mode}"
+ALLOW_SHORT = ${allow_short}
+TARGET_FILE_PREFIX = ""  # Optional uploaded-file prefix, for example "run1/"
+_EMBEDDED_TARGETS = ${embedded_targets}
+
+
+def initialize(context):
+    set_benchmark("000300.XSHG")
+    set_option("use_real_price", True)
+    g.portfolio_name = PORTFOLIO_NAME
+    g.target_mode = TARGET_MODE
+    g.targets_by_date = {}
+    run_daily(load_targets, time="before_open")
+    run_daily(rebalance_at_open, time="open")
+    run_daily(record_after_close, time="after_close")
+
+
+def _today_text(context):
+    return context.current_dt.strftime("%Y-%m-%d")
+
+
+def _today_file_name(context):
+    return context.current_dt.strftime("%Y%m%d") + ".csv"
+
+
+def _read_target_file(file_name):
+${embedded_target_read}
+    data = read_file(TARGET_FILE_PREFIX + file_name)
+    if isinstance(data, bytes):
+        data = data.decode("utf-8")
+    return data
+
+
+def _load_target_rows(context):
+    file_name = _today_file_name(context)
+    text = _read_target_file(file_name)
+    rows = list(csv.DictReader(io.StringIO(text)))
+    clean_rows = []
+    for row in rows:
+        if row.get("portfolio_name") and row["portfolio_name"] != PORTFOLIO_NAME:
+            continue
+        jq_symbol = row.get("jq_symbol") or row.get("security") or row.get("symbol")
+        if not jq_symbol:
+            log.warn("Skipping target row with no jq_symbol: %s" % row)
+            continue
+
+        if TARGET_MODE == "target_shares":
+            target = int(float(row.get("target_shares") or 0))
+        elif TARGET_MODE == "target_value":
+            target = float(row.get("target_value") or 0.0)
+        else:
+            raise ValueError("Unsupported TARGET_MODE: %s" % TARGET_MODE)
+
+        if not ALLOW_SHORT and target < 0:
+            log.warn(
+                "SHORT_NOT_SUPPORTED clipping %s target from %s to 0" %
+                (jq_symbol, target)
+            )
+            target = 0
+
+        clean_rows.append({"jq_symbol": jq_symbol, "target": target, "raw": row})
+    return clean_rows
+
+
+def load_targets(context):
+    date_text = _today_text(context)
+    try:
+        rows = _load_target_rows(context)
+    except Exception as exc:
+        log.error("Failed to load JoinQuant target file for %s: %s" % (date_text, exc))
+        rows = []
+    g.targets_by_date[date_text] = rows
+    log.info("JOINQUANT_TARGET_LOAD|%s" % json.dumps({
+        "date": date_text,
+        "portfolio_name": PORTFOLIO_NAME,
+        "target_mode": TARGET_MODE,
+        "n_targets": len(rows),
+    }, sort_keys=True))
+
+
+def rebalance_at_open(context):
+    date_text = _today_text(context)
+    rows = g.targets_by_date.get(date_text, [])
+    target_symbols = set()
+    for row in rows:
+        security = row["jq_symbol"]
+        target_symbols.add(security)
+        if TARGET_MODE == "target_shares":
+            order_target(security, int(row["target"]))
+        else:
+            order_target_value(security, float(row["target"]))
+        log.info("JOINQUANT_ORDER_SUBMIT|%s" % json.dumps({
+            "date": date_text,
+            "portfolio_name": PORTFOLIO_NAME,
+            "jq_symbol": security,
+            "target_mode": TARGET_MODE,
+            "target": row["target"],
+        }, sort_keys=True))
+
+    for security in list(context.portfolio.positions.keys()):
+        if security not in target_symbols:
+            order_target(security, 0)
+            log.info("JOINQUANT_ORDER_CLOSE|%s" % json.dumps({
+                "date": date_text,
+                "portfolio_name": PORTFOLIO_NAME,
+                "jq_symbol": security,
+            }, sort_keys=True))
+
+
+def _position_records(context):
+    records = []
+    cash = float(context.portfolio.available_cash)
+    total_value = float(context.portfolio.total_value)
+    for security, position in context.portfolio.positions.items():
+        records.append({
+            "date": _today_text(context),
+            "portfolio_name": PORTFOLIO_NAME,
+            "jq_symbol": security,
+            "position_shares": int(position.total_amount),
+            "position_value": float(position.value),
+            "cash": cash,
+            "total_value": total_value,
+        })
+    return records
+
+
+def _trade_records(context):
+    records = []
+    try:
+        trades = get_trades()
+    except Exception:
+        trades = {}
+    for trade_id, trade in trades.items():
+        amount = int(getattr(trade, "amount", 0))
+        price = float(getattr(trade, "price", 0.0))
+        security = getattr(trade, "security", "")
+        side = "buy" if amount >= 0 else "sell"
+        records.append({
+            "date": _today_text(context),
+            "portfolio_name": PORTFOLIO_NAME,
+            "jq_symbol": security,
+            "order_id": str(getattr(trade, "order_id", trade_id)),
+            "side": side,
+            "filled_shares": amount,
+            "fill_price": price,
+            "trade_value": abs(amount * price),
+            "trade_cost": float(getattr(trade, "commission", 0.0)),
+            "raw_status": "filled",
+        })
+    return records
+
+
+def record_after_close(context):
+    date_text = _today_text(context)
+    for record in _trade_records(context):
+        log.info("JOINQUANT_FILL|%s" % json.dumps(record, sort_keys=True))
+    for record in _position_records(context):
+        log.info("JOINQUANT_POSITION|%s" % json.dumps(record, sort_keys=True))
+    log.info("JOINQUANT_PNL|%s" % json.dumps({
+        "date": date_text,
+        "portfolio_name": PORTFOLIO_NAME,
+        "cash": float(context.portfolio.available_cash),
+        "total_value": float(context.portfolio.total_value),
+    }, sort_keys=True))
+'''
+)
+
+
+# Make this module itself usable as a JoinQuant strategy with defaults.
+exec(_WRAPPER_TEMPLATE.substitute(
+    portfolio_name="run1",
+    mode="target_shares",
+    allow_short="False",
+    embedded_targets="{}",
+    embedded_target_read="",
+))
+
+
+def render_wrapper_strategy(
+    *,
+    portfolio_name: str,
+    mode: WrapperMode = "target_shares",
+    allow_short: bool = False,
+    embedded_targets: dict[str, str] | None = None,
+) -> str:
+    """Render the standalone JoinQuant wrapper strategy source."""
+    if mode not in {"target_shares", "target_value"}:
+        raise ValueError("mode must be 'target_shares' or 'target_value'")
+    embedded_targets = embedded_targets or {}
+    embedded_target_read = ""
+    if embedded_targets:
+        embedded_target_read = (
+            "    if file_name in _EMBEDDED_TARGETS:\n"
+            "        return _EMBEDDED_TARGETS[file_name]\n"
+        )
+    return _WRAPPER_TEMPLATE.substitute(
+        portfolio_name=portfolio_name,
+        mode=mode,
+        allow_short="True" if allow_short else "False",
+        embedded_targets=json.dumps(embedded_targets, ensure_ascii=False, indent=4),
+        embedded_target_read=embedded_target_read,
+    )
+
+
+def _load_embedded_targets(targets_dir: str | Path | None) -> dict[str, str]:
+    if targets_dir is None:
+        return {}
+    root = Path(targets_dir)
+    targets = {
+        path.name: path.read_text(encoding="utf-8")
+        for path in sorted(root.glob("*.csv"))
+    }
+    if not targets:
+        raise ValueError(f"No CSV target files found under {root}")
+    return targets
+
+
+def write_wrapper_strategy(
+    *,
+    portfolio_name: str,
+    mode: WrapperMode = "target_shares",
+    out_path: str | Path,
+    allow_short: bool = False,
+    embedded_targets_dir: str | Path | None = None,
+) -> Path:
+    """Write a configured standalone JoinQuant wrapper strategy."""
+    path = Path(out_path)
+    path.parent.mkdir(parents=True, exist_ok=True)
+    path.write_text(
+        render_wrapper_strategy(
+            portfolio_name=portfolio_name,
+            mode=mode,
+            allow_short=allow_short,
+            embedded_targets=_load_embedded_targets(embedded_targets_dir),
+        ),
+        encoding="utf-8",
+    )
+    return path

pyproject.toml

@@ -4,7 +4,6 @@ version = "0.1.0"
 description = "A modular Chinese A-share quant research framework (daily frequency)."
 requires-python = ">=3.10"
 dependencies = [
-    "backtrader>=1.9.76.123",
     "akshare>=1.14.0",
     "baostock>=0.8.8",
     "pandas>=2.0.0",
@@ -13,10 +12,42 @@ dependencies = [
     "pyarrow>=14.0.0",
 ]
 
+[project.optional-dependencies]
+backtrader = [
+    "backtrader>=1.9.76.123",
+]
+joinquant-browser = [
+    "playwright>=1.61.0",
+]
+
 [dependency-groups]
 dev = [
+    "coverage>=7.14.1",
     "pytest>=7.0.0",
 ]
 
 [tool.uv]
 package = false
+
+[tool.pytest.ini_options]
+markers = [
+    "network: tests that call live external data providers and are skipped unless explicitly enabled",
+]
+
+[tool.coverage.run]
+branch = true
+relative_files = true
+source = [
+    ".",
+]
+
+[tool.coverage.report]
+fail_under = 95
+show_missing = true
+skip_covered = false
+omit = [
+    "tests/*",
+    "docs/*",
+    "scripts/*",
+    ".venv/*",
+]

scripts/run_coverage.py

@@ -0,0 +1,40 @@
+"""Run pytest under coverage in this environment.
+
+The plain ``coverage run -m pytest`` path reloads NumPy under the current VS Code
+Python startup environment, which breaks pandas/numpy reductions. Import NumPy
+before starting coverage so the measured test run uses one stable NumPy module.
+"""
+
+from __future__ import annotations
+
+import sys
+
+import numpy  # noqa: F401
+import coverage
+import pytest
+
+
+def main(argv: list[str]) -> int:
+    pytest_args = argv or ["tests/", "-v"]
+
+    cov = coverage.Coverage(config_file=True)
+    cov.erase()
+    cov.start()
+    test_status = pytest.main(pytest_args)
+    cov.stop()
+    cov.save()
+
+    if test_status != 0:
+        return int(test_status)
+
+    total = cov.report()
+    print(f"\nCoverage total: {total:.2f}%")
+    fail_under = float(cov.config.fail_under)
+    if total < fail_under:
+        print(f"Coverage failure: total {total:.2f}% is below {fail_under:.2f}%")
+        return 2
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main(sys.argv[1:]))

scripts/run_reversal_rank_e2e.sh

@@ -0,0 +1,62 @@
+#!/usr/bin/env bash
+# End-to-end run of the outlier-robust reversal_rank alpha on the full
+# all-universe dataset and on a per-date liquid subset. Records per-phase
+# wall-clock time to reports/reversal_rank_timings.json.
+set -euo pipefail
+cd "$(dirname "$0")/.."
+
+DATA=data/daily_bars/all
+BOOK=10000000
+TIMINGS=reports/reversal_rank_timings.json
+mkdir -p reports
+echo "{" > "$TIMINGS"
+
+run() {  # run <json_key> <cmd...>
+  local key="$1"; shift
+  local t0 t1
+  t0=$(date +%s.%N)
+  "$@"
+  t1=$(date +%s.%N)
+  printf '  "%s": %.2f,\n' "$key" "$(echo "$t1 - $t0" | bc)" >> "$TIMINGS"
+  echo ">>> $key took $(echo "$t1 - $t0" | bc)s"
+}
+
+# ---- full all-universe, robust rank weighting ----
+run full_alpha_compute uv run python cli.py alpha compute --data-path "$DATA" \
+  --alpha-name reversal_rank_all --alpha-type reversal_rank --lookback 5 --output-dir alphas
+run full_alpha_eval uv run python cli.py alpha eval \
+  --alpha-path alphas/reversal_rank_all.pq --data-path "$DATA"
+run full_combo uv run python cli.py combo combine \
+  --alpha-paths alphas/reversal_rank_all.pq --combo-name reversal_rank_all_combo \
+  --method equal_weight --output-dir combos
+run full_portfolio_build uv run python cli.py portfolio build \
+  --weights-path combos/reversal_rank_all_combo.pq --data-path "$DATA" \
+  --booksize "$BOOK" --portfolio-name reversal_rank_all_10m --output-dir portfolio
+run full_portfolio_eval uv run python cli.py portfolio eval \
+  --positions-path portfolio/reversal_rank_all_10m.pq --data-path "$DATA"
+run full_portfolio_simulate uv run python cli.py portfolio simulate \
+  --positions-path portfolio/reversal_rank_all_10m.pq --data-path "$DATA" \
+  --constraint suspension --constraint price_limit --constraint volume_cap \
+  --cost-bps 5 --slippage-bps 5 --output-dir portfolio
+
+# ---- liquid subset (per-date investable universe), robust rank weighting ----
+run liq_alpha_compute uv run python cli.py alpha compute --data-path "$DATA" \
+  --alpha-name reversal_rank_liq --alpha-type reversal_rank --lookback 5 \
+  --liquid-universe --universe-top-n 1000 --output-dir alphas
+run liq_alpha_eval uv run python cli.py alpha eval \
+  --alpha-path alphas/reversal_rank_liq.pq --data-path "$DATA"
+run liq_combo uv run python cli.py combo combine \
+  --alpha-paths alphas/reversal_rank_liq.pq --combo-name reversal_rank_liq_combo \
+  --method equal_weight --output-dir combos
+run liq_portfolio_build uv run python cli.py portfolio build \
+  --weights-path combos/reversal_rank_liq_combo.pq --data-path "$DATA" \
+  --booksize "$BOOK" --portfolio-name reversal_rank_liq_10m --output-dir portfolio
+run liq_portfolio_eval uv run python cli.py portfolio eval \
+  --positions-path portfolio/reversal_rank_liq_10m.pq --data-path "$DATA"
+run liq_portfolio_simulate uv run python cli.py portfolio simulate \
+  --positions-path portfolio/reversal_rank_liq_10m.pq --data-path "$DATA" \
+  --constraint suspension --constraint price_limit --constraint volume_cap \
+  --cost-bps 5 --slippage-bps 5 --output-dir portfolio
+
+printf '  "_done": true\n}\n' >> "$TIMINGS"
+echo "Wrote $TIMINGS"

tests/helpers.py

@@ -0,0 +1,261 @@
+"""Shared deterministic test data for offline workflow tests."""
+
+from __future__ import annotations
+
+import numpy as np
+import pandas as pd
+
+from pipeline.common.schema import (
+    ALPHA_COLUMNS,
+    COMBO_COLUMNS,
+    DATA_COLUMNS,
+    MINUTE_BAR_COLUMNS,
+)
+
+
+GENERATED_SYMBOLS: tuple[str, ...] = (
+    "sh600000",
+    "sz000001",
+    "sh600519",
+    "sz300750",
+)
+
+GENERATED_SYMBOL_NAMES: dict[str, str] = {
+    "sh600000": "PF Bank",
+    "sz000001": "Ping An Bank",
+    "sh600519": "Kweichow Moutai",
+    "sz300750": "CATL",
+}
+
+
+def generated_sessions(n_sessions: int = 12) -> pd.DatetimeIndex:
+    """Return a fixed business-day calendar used by generated fixtures."""
+    return pd.bdate_range("2024-01-02", periods=n_sessions)
+
+
+def make_generated_daily_bars(
+    n_sessions: int = 12,
+    include_missing: bool = True,
+) -> pd.DataFrame:
+    """Build daily bars with explicit edge cases and no randomness.
+
+    The panel covers four A-share symbols and includes a suspended row, an ST
+    flag, a zero-volume row, a missing symbol-date, and limit-style open/close
+    moves. Values are deterministic so tests can assert exact identities.
+    """
+    dates = generated_sessions(n_sessions)
+    base_close = {
+        "sh600000": 10.00,
+        "sz000001": 15.00,
+        "sh600519": 1200.00,
+        "sz300750": 180.00,
+    }
+    returns = {
+        "sh600000": [0.000, 0.012, -0.006, 0.018, 0.100, -0.014, 0.006, 0.000, 0.008, -0.011, 0.004, 0.009],
+        "sz000001": [0.000, -0.008, 0.011, -0.004, 0.006, 0.000, -0.012, 0.009, 0.005, -0.007, 0.010, -0.003],
+        "sh600519": [0.000, 0.006, 0.004, -0.010, 0.012, -0.006, 0.005, 0.003, -0.009, 0.007, -0.004, 0.006],
+        "sz300750": [0.000, -0.010, 0.014, 0.006, -0.008, 0.011, -0.004, 0.009, -0.200, 0.012, -0.006, 0.008],
+    }
+    base_volume = {
+        "sh600000": 1_200_000.0,
+        "sz000001": 900_000.0,
+        "sh600519": 80_000.0,
+        "sz300750": 240_000.0,
+    }
+
+    rows: list[dict[str, object]] = []
+    for sym in GENERATED_SYMBOLS:
+        closes = [base_close[sym]]
+        pattern = returns[sym]
+        for step in range(1, n_sessions):
+            ret = pattern[step % len(pattern)]
+            closes.append(closes[-1] * (1.0 + ret))
+        closes_arr = np.asarray(closes, dtype=float)
+        precloses = np.concatenate([[closes_arr[0]], closes_arr[:-1]])
+
+        for i, date in enumerate(dates):
+            preclose = float(precloses[i])
+            close = float(closes_arr[i])
+            open_price = preclose * (1.0 + 0.25 * (close / preclose - 1.0))
+            high = max(open_price, close) * 1.01
+            low = min(open_price, close) * 0.99
+            volume = base_volume[sym] + 10_000.0 * i
+            tradestatus = 1
+            is_st = 0
+
+            if sym == "sh600000" and i == 4:
+                open_price = preclose * 1.10
+                close = open_price
+                high = open_price
+                low = open_price
+            if sym == "sh600000" and i == 7:
+                volume = 0.0
+            if sym == "sz000001" and i == 5:
+                open_price = preclose
+                close = preclose
+                high = preclose
+                low = preclose
+                volume = 0.0
+                tradestatus = 0
+            if sym == "sh600519" and i == 7:
+                is_st = 1
+            if sym == "sz300750" and i == 8:
+                open_price = preclose * 0.80
+                close = open_price
+                high = open_price
+                low = open_price
+
+            amount = volume * ((open_price + close) / 2.0)
+            vwap = amount / volume if volume > 0 else np.nan
+            pct_chg = (close / preclose - 1.0) * 100.0 if preclose else 0.0
+            rows.append({
+                "symbol_id": sym,
+                "symbol_name": GENERATED_SYMBOL_NAMES[sym],
+                "date": date,
+                "open": open_price,
+                "high": high,
+                "low": low,
+                "close": close,
+                "preclose": preclose,
+                "volume": volume,
+                "amount": amount,
+                "vwap": vwap,
+                "turn": volume / 1_000_000.0,
+                "pctChg": pct_chg,
+                "tradestatus": tradestatus,
+                "isST": is_st,
+                "peTTM": 8.0 + i,
+                "pbMRQ": 1.0 + 0.05 * i,
+                "psTTM": 2.0 + 0.03 * i,
+                "pcfNcfTTM": 5.0 + 0.1 * i,
+            })
+
+    result = pd.DataFrame(rows)
+    if include_missing and n_sessions > 6:
+        missing_mask = (
+            (result["symbol_id"] == "sz300750")
+            & (result["date"] == dates[6])
+        )
+        result = result.loc[~missing_mask].copy()
+
+    result = result[DATA_COLUMNS]
+    return result.sort_values(["date", "symbol_id"]).reset_index(drop=True)
+
+
+def make_generated_minute_bars(
+    daily: pd.DataFrame | None = None,
+) -> pd.DataFrame:
+    """Expand generated daily bars into a tiny deterministic intraday panel."""
+    daily = make_generated_daily_bars() if daily is None else daily.copy()
+    rows: list[dict[str, object]] = []
+    bar_times = ("09:35:00", "10:30:00", "14:55:00")
+
+    for daily_row in daily.sort_values(["date", "symbol_id"]).itertuples(index=False):
+        if int(getattr(daily_row, "tradestatus", 1)) == 0:
+            continue
+
+        volume = float(daily_row.volume)
+        volume_slices = [0.25 * volume, 0.35 * volume, 0.40 * volume]
+        prices = np.linspace(float(daily_row.open), float(daily_row.close), len(bar_times))
+        for j, time_text in enumerate(bar_times):
+            dt = pd.Timestamp(daily_row.date) + pd.Timedelta(time_text)
+            open_price = prices[j - 1] if j else float(daily_row.open)
+            close_price = float(prices[j])
+            high = max(open_price, close_price) * 1.002
+            low = min(open_price, close_price) * 0.998
+            minute_volume = float(volume_slices[j])
+            amount = minute_volume * ((open_price + close_price) / 2.0)
+            rows.append({
+                "symbol_id": daily_row.symbol_id,
+                "symbol_name": daily_row.symbol_name,
+                "datetime": dt,
+                "date": pd.Timestamp(daily_row.date).normalize(),
+                "time": time_text,
+                "frequency": "5m",
+                "open": open_price,
+                "high": high,
+                "low": low,
+                "close": close_price,
+                "volume": minute_volume,
+                "amount": amount,
+                "vwap": amount / minute_volume if minute_volume > 0 else np.nan,
+                "adjustflag": "3",
+            })
+
+    return pd.DataFrame(rows, columns=MINUTE_BAR_COLUMNS)
+
+
+def make_generated_derived_features(
+    daily: pd.DataFrame | None = None,
+) -> pd.DataFrame:
+    """Return numeric daily derived values, including NaN and infinity cells."""
+    daily = make_generated_daily_bars() if daily is None else daily.copy()
+    keys = (
+        daily[["symbol_id", "date"]]
+        .drop_duplicates()
+        .sort_values(["date", "symbol_id"])
+        .reset_index(drop=True)
+    )
+    date_rank = keys["date"].rank(method="dense").astype(float)
+    symbol_rank = keys["symbol_id"].map({
+        "sh600000": 1.0,
+        "sz000001": 2.0,
+        "sh600519": 3.0,
+        "sz300750": 4.0,
+    })
+    out = keys.copy()
+    out["toy_feature"] = symbol_rank + date_rank / 100.0
+    out["finite_feature"] = symbol_rank * date_rank
+    out["nan_feature"] = out["toy_feature"]
+    out["inf_feature"] = out["toy_feature"]
+    if len(out) >= 2:
+        out.loc[out.index[0], "nan_feature"] = np.nan
+        out.loc[out.index[1], "inf_feature"] = np.inf
+    return out
+
+
+def make_generated_alpha_weights(
+    alpha_name: str = "alpha_a",
+    *,
+    scale: float = 1.0,
+    offset: float = 0.0,
+    zero_date_index: int | None = None,
+    n_sessions: int = 10,
+) -> pd.DataFrame:
+    """Create a deterministic long alpha grid with optional zero-gross date."""
+    dates = generated_sessions(n_sessions)
+    even = np.array([1.20, -0.80, 0.40, -0.80], dtype=float)
+    odd = np.array([-0.60, 1.10, -0.90, 0.40], dtype=float)
+    rows: list[dict[str, object]] = []
+    for i, date in enumerate(dates):
+        vector = even.copy() if i % 2 == 0 else odd.copy()
+        vector = vector + offset
+        vector = vector - vector.mean()
+        if zero_date_index is not None and i == zero_date_index:
+            vector = np.zeros_like(vector)
+        for sym, weight in zip(GENERATED_SYMBOLS, scale * vector):
+            rows.append({
+                "symbol_id": sym,
+                "date": date,
+                "alpha_name": alpha_name,
+                "weight": float(weight),
+            })
+    result = pd.DataFrame(rows, columns=ALPHA_COLUMNS)
+    return result.sort_values(["symbol_id", "date"]).reset_index(drop=True)
+
+
+def make_generated_combo_weights(
+    combo_name: str = "combo",
+    *,
+    zero_date_index: int | None = 2,
+    n_sessions: int = 10,
+) -> pd.DataFrame:
+    """Create deterministic combo weights for portfolio construction tests."""
+    alpha = make_generated_alpha_weights(
+        "combo_source",
+        zero_date_index=zero_date_index,
+        n_sessions=n_sessions,
+    )
+    combo = alpha.rename(columns={"alpha_name": "combo_name"}).copy()
+    combo["combo_name"] = combo_name
+    return combo[COMBO_COLUMNS].sort_values(["symbol_id", "date"]).reset_index(drop=True)

tests/test_alpha.py

@@ -6,7 +6,11 @@ import pandas as pd
 import pytest
 
 from pipeline.alpha.base import BaseAlpha
-from pipeline.alpha.compute import compute_alpha, evaluate_alpha
+from pipeline.alpha.compute import (
+    compute_alpha,
+    evaluate_alpha,
+    investable_universe_mask,
+)
 from pipeline.alpha.registry import (
     available_alphas,
     get_alpha,
@@ -51,7 +55,7 @@ def test_reversal_sign_matches_negative_trailing_return():
     data = _make_data()
     alpha = compute_alpha(data, "rev5", "reversal", lookback=5)
     close = data.pivot_table(index="date", columns="symbol_id", values="close").sort_index()
-    raw = -close.pct_change(5)
+    raw = -close.pct_change(5, fill_method=None)
     last = raw.index[-1]
     expected_top = raw.loc[last].idxmax()
     got = alpha[alpha["date"] == last].set_index("symbol_id")["weight"].idxmax()
@@ -74,6 +78,83 @@ def test_evaluate_alpha_keys():
         assert key in metrics
 
 
+def test_evaluate_alpha_uses_next_open_to_next_open_returns():
+    dates = pd.date_range("2024-01-01", periods=5)
+    data = pd.concat([
+        pd.DataFrame({
+            "symbol_id": "sh600000",
+            "symbol_name": "sh600000",
+            "date": dates,
+            "open": [100.0, 100.0, 100.0, 100.0, 200.0],
+            "high": [100.0, 1000.0, 1000.0, 1000.0, 1000.0],
+            "low": [100.0, 1000.0, 1000.0, 1000.0, 1000.0],
+            "close": [100.0, 1000.0, 1000.0, 1000.0, 1000.0],
+            "volume": 1_000.0,
+            "amount": 1_000.0,
+        }),
+        pd.DataFrame({
+            "symbol_id": "sz000001",
+            "symbol_name": "sz000001",
+            "date": dates,
+            "open": [100.0, 100.0, 100.0, 200.0, 200.0],
+            "high": [100.0, 10.0, 10.0, 10.0, 10.0],
+            "low": [100.0, 10.0, 10.0, 10.0, 10.0],
+            "close": [100.0, 10.0, 10.0, 10.0, 10.0],
+            "volume": 1_000.0,
+            "amount": 1_000.0,
+        }),
+    ], ignore_index=True)
+    alpha = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000", "sz000001"],
+        "date": [dates[1], dates[1], dates[2], dates[2]],
+        "alpha_name": ["toy"] * 4,
+        "weight": [-1.0, 1.0, 1.0, -1.0],
+    })
+
+    metrics = evaluate_alpha(alpha, data)
+
+    assert metrics["n_dates"] == 2
+    assert np.isclose(metrics["cumulative_return"], 1.25)
+
+
+def test_evaluate_alpha_excludes_signal_without_forward_return():
+    dates = pd.date_range("2024-01-01", periods=3)
+    data = pd.concat([
+        pd.DataFrame({
+            "symbol_id": "sh600000",
+            "symbol_name": "sh600000",
+            "date": dates,
+            "open": [100.0, 100.0, 200.0],
+            "high": [100.0, 100.0, 200.0],
+            "low": [100.0, 100.0, 200.0],
+            "close": [100.0, 100.0, 200.0],
+            "volume": 1_000.0,
+            "amount": 1_000.0,
+        }),
+        pd.DataFrame({
+            "symbol_id": "sz000001",
+            "symbol_name": "sz000001",
+            "date": dates,
+            "open": [100.0, 100.0, 100.0],
+            "high": [100.0, 100.0, 100.0],
+            "low": [100.0, 100.0, 100.0],
+            "close": [100.0, 100.0, 100.0],
+            "volume": 1_000.0,
+            "amount": 1_000.0,
+        }),
+    ], ignore_index=True)
+    alpha = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000", "sz000001"],
+        "date": [dates[0], dates[0], dates[1], dates[1]],
+        "alpha_name": ["toy"] * 4,
+        "weight": [1.0, -1.0, -1.0, 1.0],
+    })
+
+    metrics = evaluate_alpha(alpha, data)
+
+    assert metrics["n_dates"] == 1
+
+
 def test_equal_weight_is_mean_of_alphas():
     data = _make_data()
     a = compute_alpha(data, "rev", "reversal", lookback=5)
@@ -95,10 +176,34 @@ def test_combine_alphas_schema(tmp_path):
     assert (combo["combo_name"] == "eq").all()
 
 
+def test_combine_single_alpha_is_identity(tmp_path):
+    data = _make_data()
+    a = compute_alpha(data, "rev", "reversal", lookback=5)
+    a_path = tmp_path / "a.pq"
+    a.to_parquet(a_path, index=False)
+
+    combo = combine_alphas([str(a_path)], "rev_combo", method="equal_weight")
+
+    expected = a[["symbol_id", "date", "weight"]].reset_index(drop=True)
+    got = combo[["symbol_id", "date", "weight"]].reset_index(drop=True)
+    pd.testing.assert_frame_equal(got, expected)
+    assert list(combo.columns) == COMBO_COLUMNS
+    assert (combo["combo_name"] == "rev_combo").all()
+
+
+def test_combine_alphas_rejects_unknown_method(tmp_path):
+    data = _make_data()
+    alpha_path = tmp_path / "alpha.pq"
+    compute_alpha(data, "rev", "reversal", lookback=5).to_parquet(alpha_path, index=False)
+
+    with pytest.raises(ValueError, match="Unknown combo method"):
+        combine_alphas([str(alpha_path)], "bad_combo", method="does_not_exist")
+
+
 # --- registry / factory -----------------------------------------------------
 
 def test_builtins_are_registered():
-    assert {"reversal", "reversal_vol", "momentum"} <= set(available_alphas())
+    assert {"reversal", "reversal_vol", "momentum", "reversal_rank"} <= set(available_alphas())
 
 
 def test_get_alpha_filters_unaccepted_params():
@@ -109,6 +214,22 @@ def test_get_alpha_filters_unaccepted_params():
     assert not hasattr(alpha, "vol_window")
 
 
+def test_get_alpha_forwards_kwargs_to_flexible_alpha():
+    @register_alpha
+    class _FlexibleAlpha(BaseAlpha):
+        name = "_flexible_alpha_kwargs"
+
+        def __init__(self, **kwargs):
+            self.kwargs = kwargs
+
+        def signal(self, close):
+            return close
+
+    alpha = get_alpha("_flexible_alpha_kwargs", decay=0.5, label="demo")
+
+    assert alpha.kwargs == {"decay": 0.5, "label": "demo"}
+
+
 def test_get_alpha_unknown_raises():
     with pytest.raises(KeyError):
         get_alpha("does_not_exist")
@@ -131,6 +252,31 @@ def test_register_rejects_non_basealpha():
         register_alpha(object)  # type: ignore[arg-type]
 
 
+def test_register_rejects_empty_alpha_name():
+    with pytest.raises(ValueError, match="non-empty"):
+        @register_alpha
+        class NoNameAlpha(BaseAlpha):
+            def signal(self, close):
+                return close
+
+
+def test_load_alpha_module_error_paths(tmp_path, monkeypatch):
+    missing_path = tmp_path / "missing_alpha.py"
+    with pytest.raises(FileNotFoundError):
+        load_alpha_module(str(missing_path))
+
+    bad_path = tmp_path / "bad_alpha.py"
+    bad_path.write_text("x = 1\n")
+    monkeypatch.setattr(
+        "pipeline.alpha.registry.importlib.util.spec_from_file_location",
+        lambda *args, **kwargs: None,
+    )
+    with pytest.raises(ImportError, match="Cannot load alpha module"):
+        load_alpha_module(str(bad_path))
+
+    load_alpha_module("math")
+
+
 # --- base class --------------------------------------------------------------
 
 def test_to_weights_are_per_date_zscore():
@@ -146,6 +292,20 @@ def test_to_weights_are_per_date_zscore():
     assert (weights.mean(axis=1).abs() < 1e-9).all()
 
 
+def test_base_alpha_default_signal_and_repr():
+    alpha = BaseAlpha()
+    alpha.example = 3
+
+    with pytest.raises(NotImplementedError, match="signal"):
+        alpha.signal(pd.DataFrame({"x": [1.0]}))
+    with pytest.raises(NotImplementedError, match="signal"):
+        alpha.signal_from_data(
+            pd.DataFrame({"symbol_id": ["sh600000"]}),
+            pd.DataFrame({"sh600000": [1.0]}),
+        )
+    assert repr(alpha) == "BaseAlpha(example=3)"
+
+
 # --- external plugin loading -------------------------------------------------
 
 def test_load_external_alpha_module(tmp_path):
@@ -163,7 +323,7 @@ def test_load_external_alpha_module(tmp_path):
                 self.span = span
 
             def signal(self, close: pd.DataFrame) -> pd.DataFrame:
-                return -close.pct_change(self.span)
+                return -close.pct_change(self.span, fill_method=None)
     '''))
 
     load_alpha_module(str(module_path))
@@ -178,3 +338,246 @@ def test_load_external_alpha_module(tmp_path):
     assert list(result.columns) == ALPHA_COLUMNS
     assert (result["alpha_name"] == "ext").all()
 
+
+# --- rank reversal + investable universe filter ------------------------------
+
+def _make_rich_data(n_days: int = 70, symbols=("sh600000", "sz000001", "sh600519", "sz300750")):
+    """Long-format data with the columns the universe filter needs."""
+    dates = pd.date_range("2024-01-01", periods=n_days)
+    rng = np.random.default_rng(1)
+    frames = []
+    for i, sym in enumerate(symbols):
+        close = 100.0 + i * 5 + np.cumsum(rng.standard_normal(n_days))
+        frames.append(pd.DataFrame({
+            "symbol_id": sym,
+            "symbol_name": sym,
+            "date": dates,
+            "open": close, "high": close, "low": close, "close": close,
+            "volume": 1_000.0,
+            "amount": (1_000.0 + i * 5_000.0) * close,  # higher i = more liquid
+            "isST": 0,
+            "tradestatus": 1,
+        }))
+    return pd.concat(frames, ignore_index=True)
+
+
+def test_reversal_rank_registered_and_bounded():
+    data = _make_data(n_days=30)
+    alpha = compute_alpha(data, "rr", "reversal_rank", lookback=5)
+    assert list(alpha.columns) == ALPHA_COLUMNS
+    # Rank-demeaned weights are per-date zero-mean and bounded by the
+    # cross-section size, never blowing up the way a z-score outlier can.
+    per_date_mean = alpha.groupby("date")["weight"].mean().abs()
+    assert (per_date_mean < 1e-9).all()
+    assert alpha["weight"].abs().max() <= len(data["symbol_id"].unique())
+
+
+def test_investable_universe_mask_excludes_st_and_suspended():
+    data = _make_rich_data()
+    # Flag one name ST throughout, and suspend another on the last date.
+    data.loc[data["symbol_id"] == "sh600000", "isST"] = 1
+    last = data["date"].max()
+    data.loc[(data["symbol_id"] == "sz000001") & (data["date"] == last), "tradestatus"] = 0
+
+    close = data.pivot_table(index="date", columns="symbol_id", values="close").sort_index()
+    mask = investable_universe_mask(data, close, top_n=10, min_history=5)
+
+    assert not mask["sh600000"].any()          # ST excluded on every date
+    assert not bool(mask.loc[last, "sz000001"])  # suspended on the last date
+    assert bool(mask.loc[last, "sh600519"])      # a normal name stays investable
+
+
+def test_compute_alpha_universe_filter_zeros_excluded_names():
+    data = _make_rich_data()
+    data.loc[data["symbol_id"] == "sh600000", "isST"] = 1
+
+    alpha = compute_alpha(
+        data, "rr_liq", "reversal_rank", lookback=5,
+        universe={"top_n": 10, "min_history": 5},
+    )
+    # The ST name is never held; an investable name is.
+    st_w = alpha.loc[alpha["symbol_id"] == "sh600000", "weight"]
+    assert (st_w.fillna(0.0) == 0.0).all()
+    assert alpha.loc[alpha["symbol_id"] == "sz300750", "weight"].abs().sum() > 0.0
+
+
+def test_universe_filter_does_not_corrupt_signal_history():
+    # Masking happens on the signal, not the price history, so weights on
+    # investable names match the unfiltered weights restricted to that set.
+    data = _make_rich_data()
+    universe = {"top_n": 2, "min_history": 5}  # keep only the 2 most liquid names
+
+    filtered = compute_alpha(data, "f", "reversal_rank", lookback=5, universe=universe)
+    held = set(filtered.loc[filtered["weight"] != 0.0, "symbol_id"].unique())
+    # The two most liquid names (highest amount) are sh600519, sz300750.
+    assert held == {"sh600519", "sz300750"}
+
+
+# --- feature-aware alpha integration ----------------------------------------
+
+def test_compute_alpha_without_feature_path_matches_empty_feature_paths():
+    data = _make_data()
+
+    base = compute_alpha(data, "rev5", "reversal", lookback=5)
+    with_empty_features = compute_alpha(
+        data,
+        "rev5",
+        "reversal",
+        lookback=5,
+        feature_paths=[],
+    )
+
+    pd.testing.assert_frame_equal(base, with_empty_features)
+
+
+def test_feature_aware_alpha_reads_joined_feature_column(tmp_path):
+    module_path = tmp_path / "feature_aware_alpha.py"
+    module_path.write_text(textwrap.dedent('''
+        import pandas as pd
+        from pipeline.alpha.base import BaseAlpha
+        from pipeline.alpha.registry import register_alpha
+
+        @register_alpha
+        class FeatureAwareAlpha(BaseAlpha):
+            name = "feature_aware_test_alpha"
+
+            def signal_from_data(
+                self,
+                data: pd.DataFrame,
+                close: pd.DataFrame,
+            ) -> pd.DataFrame:
+                signal = data.pivot_table(
+                    index="date",
+                    columns="symbol_id",
+                    values="toy_feature",
+                    aggfunc="first",
+                )
+                return signal.reindex(index=close.index, columns=close.columns)
+    '''))
+
+    data = _make_data()
+    feature = data[["symbol_id", "date"]].copy()
+    feature["toy_feature"] = feature["symbol_id"].map({
+        "sh600000": 1.0,
+        "sz000001": 2.0,
+        "sh600519": 3.0,
+    })
+    feature_path = tmp_path / "toy_feature.pq"
+    feature.to_parquet(feature_path, index=False)
+
+    load_alpha_module(str(module_path))
+    result = compute_alpha(
+        data,
+        "feature_run",
+        "feature_aware_test_alpha",
+        feature_paths=[str(feature_path)],
+    )
+
+    assert list(result.columns) == ALPHA_COLUMNS
+    assert (result["alpha_name"] == "feature_run").all()
+    last = result[result["date"] == result["date"].max()]
+    assert last.set_index("symbol_id")["weight"].idxmax() == "sh600519"
+
+
+def test_feature_paths_join_multiple_files_and_normalize_dates(tmp_path):
+    module_path = tmp_path / "multi_feature_alpha.py"
+    module_path.write_text(textwrap.dedent('''
+        import pandas as pd
+        from pipeline.alpha.base import BaseAlpha
+        from pipeline.alpha.registry import register_alpha
+
+        @register_alpha
+        class MultiFeatureAlpha(BaseAlpha):
+            name = "multi_feature_test_alpha"
+
+            def signal_from_data(
+                self,
+                data: pd.DataFrame,
+                close: pd.DataFrame,
+            ) -> pd.DataFrame:
+                data = data.copy()
+                data["combined_feature"] = data["toy_a"] + data["toy_b"]
+                signal = data.pivot_table(
+                    index="date",
+                    columns="symbol_id",
+                    values="combined_feature",
+                    aggfunc="first",
+                )
+                return signal.reindex(index=close.index, columns=close.columns)
+    '''))
+
+    data = _make_data(n_days=8)
+    symbol_score = {"sh600000": 1.0, "sz000001": 2.0, "sh600519": 3.0}
+
+    feature_a = data[["symbol_id", "date"]].copy()
+    feature_a["date"] = feature_a["date"] + pd.Timedelta(hours=15)
+    feature_a["toy_a"] = feature_a["symbol_id"].map(symbol_score)
+
+    feature_b = data[["symbol_id", "date"]].copy()
+    feature_b["date"] = feature_b["date"].dt.strftime("%Y-%m-%d 09:30:00")
+    feature_b["toy_b"] = feature_b["symbol_id"].map(symbol_score) * 10.0
+
+    feature_a_path = tmp_path / "toy_a.pq"
+    feature_b_path = tmp_path / "toy_b.pq"
+    feature_a.to_parquet(feature_a_path, index=False)
+    feature_b.to_parquet(feature_b_path, index=False)
+
+    load_alpha_module(str(module_path))
+    result = compute_alpha(
+        data,
+        "multi_feature_run",
+        "multi_feature_test_alpha",
+        feature_paths=[str(feature_a_path), str(feature_b_path)],
+    )
+
+    assert list(result.columns) == ALPHA_COLUMNS
+    assert (result["alpha_name"] == "multi_feature_run").all()
+    last = result[result["date"] == result["date"].max()]
+    assert last.set_index("symbol_id")["weight"].idxmax() == "sh600519"
+
+
+def test_compute_alpha_rejects_duplicate_feature_frame_columns():
+    data = _make_data()
+    duplicate_columns = pd.DataFrame(
+        [["sh600000", pd.Timestamp("2024-01-01"), 1.0, 2.0]],
+        columns=["symbol_id", "date", "toy_feature", "toy_feature"],
+    )
+
+    with pytest.raises(ValueError, match="duplicate columns"):
+        compute_alpha(
+            data,
+            "bad_features",
+            "reversal",
+            feature_frames=[duplicate_columns],
+        )
+
+
+def test_compute_alpha_rejects_feature_path_collision_with_daily_data(tmp_path):
+    data = _make_data()
+    close_collision = data[["symbol_id", "date"]].copy()
+    close_collision["close"] = 1.0
+    close_collision_path = tmp_path / "close_collision.pq"
+    close_collision.to_parquet(close_collision_path, index=False)
+
+    with pytest.raises(ValueError, match="conflict"):
+        compute_alpha(
+            data,
+            "close_collision",
+            "reversal",
+            feature_paths=[str(close_collision_path)],
+        )
+
+
+def test_evaluate_alpha_empty_when_signal_dates_not_on_market_calendar():
+    data = _make_data(n_days=3)
+    alpha = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [pd.Timestamp("2030-01-01")],
+        "alpha_name": ["future"],
+        "weight": [1.0],
+    })
+
+    metrics = evaluate_alpha(alpha, data)
+
+    assert metrics["n_dates"] == 0
+    assert metrics["cumulative_return"] == 0.0

tests/test_cli_workflow.py

@@ -0,0 +1,771 @@
+"""CLI handoff tests for the offline daily workflow."""
+
+from __future__ import annotations
+
+import textwrap
+from pathlib import Path
+
+import click
+import pandas as pd
+from click.testing import CliRunner
+
+from cli import cli
+import pipeline.derived.cli as derived_cli
+import pipeline.features.cli as features_cli
+from tests.helpers import (
+    make_generated_daily_bars,
+    make_generated_derived_features,
+    make_generated_minute_bars,
+)
+
+
+FIXTURE_PATH = Path(__file__).parent / "fixtures" / "daily_bars_real_2024_01_sample.pq"
+
+
+def _invoke_ok(runner: CliRunner, args: list[str]):
+    result = runner.invoke(cli, args)
+    assert result.exit_code == 0, result.output
+    return result
+
+
+def _invoke_error(runner: CliRunner, args: list[str]):
+    result = runner.invoke(cli, args)
+    assert result.exit_code != 0, result.output
+    return result
+
+
+def test_cli_daily_workflow_handoffs_stay_in_tmp_path(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars()
+    minute_bars = make_generated_minute_bars(daily_bars)
+    derived_features = make_generated_derived_features(daily_bars)
+
+    daily_path = tmp_path / "daily_bars.pq"
+    minute_path = tmp_path / "minute_bars.pq"
+    derived_input_path = tmp_path / "derived_input.pq"
+    daily_bars.to_parquet(daily_path, index=False)
+    minute_bars.to_parquet(minute_path, index=False)
+    derived_features.to_parquet(derived_input_path, index=False)
+
+    ingest_dir = tmp_path / "derived_ingested"
+    ingest_result = _invoke_ok(runner, [
+        "derived", "ingest",
+        "--input-path", str(derived_input_path),
+        "--derived-name", "toy_features",
+        "--output-dir", str(ingest_dir),
+    ])
+    ingested_feature_path = ingest_dir / "toy_features.pq"
+    assert "Saved derived data:" in ingest_result.output
+    assert ingested_feature_path.exists()
+
+    validate_result = _invoke_ok(runner, [
+        "derived", "validate",
+        "--input-path", str(ingested_feature_path),
+    ])
+    assert "Valid derived data:" in validate_result.output
+    assert "rows" in validate_result.output
+
+    computed_derived_dir = tmp_path / "derived_computed"
+    derived_compute_result = _invoke_ok(runner, [
+        "derived", "compute",
+        "--daily-path", str(daily_path),
+        "--minute-path", str(minute_path),
+        "--derived-type", "minute_daily_summary",
+        "--derived-name", "minute_summary",
+        "--output-dir", str(computed_derived_dir),
+    ])
+    minute_summary_path = computed_derived_dir / "minute_summary.pq"
+    assert "Loaded daily data:" in derived_compute_result.output
+    assert "Loaded minute bars:" in derived_compute_result.output
+    assert "Saved derived data:" in derived_compute_result.output
+    assert minute_summary_path.exists()
+    assert "minute_vwap" in pd.read_parquet(minute_summary_path).columns
+
+    alpha_module_path = tmp_path / "cli_feature_alpha.py"
+    alpha_module_path.write_text(textwrap.dedent("""
+        import pandas as pd
+        from pipeline.alpha.base import BaseAlpha
+        from pipeline.alpha.registry import register_alpha
+
+        @register_alpha
+        class CliFeatureAlpha(BaseAlpha):
+            name = "cli_feature_alpha_workflow"
+
+            def __init__(self, **kwargs):
+                self.kwargs = kwargs
+
+            def signal_from_data(
+                self,
+                data: pd.DataFrame,
+                close: pd.DataFrame,
+            ) -> pd.DataFrame:
+                signal = data.pivot_table(
+                    index="date",
+                    columns="symbol_id",
+                    values="minute_intraday_return",
+                    aggfunc="first",
+                )
+                fallback = close.pct_change(1, fill_method=None)
+                feature_signal = signal.reindex(index=close.index, columns=close.columns)
+                toy_signal = data.pivot_table(
+                    index="date",
+                    columns="symbol_id",
+                    values="toy_feature",
+                    aggfunc="first",
+                )
+                toy_signal = toy_signal.reindex(index=close.index, columns=close.columns)
+                return feature_signal.fillna(fallback) + toy_signal / 1000.0
+    """))
+
+    alpha_dir = tmp_path / "alphas"
+    alpha_result = _invoke_ok(runner, [
+        "alpha", "compute",
+        "--data-path", str(daily_path),
+        "--feature-path", str(minute_summary_path),
+        "--feature-path", str(ingested_feature_path),
+        "--alpha-module", str(alpha_module_path),
+        "--alpha-type", "cli_feature_alpha_workflow",
+        "--alpha-name", "cli_feature_alpha",
+        "--output-dir", str(alpha_dir),
+    ])
+    alpha_path = alpha_dir / "cli_feature_alpha.pq"
+    assert "Loaded data:" in alpha_result.output
+    assert "Saved alpha:" in alpha_result.output
+    assert "Weight stats" in alpha_result.output
+    assert alpha_path.exists()
+    assert not pd.read_parquet(alpha_path).empty
+
+    alpha_report_dir = tmp_path / "alpha_reports"
+    alpha_eval_result = _invoke_ok(runner, [
+        "alpha", "eval",
+        "--alpha-path", str(alpha_path),
+        "--data-path", str(daily_path),
+        "--report-dir", str(alpha_report_dir),
+    ])
+    alpha_report_path = alpha_report_dir / "cli_feature_alpha_eval.json"
+    assert "ALPHA EVALUATION" in alpha_eval_result.output
+    assert "Report saved:" in alpha_eval_result.output
+    assert alpha_report_path.exists()
+
+    combo_dir = tmp_path / "combos"
+    combo_result = _invoke_ok(runner, [
+        "combo", "combine",
+        "--alpha-paths", f"{alpha_path},{alpha_path}",
+        "--combo-name", "cli_combo",
+        "--method", "equal_weight",
+        "--output-dir", str(combo_dir),
+    ])
+    combo_path = combo_dir / "cli_combo.pq"
+    assert "Saved combo:" in combo_result.output
+    assert "Weight stats" in combo_result.output
+    assert combo_path.exists()
+
+    portfolio_dir = tmp_path / "portfolio"
+    build_result = _invoke_ok(runner, [
+        "portfolio", "build",
+        "--weights-path", str(combo_path),
+        "--data-path", str(daily_path),
+        "--booksize", "2000000",
+        "--portfolio-name", "cli_portfolio",
+        "--output-dir", str(portfolio_dir),
+    ])
+    positions_path = portfolio_dir / "cli_portfolio.pq"
+    assert "Saved positions:" in build_result.output
+    assert "Gross exposure" in build_result.output
+    assert positions_path.exists()
+
+    execution_dir = tmp_path / "execution"
+    simulate_result = _invoke_ok(runner, [
+        "portfolio", "simulate",
+        "--positions-path", str(positions_path),
+        "--data-path", str(daily_path),
+        "--constraint", "suspension",
+        "--constraint", "price_limit",
+        "--constraint", "volume_cap",
+        "--cost-bps", "5",
+        "--slippage-bps", "5",
+        "--volume-frac", "0.02",
+        "--output-dir", str(execution_dir),
+    ])
+    fills_path = execution_dir / "fills" / "cli_portfolio.pq"
+    pnl_path = execution_dir / "pnl" / "cli_portfolio.pq"
+    assert "Saved fills:" in simulate_result.output
+    assert "Saved pnl:" in simulate_result.output
+    assert "Total PnL:" in simulate_result.output
+    assert fills_path.exists()
+    assert pnl_path.exists()
+
+    eval_result = _invoke_ok(runner, [
+        "portfolio", "eval",
+        "--positions-path", str(positions_path),
+        "--data-path", str(daily_path),
+    ])
+    assert "Research-portfolio metrics:" in eval_result.output
+    assert "cumulative_return" in eval_result.output
+    assert "fitness" in eval_result.output
+
+    pqcat_result = _invoke_ok(runner, [
+        "pqcat",
+        str(positions_path),
+        "--info",
+    ])
+    assert "shape:" in pqcat_result.output
+    assert "dtypes:" in pqcat_result.output
+    assert "position_shares" in pqcat_result.output
+
+    alphaview_result = _invoke_ok(runner, [
+        "alphaview",
+        "--data-path", str(daily_path),
+        "--alpha-path", str(alpha_path),
+        "--symbol", "sh600000",
+        "--start-date", "2024-01-02",
+        "--end-date", "2024-01-12",
+        "--columns", "close,volume",
+    ])
+    assert "symbol: sh600000" in alphaview_result.output
+    assert "cli_feature_alpha" in alphaview_result.output
+
+
+def test_cli_pipeline_accepts_partitioned_daily_dataset(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars(include_missing=False)
+    dataset_dir = tmp_path / "daily_dataset"
+    dataset_frame = daily_bars.copy()
+    dataset_frame["month"] = dataset_frame["date"].dt.strftime("%Y-%m")
+    dataset_frame.to_parquet(dataset_dir, partition_cols=["month"], index=False)
+
+    alpha_dir = tmp_path / "alphas"
+    alpha_result = _invoke_ok(runner, [
+        "alpha", "compute",
+        "--data-path", str(dataset_dir),
+        "--alpha-type", "reversal",
+        "--alpha-name", "dataset_reversal",
+        "--lookback", "3",
+        "--output-dir", str(alpha_dir),
+    ])
+    alpha_path = alpha_dir / "dataset_reversal.pq"
+    assert "Loaded data:" in alpha_result.output
+    assert alpha_path.exists()
+
+    combo_dir = tmp_path / "combos"
+    _invoke_ok(runner, [
+        "combo", "combine",
+        "--alpha-paths", str(alpha_path),
+        "--combo-name", "dataset_combo",
+        "--output-dir", str(combo_dir),
+    ])
+    combo_path = combo_dir / "dataset_combo.pq"
+    assert combo_path.exists()
+
+    portfolio_dir = tmp_path / "portfolio"
+    _invoke_ok(runner, [
+        "portfolio", "build",
+        "--weights-path", str(combo_path),
+        "--data-path", str(dataset_dir),
+        "--booksize", "1000000",
+        "--portfolio-name", "dataset_portfolio",
+        "--output-dir", str(portfolio_dir),
+    ])
+    positions_path = portfolio_dir / "dataset_portfolio.pq"
+    assert positions_path.exists()
+
+    execution_dir = tmp_path / "execution"
+    simulate_result = _invoke_ok(runner, [
+        "portfolio", "simulate",
+        "--positions-path", str(positions_path),
+        "--data-path", str(dataset_dir),
+        "--constraint", "suspension",
+        "--output-dir", str(execution_dir),
+    ])
+    assert "Saved fills:" in simulate_result.output
+    assert (execution_dir / "fills" / "dataset_portfolio.pq").exists()
+    assert (execution_dir / "pnl" / "dataset_portfolio.pq").exists()
+
+
+def test_cli_liquid_universe_masks_to_top_liquid_names(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars(n_sessions=75, include_missing=False)
+    daily_path = tmp_path / "daily_bars_75d.pq"
+    daily_bars.to_parquet(daily_path, index=False)
+
+    alpha_dir = tmp_path / "alphas"
+    result = _invoke_ok(runner, [
+        "alpha", "compute",
+        "--data-path", str(daily_path),
+        "--alpha-type", "reversal_rank",
+        "--alpha-name", "liquid_rank",
+        "--lookback", "3",
+        "--liquid-universe",
+        "--universe-top-n", "2",
+        "--output-dir", str(alpha_dir),
+    ])
+
+    alpha_path = alpha_dir / "liquid_rank.pq"
+    alpha = pd.read_parquet(alpha_path)
+    nonzero = alpha[alpha["weight"] != 0.0]
+    assert "Saved alpha:" in result.output
+    assert alpha_path.exists()
+    assert not nonzero.empty
+    assert nonzero.groupby("date")["symbol_id"].nunique().max() <= 2
+
+
+def test_cli_real_fixture_round_trips_through_portfolio(tmp_path):
+    runner = CliRunner()
+
+    alpha_dir = tmp_path / "alphas"
+    _invoke_ok(runner, [
+        "alpha", "compute",
+        "--data-path", str(FIXTURE_PATH),
+        "--alpha-type", "reversal_vol",
+        "--alpha-name", "real_cli_reversal_vol",
+        "--lookback", "3",
+        "--vol-window", "3",
+        "--output-dir", str(alpha_dir),
+    ])
+    alpha_path = alpha_dir / "real_cli_reversal_vol.pq"
+    assert alpha_path.exists()
+    assert not pd.read_parquet(alpha_path).empty
+
+    combo_dir = tmp_path / "combos"
+    _invoke_ok(runner, [
+        "combo", "combine",
+        "--alpha-paths", str(alpha_path),
+        "--combo-name", "real_cli_combo",
+        "--output-dir", str(combo_dir),
+    ])
+    combo_path = combo_dir / "real_cli_combo.pq"
+    assert combo_path.exists()
+
+    portfolio_dir = tmp_path / "portfolio"
+    _invoke_ok(runner, [
+        "portfolio", "build",
+        "--weights-path", str(combo_path),
+        "--data-path", str(FIXTURE_PATH),
+        "--booksize", "1000000",
+        "--portfolio-name", "real_cli_portfolio",
+        "--output-dir", str(portfolio_dir),
+    ])
+    positions_path = portfolio_dir / "real_cli_portfolio.pq"
+    positions = pd.read_parquet(positions_path)
+    assert not positions.empty
+
+    eval_result = _invoke_ok(runner, [
+        "portfolio", "eval",
+        "--positions-path", str(positions_path),
+        "--data-path", str(FIXTURE_PATH),
+    ])
+    assert "Research-portfolio metrics:" in eval_result.output
+
+
+def test_cli_error_paths_are_clear_for_bad_user_inputs(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars()
+    daily_path = tmp_path / "daily_bars.pq"
+    daily_bars.to_parquet(daily_path, index=False)
+
+    unknown_alpha = _invoke_error(runner, [
+        "alpha", "compute",
+        "--data-path", str(daily_path),
+        "--alpha-type", "does_not_exist",
+        "--alpha-name", "bad",
+        "--output-dir", str(tmp_path / "alphas"),
+    ])
+    assert "Unknown alpha-type" in unknown_alpha.output
+
+    malformed_param = _invoke_error(runner, [
+        "alpha", "compute",
+        "--data-path", str(daily_path),
+        "--alpha-type", "reversal",
+        "--alpha-name", "bad_param",
+        "--param", "not-an-assignment",
+        "--output-dir", str(tmp_path / "alphas"),
+    ])
+    assert "--param must be name=value" in malformed_param.output
+
+    unknown_derived = _invoke_error(runner, [
+        "derived", "compute",
+        "--daily-path", str(daily_path),
+        "--derived-type", "does_not_exist",
+        "--derived-name", "bad",
+        "--output-dir", str(tmp_path / "derived"),
+    ])
+    assert "Unknown derived-type" in unknown_derived.output
+
+    bad_constraint_positions = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [pd.Timestamp("2024-01-02")],
+        "portfolio_name": ["bad_constraint"],
+        "target_weight": [1.0],
+        "target_value": [1000.0],
+        "target_shares": [100.0],
+        "position_shares": [100],
+        "position_value": [1000.0],
+        "price": [10.0],
+    })
+    positions_path = tmp_path / "positions.pq"
+    bad_constraint_positions.to_parquet(positions_path, index=False)
+    unknown_constraint = _invoke_error(runner, [
+        "portfolio", "simulate",
+        "--positions-path", str(positions_path),
+        "--data-path", str(daily_path),
+        "--constraint", "not_a_constraint",
+        "--output-dir", str(tmp_path / "execution"),
+    ])
+    assert isinstance(unknown_constraint.exception, KeyError)
+    assert "not_a_constraint" in str(unknown_constraint.exception)
+
+    pqcat_missing_column = _invoke_error(runner, [
+        "pqcat",
+        str(daily_path),
+        "--columns", "close,not_a_column",
+    ])
+    assert "Columns not found: not_a_column" in pqcat_missing_column.output
+
+    alphaview_missing_symbol = _invoke_error(runner, [
+        "alphaview",
+        "--data-path", str(daily_path),
+        "--alpha-path", str(positions_path),
+        "--symbol", "sh999999",
+    ])
+    assert "Symbol 'sh999999' not found" in alphaview_missing_symbol.output
+
+    alphaview_missing_column = _invoke_error(runner, [
+        "alphaview",
+        "--data-path", str(daily_path),
+        "--alpha-path", str(positions_path),
+        "--symbol", "sh600000",
+        "--columns", "close,missing_bar_col",
+    ])
+    assert "Bar columns not found: missing_bar_col" in alphaview_missing_column.output
+
+    alphaview_alpha = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [daily_bars["date"].min()],
+        "alpha_name": ["toy_alpha"],
+        "weight": [1.0],
+    })
+    alphaview_alpha_path = tmp_path / "alphaview_alpha.pq"
+    alphaview_alpha.to_parquet(alphaview_alpha_path, index=False)
+    alphaview_empty_range = _invoke_error(runner, [
+        "alphaview",
+        "--data-path", str(daily_path),
+        "--alpha-path", str(alphaview_alpha_path),
+        "--symbol", "sh600000",
+        "--start-date", "2030-01-01",
+    ])
+    assert "No rows in the requested date range" in alphaview_empty_range.output
+
+    empty_combo_paths = _invoke_ok(runner, [
+        "combo", "combine",
+        "--alpha-paths", " , ",
+        "--combo-name", "empty",
+        "--output-dir", str(tmp_path / "combos"),
+    ])
+    assert "requires at least 1 path" in empty_combo_paths.output
+
+
+def test_cli_parser_helpers_cover_string_coercion_and_bad_params():
+    assert derived_cli._parse_params(("n=7", "scale=2.5", "label=demo")) == {
+        "n": 7,
+        "scale": 2.5,
+        "label": "demo",
+    }
+    assert features_cli._parse_params(("n=7", "scale=2.5", "label=demo")) == {
+        "n": 7,
+        "scale": 2.5,
+        "label": "demo",
+    }
+
+    for module in (derived_cli, features_cli):
+        try:
+            module._parse_params(("not-an-assignment",))
+        except click.BadParameter as exc:
+            assert "--param must be name=value" in str(exc)
+        else:
+            raise AssertionError("expected BadParameter")
+
+
+def test_cli_list_and_legacy_feature_paths(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars(n_sessions=3, include_missing=False)
+    minute_bars = make_generated_minute_bars(daily_bars)
+    daily_path = tmp_path / "daily_bars.pq"
+    minute_path = tmp_path / "minute_bars.pq"
+    daily_bars.to_parquet(daily_path, index=False)
+    minute_bars.to_parquet(minute_path, index=False)
+
+    derived_list = _invoke_ok(runner, ["derived", "list"])
+    feature_list = _invoke_ok(runner, ["feature", "list"])
+    assert "minute_daily_summary" in derived_list.output
+    assert "minute_daily_summary" in feature_list.output
+
+    feature_dir = tmp_path / "features"
+    feature_compute = _invoke_ok(runner, [
+        "feature",
+        "compute",
+        "--minute-path",
+        str(minute_path),
+        "--daily-path",
+        str(daily_path),
+        "--feature-type",
+        "minute_daily_summary",
+        "--feature-name",
+        "legacy_summary",
+        "--output-dir",
+        str(feature_dir),
+    ])
+    feature_path = feature_dir / "legacy_summary.pq"
+    assert "Loaded minute bars:" in feature_compute.output
+    assert "Loaded daily data:" in feature_compute.output
+    assert "Saved feature:" in feature_compute.output
+    assert feature_path.exists()
+
+    no_input = _invoke_error(runner, [
+        "derived",
+        "compute",
+        "--derived-type",
+        "minute_daily_summary",
+        "--derived-name",
+        "missing_inputs",
+        "--output-dir",
+        str(tmp_path / "derived"),
+    ])
+    assert "At least one of --daily-path or --minute-path is required" in no_input.output
+
+    missing_minute = _invoke_error(runner, [
+        "derived",
+        "compute",
+        "--daily-path",
+        str(daily_path),
+        "--derived-type",
+        "minute_daily_summary",
+        "--derived-name",
+        "daily_only",
+        "--output-dir",
+        str(tmp_path / "derived"),
+    ])
+    assert "minute_daily_summary requires minute input" in missing_minute.output
+
+    malformed_feature_param = _invoke_error(runner, [
+        "feature",
+        "compute",
+        "--minute-path",
+        str(minute_path),
+        "--feature-type",
+        "minute_daily_summary",
+        "--feature-name",
+        "bad_param",
+        "--param",
+        "not-an-assignment",
+        "--output-dir",
+        str(tmp_path / "features_bad"),
+    ])
+    assert "--param must be name=value" in malformed_feature_param.output
+
+    unknown_feature = _invoke_error(runner, [
+        "feature",
+        "compute",
+        "--minute-path",
+        str(minute_path),
+        "--feature-type",
+        "does_not_exist",
+        "--feature-name",
+        "bad_feature",
+        "--output-dir",
+        str(tmp_path / "features_unknown"),
+    ])
+    assert "Unknown feature-type" in unknown_feature.output
+
+
+def test_cli_shortcuts_and_external_module_loading(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars(n_sessions=8, include_missing=False)
+    minute_bars = make_generated_minute_bars(daily_bars)
+    daily_path = tmp_path / "daily_bars.pq"
+    minute_path = tmp_path / "minute_bars.pq"
+    daily_bars.to_parquet(daily_path, index=False)
+    minute_bars.to_parquet(minute_path, index=False)
+
+    alpha_module = tmp_path / "listed_alpha.py"
+    alpha_module.write_text(textwrap.dedent("""
+        import pandas as pd
+        from pipeline.alpha.base import BaseAlpha
+        from pipeline.alpha.registry import register_alpha
+
+        @register_alpha
+        class ListedAlpha(BaseAlpha):
+            name = "listed_alpha_cli"
+
+            def __init__(self, scale: float = 1.0, label: str = "x"):
+                self.scale = scale
+                self.label = label
+
+            def signal(self, close: pd.DataFrame) -> pd.DataFrame:
+                return close.pct_change(1, fill_method=None) * self.scale
+    """))
+    derived_module = tmp_path / "listed_derived.py"
+    derived_module.write_text(textwrap.dedent("""
+        import pandas as pd
+        from pipeline.derived.base import BaseDerivedData
+        from pipeline.derived.registry import register_derived
+
+        @register_derived
+        class ListedDerived(BaseDerivedData):
+            name = "listed_derived_cli"
+
+            def __init__(self, scale: float = 1.0):
+                self.scale = scale
+
+            def compute(self, daily=None, minute=None) -> pd.DataFrame:
+                out = daily[["symbol_id", "date", "close"]].copy()
+                out["listed_value"] = out.pop("close") * self.scale
+                return out
+    """))
+    derived_compute_module = tmp_path / "computed_derived.py"
+    derived_compute_module.write_text(textwrap.dedent("""
+        import pandas as pd
+        from pipeline.derived.base import BaseDerivedData
+        from pipeline.derived.registry import register_derived
+
+        @register_derived
+        class ComputedDerived(BaseDerivedData):
+            name = "computed_derived_cli"
+
+            def __init__(self, scale: float = 1.0):
+                self.scale = scale
+
+            def compute(self, daily=None, minute=None) -> pd.DataFrame:
+                out = daily[["symbol_id", "date", "close"]].copy()
+                out["computed_value"] = out.pop("close") * self.scale
+                return out
+    """))
+    feature_module = tmp_path / "listed_feature.py"
+    feature_module.write_text(textwrap.dedent("""
+        import pandas as pd
+        from pipeline.features.base import BaseFeature
+        from pipeline.features.registry import register_feature
+
+        @register_feature
+        class ListedFeature(BaseFeature):
+            name = "listed_feature_cli"
+
+            def compute(self, daily=None, minute=None) -> pd.DataFrame:
+                out = minute[["symbol_id", "date", "close"]].copy()
+                out["date"] = pd.to_datetime(out["date"]).dt.normalize()
+                out = out.groupby(["symbol_id", "date"], as_index=False)["close"].mean()
+                return out.rename(columns={"close": "listed_feature_value"})
+    """))
+    feature_compute_module = tmp_path / "computed_feature.py"
+    feature_compute_module.write_text(textwrap.dedent("""
+        import pandas as pd
+        from pipeline.features.base import BaseFeature
+        from pipeline.features.registry import register_feature
+
+        @register_feature
+        class ComputedFeature(BaseFeature):
+            name = "computed_feature_cli"
+
+            def compute(self, daily=None, minute=None) -> pd.DataFrame:
+                out = minute[["symbol_id", "date", "close"]].copy()
+                out["date"] = pd.to_datetime(out["date"]).dt.normalize()
+                out = out.groupby(["symbol_id", "date"], as_index=False)["close"].mean()
+                return out.rename(columns={"close": "computed_feature_value"})
+    """))
+
+    alpha_list = _invoke_ok(runner, [
+        "alpha", "list",
+        "--alpha-module", str(alpha_module),
+    ])
+    assert "listed_alpha_cli" in alpha_list.output
+
+    alpha_dir = tmp_path / "alphas"
+    reversal = _invoke_ok(runner, [
+        "alpha", "reversal",
+        "--data-path", str(daily_path),
+        "--output-dir", str(alpha_dir),
+        "--lookback", "3",
+    ])
+    reversal_vol = _invoke_ok(runner, [
+        "alpha", "reversal-vol",
+        "--data-path", str(daily_path),
+        "--output-dir", str(alpha_dir),
+        "--lookback", "3",
+        "--vol-window", "3",
+    ])
+    external_alpha = _invoke_ok(runner, [
+        "alpha", "compute",
+        "--data-path", str(daily_path),
+        "--alpha-type", "listed_alpha_cli",
+        "--alpha-name", "listed_alpha_run",
+        "--param", "scale=2.5",
+        "--param", "label=demo",
+        "--output-dir", str(alpha_dir),
+    ])
+
+    assert "Saved alpha:" in reversal.output
+    assert "Saved alpha:" in reversal_vol.output
+    assert "Saved alpha:" in external_alpha.output
+    assert (alpha_dir / "reversal_3d.pq").exists()
+    assert (alpha_dir / "reversal_vol_3d_3d.pq").exists()
+    assert (alpha_dir / "listed_alpha_run.pq").exists()
+
+    derived_list = _invoke_ok(runner, [
+        "derived", "list",
+        "--derived-module", str(derived_module),
+    ])
+    assert "listed_derived_cli" in derived_list.output
+    derived_dir = tmp_path / "derived_external"
+    derived_compute = _invoke_ok(runner, [
+        "derived", "compute",
+        "--daily-path", str(daily_path),
+        "--derived-module", str(derived_compute_module),
+        "--derived-type", "computed_derived_cli",
+        "--derived-name", "listed_derived_run",
+        "--param", "scale=3",
+        "--output-dir", str(derived_dir),
+    ])
+    assert "Saved derived data:" in derived_compute.output
+    assert (derived_dir / "listed_derived_run.pq").exists()
+
+    feature_list = _invoke_ok(runner, [
+        "feature", "list",
+        "--feature-module", str(feature_module),
+    ])
+    assert "listed_feature_cli" in feature_list.output
+    feature_dir = tmp_path / "features_external"
+    feature_compute = _invoke_ok(runner, [
+        "feature", "compute",
+        "--minute-path", str(minute_path),
+        "--feature-module", str(feature_compute_module),
+        "--feature-type", "computed_feature_cli",
+        "--feature-name", "listed_feature_run",
+        "--output-dir", str(feature_dir),
+    ])
+    assert "Saved feature:" in feature_compute.output
+    assert (feature_dir / "listed_feature_run.pq").exists()
+
+
+def test_cli_pqcat_row_modes(tmp_path):
+    runner = CliRunner()
+    daily_bars = make_generated_daily_bars(n_sessions=3, include_missing=False)
+    daily_path = tmp_path / "daily_bars.pq"
+    daily_bars.to_parquet(daily_path, index=False)
+
+    head_result = _invoke_ok(runner, [
+        "pqcat",
+        str(daily_path),
+        "--head",
+        "2",
+        "--columns",
+        "symbol_id,close",
+    ])
+    tail_result = _invoke_ok(runner, [
+        "pqcat",
+        str(daily_path),
+        "--tail",
+        "1",
+    ])
+
+    assert "symbol_id" in head_result.output
+    assert "close" in head_result.output
+    assert "date" in tail_result.output

tests/test_data_cli_and_resolve.py

@@ -0,0 +1,148 @@
+"""Offline coverage for data CLI and universe resolution glue."""
+
+from __future__ import annotations
+
+import pandas as pd
+from click.testing import CliRunner
+
+from cli import cli
+import pipeline.data.cli as data_cli
+import pipeline.data.downloader as pipeline_downloader
+
+
+def test_resolve_universe_handles_named_file_all_and_symbol_list(tmp_path, monkeypatch):
+    hs300_raw = pd.DataFrame({
+        "updateDate": ["2024-01-12", "2024-01-12"],
+        "stockName": ["浦发银行", "平安银行"],
+        "stockCode": ["sh.600000", "sz.000001"],
+    })
+    zz500_raw = pd.DataFrame({
+        "name": ["东风汽车"],
+        "code": ["sh.600006"],
+        "date": ["2024-01-12"],
+    })
+
+    monkeypatch.setattr(pipeline_downloader, "get_hs300_stocks", lambda: hs300_raw)
+    monkeypatch.setattr(pipeline_downloader, "get_zz500_stocks", lambda: zz500_raw)
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "get_all_stocks",
+        lambda: pd.DataFrame({
+            "code": ["sh600000", "sz000001", "sh600519"],
+            "name": ["浦发银行", "平安银行", "贵州茅台"],
+        }),
+    )
+
+    symbol_file = tmp_path / "symbols.txt"
+    symbol_file.write_text("sh600000\n\nsz000001\n")
+
+    hs300 = pipeline_downloader._resolve_universe("hs300")
+    zz500 = pipeline_downloader._resolve_universe("csi500")
+    all_capped = pipeline_downloader._resolve_universe("all", max_symbols=2)
+    from_file = pipeline_downloader._resolve_universe(str(symbol_file))
+    from_list = pipeline_downloader._resolve_universe("sh600000, sz000001")
+
+    assert hs300.to_dict("list") == {
+        "symbol_name": ["浦发银行", "平安银行"],
+        "symbol_id": ["sh600000", "sz000001"],
+    }
+    assert zz500.to_dict("list") == {
+        "symbol_name": ["东风汽车"],
+        "symbol_id": ["sh600006"],
+    }
+    assert all_capped["symbol_id"].tolist() == ["sh600000", "sz000001"]
+    assert from_file["symbol_id"].tolist() == ["sh600000", "sz000001"]
+    assert from_file["symbol_name"].tolist() == ["sh600000", "sz000001"]
+    assert from_list["symbol_id"].tolist() == ["sh600000", "sz000001"]
+
+
+def test_data_cli_download_commands_print_summaries_without_network(monkeypatch, tmp_path):
+    runner = CliRunner()
+    daily_calls: list[dict] = []
+    minute_calls: list[dict] = []
+
+    def fake_daily(**kwargs):
+        daily_calls.append(kwargs)
+        return {
+            "dataset_path": str(tmp_path / "daily" / kwargs["universe"]),
+            "n_symbols": 2,
+            "n_requested": 3,
+            "n_rows": 18,
+            "date_min": "2024-01-02",
+            "date_max": "2024-01-12",
+        }
+
+    def fake_minute(**kwargs):
+        minute_calls.append(kwargs)
+        return {
+            "dataset_path": str(tmp_path / "minute" / kwargs["universe"]),
+            "frequency": "15m",
+            "n_symbols": 1,
+            "n_requested": 1,
+            "n_rows": 32,
+            "date_min": "2024-01-02",
+            "date_max": "2024-01-03",
+        }
+
+    monkeypatch.setattr(data_cli, "download_universe", fake_daily)
+    monkeypatch.setattr(data_cli, "download_minute_universe", fake_minute)
+
+    daily_result = runner.invoke(cli, [
+        "data",
+        "download",
+        "--universe",
+        "sh600000,sz000001",
+        "--start-date",
+        "2024-01-02",
+        "--end-date",
+        "2024-01-12",
+        "--output-dir",
+        str(tmp_path / "daily"),
+        "--symbols",
+        "3",
+        "--chunk-size",
+        "2",
+        "--adjust",
+        "none",
+    ])
+    minute_result = runner.invoke(cli, [
+        "data",
+        "download-minute",
+        "--universe",
+        "toy",
+        "--start-date",
+        "2024-01-02",
+        "--end-date",
+        "2024-01-03",
+        "--output-dir",
+        str(tmp_path / "minute"),
+        "--symbols",
+        "1",
+        "--chunk-size",
+        "1",
+        "--frequency",
+        "15",
+    ])
+
+    assert daily_result.exit_code == 0, daily_result.output
+    assert "Summary: 2/3 symbols, 18 bars" in daily_result.output
+    assert daily_calls == [{
+        "universe": "sh600000,sz000001",
+        "start_date": "2024-01-02",
+        "end_date": "2024-01-12",
+        "output_dir": str(tmp_path / "daily"),
+        "max_symbols": 3,
+        "chunk_size": 2,
+        "adjust": "none",
+    }]
+    assert minute_result.exit_code == 0, minute_result.output
+    assert "frequency=15m" in minute_result.output
+    assert minute_calls == [{
+        "universe": "toy",
+        "start_date": "2024-01-02",
+        "end_date": "2024-01-03",
+        "output_dir": str(tmp_path / "minute"),
+        "max_symbols": 1,
+        "chunk_size": 1,
+        "frequency": "15",
+    }]

tests/test_derived.py

@@ -0,0 +1,376 @@
+"""Tests for daily derived-data ingestion and plugins."""
+
+import textwrap
+
+import numpy as np
+import pandas as pd
+import pytest
+from click.testing import CliRunner
+
+from cli import cli
+from pipeline.alpha.compute import join_feature_frames
+from pipeline.derived.base import BaseDerivedData
+from pipeline.derived.compute import compute_derived, read_derived_frame, validate_derived_frame
+from pipeline.derived.registry import (
+    available_derived,
+    get_derived,
+    load_derived_module,
+    register_derived,
+)
+
+
+def _daily_bars() -> pd.DataFrame:
+    return pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000"],
+        "date": pd.to_datetime(["2024-01-02", "2024-01-02", "2024-01-03"]),
+        "open": [10.0, 20.0, 11.0],
+        "close": [10.5, 20.5, 11.5],
+        "volume": [1000.0, 2000.0, 1200.0],
+    })
+
+
+def _minute_bars() -> pd.DataFrame:
+    return pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000", "sz000001"],
+        "datetime": pd.to_datetime([
+            "2024-01-02 09:35:00",
+            "2024-01-02 09:40:00",
+            "2024-01-02 09:35:00",
+        ]),
+        "date": pd.to_datetime(["2024-01-02", "2024-01-02", "2024-01-02"]),
+        "time": ["09:35:00", "09:40:00", "09:35:00"],
+        "open": [10.0, 10.5, 20.0],
+        "high": [11.0, 12.0, 21.0],
+        "low": [9.0, 10.0, 19.0],
+        "close": [10.5, 11.0, 20.5],
+        "volume": [100.0, 300.0, 200.0],
+        "amount": [1000.0, 3300.0, 4100.0],
+    })
+
+
+def test_validate_derived_frame_normalizes_and_sorts():
+    result = validate_derived_frame(pd.DataFrame({
+        "symbol_id": ["sz000001", "sh600000"],
+        "date": ["2024-01-02 15:00:00", "2024-01-02 09:30:00"],
+        "custom_value": [2.0, 1.0],
+    }))
+
+    assert result["symbol_id"].tolist() == ["sh600000", "sz000001"]
+    assert result["date"].tolist() == [
+        pd.Timestamp("2024-01-02"),
+        pd.Timestamp("2024-01-02"),
+    ]
+
+
+def test_validate_derived_frame_rejects_missing_keys():
+    with pytest.raises(ValueError, match="missing required"):
+        validate_derived_frame(pd.DataFrame({"symbol_id": ["sh600000"], "x": [1.0]}))
+
+
+def test_validate_derived_frame_rejects_duplicate_normalized_keys():
+    with pytest.raises(ValueError, match="duplicate symbol_id,date"):
+        validate_derived_frame(pd.DataFrame({
+            "symbol_id": ["sh600000", "sh600000"],
+            "date": ["2024-01-02 09:30:00", "2024-01-02 15:00:00"],
+            "x": [1.0, 2.0],
+        }))
+
+
+def test_validate_derived_frame_rejects_duplicate_columns():
+    bad = pd.DataFrame(
+        [["sh600000", pd.Timestamp("2024-01-02"), 1.0, 2.0]],
+        columns=["symbol_id", "date", "dup", "dup"],
+    )
+    with pytest.raises(ValueError, match="duplicate columns"):
+        validate_derived_frame(bad)
+
+
+def test_validate_derived_frame_rejects_non_numeric_values():
+    with pytest.raises(ValueError, match="numeric"):
+        validate_derived_frame(pd.DataFrame({
+            "symbol_id": ["sh600000"],
+            "date": [pd.Timestamp("2024-01-02")],
+            "bad": ["not numeric"],
+        }))
+
+
+def test_validate_derived_frame_rejects_missing_value_columns():
+    with pytest.raises(ValueError, match="at least one value column"):
+        validate_derived_frame(pd.DataFrame({
+            "symbol_id": ["sh600000"],
+            "date": [pd.Timestamp("2024-01-02")],
+        }))
+
+
+def test_read_derived_frame_rejects_empty_csv(tmp_path):
+    empty_csv = tmp_path / "empty.csv"
+    empty_csv.write_text("")
+
+    with pytest.raises(ValueError, match="CSV input is empty"):
+        read_derived_frame(empty_csv)
+
+
+def test_compute_derived_rejects_missing_inputs():
+    with pytest.raises(ValueError, match="requires --daily-path or --minute-path"):
+        compute_derived("minute_daily_summary")
+
+
+def test_derived_ingest_cli_accepts_csv_and_parquet(tmp_path):
+    runner = CliRunner()
+    source = pd.DataFrame({
+        "symbol_id": ["sz000001", "sh600000"],
+        "date": ["2024-01-02", "2024-01-02"],
+        "custom_value": [2.0, 1.0],
+    })
+    csv_path = tmp_path / "custom.csv"
+    parquet_path = tmp_path / "custom.pq"
+    out_dir = tmp_path / "derived"
+    source.to_csv(csv_path, index=False)
+    source.to_parquet(parquet_path, index=False)
+
+    csv_result = runner.invoke(cli, [
+        "derived", "ingest",
+        "--input-path", str(csv_path),
+        "--derived-name", "csv_custom",
+        "--output-dir", str(out_dir),
+    ])
+    assert csv_result.exit_code == 0, csv_result.output
+
+    parquet_result = runner.invoke(cli, [
+        "derived", "ingest",
+        "--input-path", str(parquet_path),
+        "--derived-name", "parquet_custom",
+        "--output-dir", str(out_dir),
+    ])
+    assert parquet_result.exit_code == 0, parquet_result.output
+
+    written = pd.read_parquet(out_dir / "csv_custom.pq")
+    assert written["symbol_id"].tolist() == ["sh600000", "sz000001"]
+    assert (out_dir / "parquet_custom.pq").exists()
+
+
+def test_derived_validate_cli_rejects_duplicate_csv_columns(tmp_path):
+    runner = CliRunner()
+    csv_path = tmp_path / "bad.csv"
+    csv_path.write_text("symbol_id,date,x,x\nsh600000,2024-01-02,1.0,2.0\n")
+
+    result = runner.invoke(cli, [
+        "derived", "validate",
+        "--input-path", str(csv_path),
+    ])
+
+    assert result.exit_code != 0
+    assert "duplicate columns" in result.output
+
+
+def test_derived_ingest_cli_wraps_validation_errors(tmp_path):
+    runner = CliRunner()
+    csv_path = tmp_path / "bad_ingest.csv"
+    csv_path.write_text("symbol_id,date\nsh600000,2024-01-02\n")
+
+    result = runner.invoke(cli, [
+        "derived",
+        "ingest",
+        "--input-path",
+        str(csv_path),
+        "--derived-name",
+        "bad",
+        "--output-dir",
+        str(tmp_path / "derived"),
+    ])
+
+    assert result.exit_code != 0
+    assert "at least one value column" in result.output
+
+
+def test_external_derived_plugin_loads_filters_params_and_uses_inputs(tmp_path):
+    module_path = tmp_path / "external_derived.py"
+    module_path.write_text(textwrap.dedent('''
+        import pandas as pd
+        from pipeline.derived.base import BaseDerivedData
+        from pipeline.derived.registry import register_derived
+
+        @register_derived
+        class FlexibleDerived(BaseDerivedData):
+            name = "flexible_derived_test"
+
+            def __init__(self, scale: float = 1.0):
+                self.scale = scale
+
+            def compute(self, daily=None, minute=None) -> pd.DataFrame:
+                result = None
+                if daily is not None:
+                    result = daily[["symbol_id", "date", "close"]].copy()
+                    result["daily_scaled_close"] = result.pop("close") * self.scale
+                if minute is not None:
+                    minute_out = (
+                        minute.groupby(["symbol_id", "date"], as_index=False)["volume"]
+                        .sum()
+                        .rename(columns={"volume": "minute_volume_sum"})
+                    )
+                    minute_out["minute_volume_sum"] *= self.scale
+                    result = minute_out if result is None else result.merge(
+                        minute_out, on=["symbol_id", "date"], how="left"
+                    )
+                return result
+    '''))
+
+    load_derived_module(str(module_path))
+    assert "flexible_derived_test" in available_derived()
+
+    instance = get_derived("flexible_derived_test", scale=2.0, ignored=99)
+    assert instance.scale == 2.0
+    assert not hasattr(instance, "ignored")
+
+    daily_result = compute_derived(
+        "flexible_derived_test",
+        daily=_daily_bars(),
+        scale=2.0,
+        ignored=99,
+    )
+    assert "daily_scaled_close" in daily_result.columns
+    assert np.isclose(daily_result["daily_scaled_close"].iloc[0], 21.0)
+
+    minute_result = compute_derived(
+        "flexible_derived_test",
+        minute=_minute_bars(),
+        scale=2.0,
+    )
+    assert "minute_volume_sum" in minute_result.columns
+    assert np.isclose(
+        minute_result.loc[minute_result["symbol_id"] == "sh600000", "minute_volume_sum"].iloc[0],
+        800.0,
+    )
+
+    both_result = compute_derived(
+        "flexible_derived_test",
+        daily=_daily_bars(),
+        minute=_minute_bars(),
+        scale=1.0,
+    )
+    assert {"daily_scaled_close", "minute_volume_sum"}.issubset(both_result.columns)
+
+
+def test_derived_registry_rejects_bad_plugins_and_load_failures(tmp_path, monkeypatch):
+    with pytest.raises(TypeError):
+        register_derived(object)  # type: ignore[arg-type]
+
+    with pytest.raises(ValueError, match="non-empty"):
+        @register_derived
+        class NoNameDerived(BaseDerivedData):
+            def compute(self, daily=None, minute=None):
+                return pd.DataFrame()
+
+    @register_derived
+    class _CoverageDerived(BaseDerivedData):
+        name = "_coverage_derived_registry"
+
+        def compute(self, daily=None, minute=None):
+            return pd.DataFrame({
+                "symbol_id": ["sh600000"],
+                "date": [pd.Timestamp("2024-01-02")],
+                "x": [1.0],
+            })
+
+    with pytest.raises(ValueError, match="already registered"):
+        @register_derived
+        class _CoverageDerivedDuplicate(BaseDerivedData):
+            name = "_coverage_derived_registry"
+
+            def compute(self, daily=None, minute=None):
+                return pd.DataFrame()
+
+    with pytest.raises(KeyError, match="Unknown derived data"):
+        get_derived("does_not_exist")
+
+    missing_path = tmp_path / "missing_derived.py"
+    with pytest.raises(FileNotFoundError):
+        load_derived_module(str(missing_path))
+
+    bad_path = tmp_path / "bad_derived.py"
+    bad_path.write_text("x = 1\n")
+    monkeypatch.setattr(
+        "pipeline.derived.registry.importlib.util.spec_from_file_location",
+        lambda *args, **kwargs: None,
+    )
+    with pytest.raises(ImportError, match="Cannot load derived data module"):
+        load_derived_module(str(bad_path))
+
+    load_derived_module("math")
+    instance = _CoverageDerived()
+    instance.scale = 2
+    assert repr(instance) == "_CoverageDerived(scale=2)"
+
+
+def test_derived_compute_cli_writes_builtin_minute_summary(tmp_path):
+    runner = CliRunner()
+    minute_path = tmp_path / "minute.pq"
+    out_dir = tmp_path / "derived"
+    _minute_bars().to_parquet(minute_path, index=False)
+
+    result = runner.invoke(cli, [
+        "derived", "compute",
+        "--minute-path", str(minute_path),
+        "--derived-type", "minute_daily_summary",
+        "--derived-name", "minute_summary",
+        "--output-dir", str(out_dir),
+    ])
+
+    assert result.exit_code == 0, result.output
+    written = pd.read_parquet(out_dir / "minute_summary.pq")
+    assert "minute_vwap" in written.columns
+
+
+def test_minute_summary_uses_time_sort_and_daily_without_close():
+    minute = _minute_bars().drop(columns=["datetime"]).sample(frac=1.0, random_state=0)
+    daily = _daily_bars()[["symbol_id", "date", "open"]]
+
+    result = compute_derived(
+        "minute_daily_summary",
+        daily=daily,
+        minute=minute,
+    )
+
+    by_symbol = result.set_index(["symbol_id", "date"])
+    assert np.isclose(
+        by_symbol.loc[("sh600000", pd.Timestamp("2024-01-02")), "minute_intraday_return"],
+        0.10,
+    )
+    assert "minute_vwap_deviation" in result.columns
+
+
+def test_alpha_feature_join_rejects_derived_column_collisions():
+    data = _daily_bars()
+    derived_a = data[["symbol_id", "date"]].copy()
+    derived_a["custom_value"] = 1.0
+    derived_b = data[["symbol_id", "date"]].copy()
+    derived_b["custom_value"] = 2.0
+
+    with pytest.raises(ValueError, match="conflict"):
+        join_feature_frames(data, [derived_a, derived_b])
+
+    close_collision = data[["symbol_id", "date"]].copy()
+    close_collision["close"] = 1.0
+    with pytest.raises(ValueError, match="conflict"):
+        join_feature_frames(data, [close_collision])
+
+
+def test_legacy_feature_cli_delegates_to_derived_registry(tmp_path):
+    runner = CliRunner()
+    minute_path = tmp_path / "minute.pq"
+    out_dir = tmp_path / "features"
+    _minute_bars().to_parquet(minute_path, index=False)
+
+    list_result = runner.invoke(cli, ["feature", "list"])
+    assert list_result.exit_code == 0, list_result.output
+    assert "minute_daily_summary" in list_result.output
+
+    compute_result = runner.invoke(cli, [
+        "feature", "compute",
+        "--minute-path", str(minute_path),
+        "--feature-type", "minute_daily_summary",
+        "--feature-name", "minute_summary",
+        "--output-dir", str(out_dir),
+    ])
+    assert compute_result.exit_code == 0, compute_result.output
+    assert (out_dir / "minute_summary.pq").exists()

tests/test_downloader.py

@@ -1,6 +1,16 @@
+import os
+
 import pytest
 from data.downloader import download_daily
 
+pytestmark = [
+    pytest.mark.network,
+    pytest.mark.skipif(
+        os.environ.get("CEQ_RUN_LIVE_DOWNLOADER") != "1",
+        reason="set CEQ_RUN_LIVE_DOWNLOADER=1 to run live baostock/akshare smoke tests",
+    ),
+]
+
 
 def test_download_single_stock():
     """Smoke test: download data for 浦发银行 for a short window."""

tests/test_downloader_contracts.py

@@ -0,0 +1,631 @@
+"""Offline downloader contract tests with mocked data providers."""
+
+from __future__ import annotations
+
+import numpy as np
+import pandas as pd
+import pytest
+
+import data.downloader as downloader
+import pipeline.data.downloader as pipeline_downloader
+from data.downloader import download_daily, download_daily_batch
+from pipeline.common.schema import DATA_COLUMNS
+from pipeline.data.downloader import download_universe
+
+
+class _FakeResult:
+    def __init__(self, rows, error_code="0", error_msg=""):
+        self.rows = rows
+        self.error_code = error_code
+        self.error_msg = error_msg
+        self._idx = -1
+
+    def next(self):
+        self._idx += 1
+        return self._idx < len(self.rows)
+
+    def get_row_data(self):
+        return self.rows[self._idx]
+
+
+def _daily_batch_row(
+    date: str = "2024-01-02",
+    open_: str = "10",
+    high: str = "11",
+    low: str = "9",
+    close: str = "10.5",
+    preclose: str = "10",
+    volume: str = "1000",
+    amount: str = "10500",
+) -> list[str]:
+    return [
+        date,
+        open_,
+        high,
+        low,
+        close,
+        preclose,
+        volume,
+        amount,
+        "1.23",
+        "5.0",
+        "1",
+        "0",
+        "8.0",
+        "1.1",
+        "2.2",
+        "3.3",
+    ]
+
+
+def test_download_daily_uses_baostock_before_akshare_in_auto(monkeypatch):
+    calls: list[str] = []
+    expected = pd.DataFrame({
+        "symbol": ["sh600000"],
+        "date": ["2024-01-02"],
+        "open": [10.0],
+        "high": [11.0],
+        "low": [9.0],
+        "close": [10.5],
+        "volume": [1000.0],
+        "amount": [10500.0],
+    })
+
+    def fake_baostock(symbol, start, end, adjust):
+        calls.append("baostock")
+        return expected
+
+    def fake_akshare(symbol, start, end, adjust):
+        calls.append("akshare")
+        raise AssertionError("akshare should not be called after baostock succeeds")
+
+    monkeypatch.setattr(downloader, "_download_baostock", fake_baostock)
+    monkeypatch.setattr(downloader, "_download_akshare", fake_akshare)
+
+    result = download_daily("sh600000", "2024-01-02", "2024-01-02", source="auto")
+
+    assert calls == ["baostock"]
+    assert result["date"].tolist() == [pd.Timestamp("2024-01-02")]
+    assert result["close"].tolist() == [10.5]
+
+
+def test_download_daily_falls_back_to_akshare_when_baostock_empty(monkeypatch):
+    calls: list[str] = []
+    fallback = pd.DataFrame({
+        "symbol": ["sz000001"],
+        "date": ["2024-01-02"],
+        "open": [20.0],
+        "high": [21.0],
+        "low": [19.0],
+        "close": [20.5],
+        "volume": [2000.0],
+        "amount": [41000.0],
+    })
+
+    monkeypatch.setattr(
+        downloader,
+        "_download_baostock",
+        lambda symbol, start, end, adjust: calls.append("baostock") or None,
+    )
+    monkeypatch.setattr(
+        downloader,
+        "_download_akshare",
+        lambda symbol, start, end, adjust: calls.append("akshare") or fallback,
+    )
+
+    result = download_daily("sz000001", "2024-01-02", "2024-01-02", source="auto")
+
+    assert calls == ["baostock", "akshare"]
+    assert result["symbol"].tolist() == ["sz000001"]
+    assert result["date"].tolist() == [pd.Timestamp("2024-01-02")]
+
+
+def test_download_daily_raises_when_requested_source_has_no_data(monkeypatch):
+    monkeypatch.setattr(downloader, "_download_baostock", lambda *args: None)
+
+    with pytest.raises(RuntimeError, match="Failed to download data for sh600000"):
+        download_daily(
+            "sh600000",
+            "2024-01-02",
+            "2024-01-02",
+            source="baostock",
+        )
+
+
+def test_download_daily_akshare_source_skips_baostock(monkeypatch):
+    calls: list[str] = []
+    fallback = pd.DataFrame({
+        "symbol": ["sh600000"],
+        "date": ["2024-01-02"],
+        "open": [10.0],
+        "high": [11.0],
+        "low": [9.0],
+        "close": [10.5],
+        "volume": [1000.0],
+        "amount": [10500.0],
+    })
+
+    monkeypatch.setattr(
+        downloader,
+        "_download_baostock",
+        lambda *args: calls.append("baostock") or None,
+    )
+    monkeypatch.setattr(
+        downloader,
+        "_download_akshare",
+        lambda *args: calls.append("akshare") or fallback,
+    )
+
+    result = download_daily(
+        "sh600000",
+        "2024-01-02",
+        "2024-01-02",
+        source="akshare",
+    )
+
+    assert calls == ["akshare"]
+    assert result["date"].tolist() == [pd.Timestamp("2024-01-02")]
+
+
+def test_akshare_daily_downloader_maps_columns_and_failures(monkeypatch):
+    calls: list[dict] = []
+    raw = pd.DataFrame({
+        "日期": ["2024-01-02"],
+        "开盘": [10.0],
+        "最高": [11.0],
+        "最低": [9.0],
+        "收盘": [10.5],
+        "成交量": [1000.0],
+        "成交额": [10500.0],
+        "换手率": [1.2],
+    })
+
+    def fake_hist(**kwargs):
+        calls.append(kwargs)
+        return raw.copy()
+
+    monkeypatch.setattr(downloader.ak, "stock_zh_a_hist", fake_hist)
+
+    result = downloader._download_akshare(
+        "sh600000",
+        "20240102",
+        "20240102",
+        adjust="",
+    )
+
+    assert calls == [{
+        "symbol": "600000",
+        "period": "daily",
+        "start_date": "20240102",
+        "end_date": "20240102",
+        "adjust": "",
+    }]
+    assert result is not None
+    assert result.columns.tolist() == [
+        "symbol", "date", "open", "high", "low", "close", "volume", "amount",
+    ]
+    assert result["symbol"].tolist() == ["sh600000"]
+
+    monkeypatch.setattr(downloader.ak, "stock_zh_a_hist", lambda **kwargs: pd.DataFrame())
+    assert downloader._download_akshare("sh600000", "20240102", "20240102") is None
+
+    monkeypatch.setattr(
+        downloader.ak,
+        "stock_zh_a_hist",
+        lambda **kwargs: (_ for _ in ()).throw(RuntimeError("provider down")),
+    )
+    assert downloader._download_akshare("sh600000", "20240102", "20240102") is None
+
+
+def test_baostock_daily_downloader_maps_errors_and_logout_failures(monkeypatch):
+    query_calls: list[dict] = []
+    row = ["2024-01-02", "10", "11", "9", "10.5", "1000", "10500"]
+
+    monkeypatch.setattr(downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(
+        downloader.bs,
+        "logout",
+        lambda: (_ for _ in ()).throw(RuntimeError("logout failed")),
+    )
+
+    def fake_query(**kwargs):
+        query_calls.append(kwargs)
+        return _FakeResult([row])
+
+    monkeypatch.setattr(downloader.bs, "query_history_k_data_plus", fake_query)
+
+    result = downloader._download_baostock(
+        "sz000001",
+        "2024-01-02",
+        "2024-01-02",
+        adjust="none",
+    )
+
+    assert query_calls[0]["code"] == "sz.000001"
+    assert query_calls[0]["adjustflag"] == "3"
+    assert result is not None
+    assert result["symbol"].tolist() == ["sz000001"]
+    assert pd.api.types.is_numeric_dtype(result["close"])
+
+    monkeypatch.setattr(downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: _FakeResult([], error_code="1", error_msg="bad symbol"),
+    )
+    assert downloader._download_baostock("sz000001", "2024-01-02", "2024-01-02") is None
+
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: _FakeResult([]),
+    )
+    assert downloader._download_baostock("sz000001", "2024-01-02", "2024-01-02") is None
+
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: (_ for _ in ()).throw(RuntimeError("query failed")),
+    )
+    assert downloader._download_baostock("sz000001", "2024-01-02", "2024-01-02") is None
+
+
+def test_download_daily_batch_maps_rich_schema_and_vwap(monkeypatch):
+    query_calls: list[dict] = []
+    login_count = 0
+    logout_count = 0
+
+    def fake_login():
+        nonlocal login_count
+        login_count += 1
+
+    def fake_logout():
+        nonlocal logout_count
+        logout_count += 1
+
+    def fake_query(**kwargs):
+        query_calls.append(kwargs)
+        rows = [
+            _daily_batch_row(volume="1000", amount="10500"),
+            _daily_batch_row(date="2024-01-03", volume="0", amount="0"),
+        ]
+        return _FakeResult(rows)
+
+    monkeypatch.setattr(downloader.bs, "login", fake_login)
+    monkeypatch.setattr(downloader.bs, "logout", fake_logout)
+    monkeypatch.setattr(downloader.bs, "query_history_k_data_plus", fake_query)
+
+    [(symbol, frame)] = list(
+        download_daily_batch(
+            ["sh600000"],
+            "2024-01-02",
+            "2024-01-03",
+            adjust="hfq",
+        )
+    )
+
+    assert symbol == "sh600000"
+    assert query_calls[0]["code"] == "sh.600000"
+    assert query_calls[0]["adjustflag"] == "1"
+    assert login_count == 1
+    assert logout_count == 1
+    assert frame is not None
+    assert frame.columns.tolist() == [
+        "symbol", "date", "open", "high", "low", "close", "preclose",
+        "volume", "amount", "vwap", "turn", "pctChg", "tradestatus", "isST",
+        "peTTM", "pbMRQ", "psTTM", "pcfNcfTTM",
+    ]
+    assert np.isclose(frame["vwap"].iloc[0], 10.5)
+    assert pd.isna(frame["vwap"].iloc[1])
+    assert pd.api.types.is_datetime64_any_dtype(frame["date"])
+    assert pd.api.types.is_numeric_dtype(frame["tradestatus"])
+
+
+def test_download_daily_batch_periodic_relogin_and_none_result(monkeypatch):
+    responses = [
+        _FakeResult([], error_code="1", error_msg="bad symbol"),
+        _FakeResult([_daily_batch_row(date="2024-01-03")]),
+    ]
+    login_count = 0
+    logout_count = 0
+
+    def fake_login():
+        nonlocal login_count
+        login_count += 1
+
+    def fake_logout():
+        nonlocal logout_count
+        logout_count += 1
+
+    monkeypatch.setattr(downloader.bs, "login", fake_login)
+    monkeypatch.setattr(downloader.bs, "logout", fake_logout)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: responses.pop(0),
+    )
+
+    results = list(
+        download_daily_batch(
+            ["sh600000", "sz000001"],
+            "2024-01-02",
+            "2024-01-03",
+            relogin_every=1,
+        )
+    )
+
+    assert results[0] == ("sh600000", None)
+    assert results[1][0] == "sz000001"
+    assert results[1][1] is not None
+    assert login_count == 2
+    assert logout_count == 2
+
+
+def test_download_daily_batch_empty_rows_yields_none(monkeypatch):
+    monkeypatch.setattr(downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: _FakeResult([]),
+    )
+
+    assert list(download_daily_batch(["sh600000"], "2024-01-02", "2024-01-02")) == [
+        ("sh600000", None)
+    ]
+
+
+def test_download_daily_batch_generic_exception_yields_none(monkeypatch):
+    monkeypatch.setattr(downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: (_ for _ in ()).throw(RuntimeError("query failed")),
+    )
+
+    assert list(download_daily_batch(["sh600000"], "2024-01-02", "2024-01-02")) == [
+        ("sh600000", None)
+    ]
+
+
+def test_download_daily_batch_relogs_and_retries_session_loss(monkeypatch):
+    responses = [
+        _FakeResult([], error_code="10002007", error_msg="用户未登录"),
+        _FakeResult([_daily_batch_row()]),
+    ]
+    login_count = 0
+    logout_count = 0
+
+    def fake_login():
+        nonlocal login_count
+        login_count += 1
+
+    def fake_logout():
+        nonlocal logout_count
+        logout_count += 1
+
+    monkeypatch.setattr(downloader.bs, "login", fake_login)
+    monkeypatch.setattr(downloader.bs, "logout", fake_logout)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: responses.pop(0),
+    )
+
+    [(symbol, frame)] = list(download_daily_batch(["sh600000"], "2024-01-02", "2024-01-02"))
+
+    assert symbol == "sh600000"
+    assert frame is not None
+    assert len(frame) == 1
+    assert login_count == 2
+    assert logout_count == 2
+
+
+def test_download_daily_batch_second_session_loss_and_logout_failure(monkeypatch):
+    responses = [
+        _FakeResult([], error_code="10002007", error_msg="用户未登录"),
+        _FakeResult([], error_code="10002007", error_msg="用户未登录"),
+    ]
+    logout_count = 0
+
+    def fake_logout():
+        nonlocal logout_count
+        logout_count += 1
+        raise RuntimeError("logout failed")
+
+    monkeypatch.setattr(downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(downloader.bs, "logout", fake_logout)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: responses.pop(0),
+    )
+
+    assert list(download_daily_batch(["sh600000"], "2024-01-02", "2024-01-02")) == [
+        ("sh600000", None)
+    ]
+    assert logout_count == 2
+
+
+def test_download_daily_batch_uses_akshare_fallback_when_enabled(monkeypatch):
+    fallback = pd.DataFrame({
+        "symbol": ["sh600000"],
+        "date": ["2024-01-02"],
+        "open": [10.0],
+        "high": [11.0],
+        "low": [9.0],
+        "close": [10.5],
+        "volume": [1000.0],
+        "amount": [10500.0],
+    })
+
+    monkeypatch.setattr(downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: _FakeResult([], error_code="1", error_msg="no data"),
+    )
+    monkeypatch.setattr(
+        downloader,
+        "_download_akshare",
+        lambda symbol, start, end, adjust: fallback.copy(),
+    )
+
+    [(symbol, frame)] = list(
+        download_daily_batch(
+            ["sh600000"],
+            "2024-01-02",
+            "2024-01-02",
+            akshare_fallback=True,
+        )
+    )
+
+    assert symbol == "sh600000"
+    assert frame is not None
+    assert frame["date"].tolist() == [pd.Timestamp("2024-01-02")]
+    assert frame["close"].tolist() == [10.5]
+
+
+def test_download_universe_writes_daily_partitions_from_mock_batch(tmp_path, monkeypatch):
+    batch_frame = pd.DataFrame({
+        "symbol": ["sh600000", "sh600000"],
+        "date": pd.to_datetime(["2024-01-02", "2024-02-01"]),
+        "open": [10.0, 11.0],
+        "high": [11.0, 12.0],
+        "low": [9.0, 10.0],
+        "close": [10.5, 11.5],
+        "preclose": [10.0, 10.5],
+        "volume": [1000.0, 1200.0],
+        "amount": [10500.0, 13800.0],
+        "vwap": [10.5, 11.5],
+        "turn": [1.0, 1.1],
+        "pctChg": [5.0, 9.5],
+        "tradestatus": [1, 1],
+        "isST": [0, 0],
+        "peTTM": [8.0, 8.1],
+        "pbMRQ": [1.1, 1.2],
+        "psTTM": [2.1, 2.2],
+        "pcfNcfTTM": [3.1, 3.2],
+    })
+
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "_resolve_universe",
+        lambda universe, max_symbols=0: pd.DataFrame({
+            "symbol_id": ["sh600000", "sz000001"],
+            "symbol_name": ["PF Bank", "Ping An Bank"],
+        }),
+    )
+
+    def fake_batch(symbols, start, end, adjust="qfq"):
+        assert symbols == ["sh600000", "sz000001"]
+        assert adjust == "qfq"
+        yield "sh600000", batch_frame
+        yield "sz000001", None
+
+    monkeypatch.setattr(pipeline_downloader, "download_daily_batch", fake_batch)
+
+    stale_file = tmp_path / "toy" / "month=2024-01" / "stale.pq"
+    stale_file.parent.mkdir(parents=True)
+    batch_frame.iloc[[0]][DATA_COLUMNS[2:]].to_parquet(stale_file, index=False)
+
+    stats = download_universe(
+        universe="toy",
+        start_date="2024-01-02",
+        end_date="2024-02-01",
+        output_dir=str(tmp_path),
+        chunk_size=1,
+    )
+
+    dataset_path = tmp_path / "toy"
+    written = pd.read_parquet(dataset_path).sort_values(["date", "symbol_id"]).reset_index(drop=True)
+    assert stats == {
+        "dataset_path": str(dataset_path),
+        "n_symbols": 1,
+        "n_requested": 2,
+        "n_rows": 2,
+        "date_min": "2024-01-02",
+        "date_max": "2024-02-01",
+    }
+    assert (dataset_path / "month=2024-01").exists()
+    assert (dataset_path / "month=2024-02").exists()
+    assert not stale_file.exists()
+    assert written[DATA_COLUMNS].columns.tolist() == DATA_COLUMNS
+    assert written["symbol_name"].tolist() == ["PF Bank", "PF Bank"]
+
+
+def test_download_universe_raises_when_all_daily_symbols_empty(tmp_path, monkeypatch):
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "_resolve_universe",
+        lambda universe, max_symbols=0: pd.DataFrame({
+            "symbol_id": ["sh600000"],
+            "symbol_name": ["PF Bank"],
+        }),
+    )
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "download_daily_batch",
+        lambda symbols, start, end, adjust="qfq": iter([("sh600000", None)]),
+    )
+
+    with pytest.raises(RuntimeError, match="No data downloaded"):
+        download_universe(
+            universe="toy",
+            start_date="2024-01-02",
+            end_date="2024-01-02",
+            output_dir=str(tmp_path),
+        )
+
+
+def test_download_universe_progress_branch_at_100_symbols(tmp_path, monkeypatch):
+    symbols = [f"sh6{i:05d}" for i in range(100)]
+    batch_frame = pd.DataFrame({
+        "symbol": ["sh600000"],
+        "date": [pd.Timestamp("2024-01-02")],
+        "open": [10.0],
+        "high": [11.0],
+        "low": [9.0],
+        "close": [10.5],
+        "preclose": [10.0],
+        "volume": [1000.0],
+        "amount": [10500.0],
+        "vwap": [10.5],
+        "turn": [1.0],
+        "pctChg": [5.0],
+        "tradestatus": [1],
+        "isST": [0],
+        "peTTM": [8.0],
+        "pbMRQ": [1.1],
+        "psTTM": [2.1],
+        "pcfNcfTTM": [3.1],
+    })
+
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "_resolve_universe",
+        lambda universe, max_symbols=0: pd.DataFrame({
+            "symbol_id": symbols,
+            "symbol_name": symbols,
+        }),
+    )
+
+    def fake_batch(requested_symbols, start, end, adjust="qfq"):
+        assert requested_symbols == symbols
+        for symbol in requested_symbols:
+            yield symbol, batch_frame.copy()
+
+    monkeypatch.setattr(pipeline_downloader, "download_daily_batch", fake_batch)
+
+    stats = download_universe(
+        universe="toy100",
+        start_date="2024-01-02",
+        end_date="2024-01-02",
+        output_dir=str(tmp_path),
+        chunk_size=200,
+    )
+
+    assert stats["n_symbols"] == 100
+    assert stats["n_rows"] == 100

tests/test_features.py

@@ -0,0 +1,188 @@
+"""Tests for minute-derived daily feature plugins."""
+
+import textwrap
+
+import numpy as np
+import pandas as pd
+import pytest
+
+from pipeline.features.compute import compute_feature, validate_feature_frame
+from pipeline.features.compute import read_feature_frames
+from pipeline.features.library.minute_daily_summary import MinuteDailySummaryFeature
+from pipeline.features.registry import (
+    available_features,
+    get_feature,
+    load_feature_module,
+)
+from pipeline.derived.compute import compute_derived
+
+
+def _minute_bars() -> pd.DataFrame:
+    return pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000", "sz000001"],
+        "symbol_name": ["PF Bank", "PF Bank", "Ping An"],
+        "datetime": pd.to_datetime([
+            "2024-01-02 09:35:00",
+            "2024-01-02 09:40:00",
+            "2024-01-02 09:35:00",
+        ]),
+        "date": pd.to_datetime(["2024-01-02", "2024-01-02", "2024-01-02"]),
+        "time": ["09:35:00", "09:40:00", "09:35:00"],
+        "frequency": ["5m", "5m", "5m"],
+        "open": [10.0, 10.5, 20.0],
+        "high": [11.0, 12.0, 21.0],
+        "low": [9.0, 10.0, 19.0],
+        "close": [10.5, 11.0, 20.5],
+        "volume": [100.0, 300.0, 200.0],
+        "amount": [1000.0, 3300.0, 4100.0],
+        "vwap": [10.0, 11.0, 20.5],
+        "adjustflag": ["3", "3", "3"],
+    })
+
+
+def test_built_in_minute_daily_summary():
+    daily = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000"],
+        "date": pd.to_datetime(["2024-01-02", "2024-01-02", "2024-01-03"]),
+        "close": [11.0, 20.5, 12.0],
+    })
+
+    result = compute_feature(
+        minute=_minute_bars(),
+        daily=daily,
+        feature_type="minute_daily_summary",
+    )
+
+    assert "minute_daily_summary" in available_features()
+    row = result[
+        (result["symbol_id"] == "sh600000")
+        & (result["date"] == pd.Timestamp("2024-01-02"))
+    ].iloc[0]
+    assert row["minute_bar_count"] == 2
+    assert np.isclose(row["minute_intraday_return"], 11.0 / 10.0 - 1.0)
+    assert np.isclose(row["minute_intraday_range"], 12.0 / 9.0 - 1.0)
+    assert np.isclose(row["minute_vwap"], 4300.0 / 400.0)
+    assert np.isclose(row["minute_vwap_deviation"], (4300.0 / 400.0) / 11.0 - 1.0)
+
+    missing = result[
+        (result["symbol_id"] == "sh600000")
+        & (result["date"] == pd.Timestamp("2024-01-03"))
+    ].iloc[0]
+    assert pd.isna(missing["minute_vwap"])
+
+
+def test_minute_daily_summary_feature_preserves_legacy_positional_compute():
+    direct = MinuteDailySummaryFeature().compute(_minute_bars())
+    via_registry = get_feature("minute_daily_summary").compute(_minute_bars())
+
+    assert "minute_vwap" in direct.columns
+    pd.testing.assert_frame_equal(direct, via_registry)
+
+
+def test_legacy_feature_compute_matches_canonical_derived_compute():
+    daily = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000"],
+        "date": pd.to_datetime(["2024-01-02", "2024-01-02", "2024-01-03"]),
+        "close": [11.0, 20.5, 12.0],
+    })
+
+    legacy_feature = compute_feature(
+        minute=_minute_bars(),
+        daily=daily,
+        feature_type="minute_daily_summary",
+    )
+    canonical_derived = compute_derived(
+        "minute_daily_summary",
+        daily=daily,
+        minute=_minute_bars(),
+    )
+
+    pd.testing.assert_frame_equal(legacy_feature, canonical_derived)
+
+
+def test_read_feature_frames_delegates_to_derived_validation(tmp_path):
+    feature = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": ["2024-01-02 15:00:00"],
+        "toy_feature": [1.5],
+    })
+    feature_path = tmp_path / "feature.pq"
+    feature.to_parquet(feature_path, index=False)
+
+    [result] = read_feature_frames([feature_path])
+
+    assert result["date"].tolist() == [pd.Timestamp("2024-01-02")]
+    assert result["toy_feature"].tolist() == [1.5]
+
+
+def test_load_external_feature_module_and_filter_params(tmp_path):
+    module_path = tmp_path / "external_feature.py"
+    module_path.write_text(textwrap.dedent('''
+        import pandas as pd
+        from pipeline.features.base import BaseFeature
+        from pipeline.features.registry import register_feature
+
+        @register_feature
+        class ExternalVolumeFeature(BaseFeature):
+            name = "external_volume_feature"
+
+            def __init__(self, scale: float = 1.0):
+                self.scale = scale
+
+            def compute(self, minute: pd.DataFrame, daily=None) -> pd.DataFrame:
+                out = (
+                    minute.groupby(["symbol_id", "date"], as_index=False)["volume"]
+                    .sum()
+                    .rename(columns={"volume": "scaled_volume"})
+                )
+                out["scaled_volume"] *= self.scale
+                return out
+    '''))
+
+    load_feature_module(str(module_path))
+    assert "external_volume_feature" in available_features()
+
+    instance = get_feature("external_volume_feature", scale=2.0, ignored=99)
+    assert instance.scale == 2.0
+    assert not hasattr(instance, "ignored")
+
+    result = compute_feature(
+        minute=_minute_bars(),
+        feature_type="external_volume_feature",
+        scale=2.0,
+        ignored=99,
+    )
+    row = result[result["symbol_id"] == "sh600000"].iloc[0]
+    assert np.isclose(row["scaled_volume"], 800.0)
+
+
+def test_validate_feature_frame_rejects_missing_keys():
+    with pytest.raises(ValueError, match="missing required"):
+        validate_feature_frame(pd.DataFrame({"symbol_id": ["sh600000"], "x": [1.0]}))
+
+
+def test_validate_feature_frame_rejects_duplicate_keys_after_date_normalization():
+    with pytest.raises(ValueError, match="duplicate symbol_id,date"):
+        validate_feature_frame(pd.DataFrame({
+            "symbol_id": ["sh600000", "sh600000"],
+            "date": ["2024-01-02", pd.Timestamp("2024-01-02")],
+            "x": [1.0, 2.0],
+        }))
+
+
+def test_validate_feature_frame_rejects_duplicate_columns():
+    bad = pd.DataFrame(
+        [["sh600000", pd.Timestamp("2024-01-02"), 1.0, 2.0]],
+        columns=["symbol_id", "date", "dup", "dup"],
+    )
+    with pytest.raises(ValueError, match="duplicate columns"):
+        validate_feature_frame(bad)
+
+
+def test_validate_feature_frame_rejects_non_numeric_feature_columns():
+    with pytest.raises(ValueError, match="numeric"):
+        validate_feature_frame(pd.DataFrame({
+            "symbol_id": ["sh600000"],
+            "date": [pd.Timestamp("2024-01-02")],
+            "bad": ["not numeric"],
+        }))

tests/test_joinquant_plugin.py

@@ -0,0 +1,650 @@
+"""Tests for the JoinQuant comparison plugin (network-free)."""
+
+from __future__ import annotations
+
+import hashlib
+import json
+from pathlib import Path
+
+import pandas as pd
+import pytest
+from click.testing import CliRunner
+
+from cli import cli
+from pipeline.common.schema import FILL_COLUMNS, PNL_COLUMNS, POSITION_COLUMNS
+from plugins.joinquant.browser import (
+    default_browser_config,
+    load_env_file,
+    resolve_template,
+    write_browser_config_template,
+)
+from plugins.joinquant.export_targets import export_targets
+from plugins.joinquant.ingest import (
+    ingest_joinquant_outputs,
+    normalize_fills_csv,
+)
+from plugins.joinquant.reconcile import reconcile_joinquant
+from plugins.joinquant.schema import (
+    JOINQUANT_FILL_COLUMNS,
+    JOINQUANT_PNL_COLUMNS,
+    JOINQUANT_POSITION_COLUMNS,
+    JOINQUANT_TARGET_COLUMNS,
+    RECONCILE_COLUMNS,
+)
+from plugins.joinquant.smoke import build_fixed_share_positions
+from plugins.joinquant.symbols import from_joinquant_symbol, to_joinquant_symbol
+from plugins.joinquant.wrapper_strategy import write_wrapper_strategy
+
+
+def _positions(
+    *,
+    symbol: str = "sh600000",
+    date: str = "2026-07-01",
+    shares: int = 1000,
+    price: float = 10.0,
+    portfolio_name: str = "run1",
+) -> pd.DataFrame:
+    target_value = float(shares * price)
+    weight = target_value / 1_000_000.0
+    return pd.DataFrame([{
+        "symbol_id": symbol,
+        "date": pd.Timestamp(date),
+        "portfolio_name": portfolio_name,
+        "target_weight": weight,
+        "target_value": target_value,
+        "target_shares": float(shares) + 0.25,
+        "position_shares": shares,
+        "position_value": target_value,
+        "price": price,
+    }], columns=POSITION_COLUMNS)
+
+
+def _our_fills(
+    *,
+    symbol: str = "sh600000",
+    date: str = "2026-07-01",
+    shares: int = 1000,
+    price: float = 10.0,
+    cost: float = 5.0,
+    portfolio_name: str = "run1",
+) -> pd.DataFrame:
+    fills = pd.DataFrame([{
+        "symbol_id": symbol,
+        "date": pd.Timestamp(date),
+        "portfolio_name": portfolio_name,
+        "prev_shares": 0,
+        "target_shares": shares,
+        "traded_shares": shares,
+        "realized_shares": shares,
+        "blocked": 0,
+        "trade_cost": cost,
+        "trade_price": price,
+    }])
+    return fills
+
+
+def _our_pnl(
+    *,
+    date: str = "2026-07-01",
+    pnl: float = 100.0,
+    cost: float = 5.0,
+    portfolio_name: str = "run1",
+) -> pd.DataFrame:
+    return pd.DataFrame([{
+        "date": pd.Timestamp(date),
+        "portfolio_name": portfolio_name,
+        "gross_exposure": 10_000.0,
+        "net_exposure": 10_000.0,
+        "pnl": pnl,
+        "cost": cost,
+        "turnover": 1.0,
+        "n_positions": 1,
+    }], columns=PNL_COLUMNS)
+
+
+def _jq_fills(
+    *,
+    symbol: str = "sh600000",
+    date: str = "2026-07-01",
+    shares: int = 1000,
+    price: float = 10.0,
+    cost: float = 5.0,
+    portfolio_name: str = "run1",
+    raw_status: str = "filled",
+) -> pd.DataFrame:
+    return pd.DataFrame([{
+        "date": date,
+        "portfolio_name": portfolio_name,
+        "symbol_id": symbol,
+        "jq_symbol": to_joinquant_symbol(symbol),
+        "order_id": "ord-1",
+        "side": "buy" if shares >= 0 else "sell",
+        "requested_shares": shares,
+        "filled_shares": shares,
+        "fill_price": price,
+        "trade_value": abs(shares * price),
+        "trade_cost": cost,
+        "blocked": 0,
+        "raw_status": raw_status,
+    }], columns=JOINQUANT_FILL_COLUMNS)
+
+
+def _jq_positions(
+    *,
+    symbol: str = "sh600000",
+    date: str = "2026-07-01",
+    shares: int = 1000,
+    price: float = 10.0,
+    portfolio_name: str = "run1",
+) -> pd.DataFrame:
+    return pd.DataFrame([{
+        "date": date,
+        "portfolio_name": portfolio_name,
+        "symbol_id": symbol,
+        "jq_symbol": to_joinquant_symbol(symbol),
+        "position_shares": shares,
+        "position_value": shares * price,
+        "cash": 990_000.0,
+        "total_value": 1_000_000.0,
+    }], columns=JOINQUANT_POSITION_COLUMNS)
+
+
+def _jq_pnl(
+    *,
+    date: str = "2026-07-01",
+    pnl: float = 100.0,
+    cost: float = 5.0,
+    portfolio_name: str = "run1",
+) -> pd.DataFrame:
+    return pd.DataFrame([{
+        "date": date,
+        "portfolio_name": portfolio_name,
+        "gross_exposure": 10_000.0,
+        "net_exposure": 10_000.0,
+        "cash": 990_000.0,
+        "total_value": 1_000_000.0,
+        "pnl": pnl,
+        "cost": cost,
+        "turnover": 1.0,
+    }], columns=JOINQUANT_PNL_COLUMNS)
+
+
+def _write_parquets(tmp_path: Path, frames: dict[str, pd.DataFrame]) -> dict[str, Path]:
+    paths = {}
+    for name, frame in frames.items():
+        path = tmp_path / f"{name}.pq"
+        frame.to_parquet(path, index=False)
+        paths[name] = path
+    return paths
+
+
+def _export_targets_for(tmp_path: Path, positions: pd.DataFrame) -> tuple[Path, Path]:
+    positions_path = tmp_path / "positions.pq"
+    positions.to_parquet(positions_path, index=False)
+    targets_root = tmp_path / "targets"
+    export_targets(
+        positions_path,
+        portfolio_name="run1",
+        out_dir=targets_root,
+        mode="target_shares",
+    )
+    return positions_path, targets_root / "run1"
+
+
+@pytest.mark.parametrize(
+    ("internal", "joinquant"),
+    [
+        ("sh600000", "600000.XSHG"),
+        ("sh688001", "688001.XSHG"),
+        ("sz000001", "000001.XSHE"),
+        ("sz001001", "001001.XSHE"),
+        ("sz002594", "002594.XSHE"),
+        ("sz300001", "300001.XSHE"),
+    ],
+)
+def test_symbol_mapping_both_directions(internal, joinquant):
+    assert to_joinquant_symbol(internal) == joinquant
+    assert from_joinquant_symbol(joinquant) == internal
+
+
+@pytest.mark.parametrize("bad", ["600000", "bj830000", "sh000001", "sz600000", "abc"])
+def test_symbol_mapping_rejects_invalid_symbols(bad):
+    with pytest.raises(ValueError):
+        to_joinquant_symbol(bad)
+
+
+@pytest.mark.parametrize("bad", ["600000", "600000.XSHE", "000001.XSHG", "abc.XSHG"])
+def test_reverse_symbol_mapping_rejects_invalid_symbols(bad):
+    with pytest.raises(ValueError):
+        from_joinquant_symbol(bad)
+
+
+def test_export_targets_schema_snapshot_hash_and_no_overwrite(tmp_path):
+    positions_path = tmp_path / "positions.pq"
+    _positions().to_parquet(positions_path, index=False)
+
+    snapshots = export_targets(
+        positions_path,
+        portfolio_name="run1",
+        out_dir=tmp_path / "targets",
+        mode="target_shares",
+    )
+
+    csv_path = tmp_path / "targets" / "run1" / "20260701.csv"
+    parquet_path = tmp_path / "targets" / "run1" / "20260701.parquet"
+    snapshot_path = tmp_path / "snapshots" / "run1" / "20260701.json"
+    assert csv_path.exists()
+    assert parquet_path.exists()
+    assert snapshot_path.exists()
+
+    target = pd.read_csv(csv_path)
+    assert list(target.columns) == JOINQUANT_TARGET_COLUMNS
+    assert int(target.loc[0, "target_shares"]) == 1000
+    assert float(target.loc[0, "target_value"]) == 10_000.0
+    assert target.loc[0, "export_mode"] == "target_shares"
+
+    snapshot = json.loads(snapshot_path.read_text())
+    actual_hash = hashlib.sha256(csv_path.read_bytes()).hexdigest()
+    assert snapshots[0]["file_sha256"] == actual_hash
+    assert snapshot["file_sha256"] == actual_hash
+    assert snapshot["n_symbols"] == 1
+
+    with pytest.raises(FileExistsError):
+        export_targets(
+            positions_path,
+            portfolio_name="run1",
+            out_dir=tmp_path / "targets",
+            mode="target_shares",
+        )
+
+
+def test_export_targets_target_value_mode_from_position_columns(tmp_path):
+    positions_path = tmp_path / "positions.pq"
+    _positions(shares=250, price=20.0).to_parquet(positions_path, index=False)
+
+    export_targets(
+        positions_path,
+        portfolio_name="run1",
+        out_dir=tmp_path / "targets_value",
+        mode="target_value",
+    )
+
+    target = pd.read_parquet(tmp_path / "targets_value" / "run1" / "20260701.parquet")
+    assert list(target.columns) == JOINQUANT_TARGET_COLUMNS
+    assert target.loc[0, "export_mode"] == "target_value"
+    assert target.loc[0, "target_value"] == 5_000.0
+    assert target.loc[0, "target_shares"] == 250
+
+
+def test_export_targets_can_shift_to_next_execution_session(tmp_path):
+    positions_path = tmp_path / "positions.pq"
+    _positions(date="2024-01-09").to_parquet(positions_path, index=False)
+    calendar_path = tmp_path / "daily.pq"
+    pd.DataFrame({
+        "date": pd.to_datetime(["2024-01-09", "2024-01-10", "2024-01-11"]),
+        "symbol_id": ["sh600000", "sh600000", "sh600000"],
+    }).to_parquet(calendar_path, index=False)
+
+    snapshots = export_targets(
+        positions_path,
+        portfolio_name="run1",
+        out_dir=tmp_path / "targets_shifted",
+        mode="target_shares",
+        start_date="2024-01-10",
+        end_date="2024-01-10",
+        execution_calendar_path=calendar_path,
+    )
+
+    assert len(snapshots) == 1
+    assert snapshots[0]["date"] == "2024-01-10"
+    assert (tmp_path / "targets_shifted" / "run1" / "20240110.csv").exists()
+    target = pd.read_csv(tmp_path / "targets_shifted" / "run1" / "20240110.csv")
+    assert target.loc[0, "date"] == "2024-01-10"
+
+
+def test_ingest_permissive_csv_column_mapping_and_output_schemas(tmp_path):
+    fills_csv = tmp_path / "jq_fills.csv"
+    positions_csv = tmp_path / "jq_positions.csv"
+    pnl_csv = tmp_path / "jq_pnl.csv"
+    pd.DataFrame([{
+        "Trade Date": "2026-07-01 09:31:00",
+        "Security": "600000.XSHG",
+        "Direction": "buy",
+        "Order Amount": 1000,
+        "Filled Amount": 1000,
+        "Price": 10.0,
+        "Status": "filled",
+    }]).to_csv(fills_csv, index=False)
+    pd.DataFrame([{
+        "Date": "2026-07-01",
+        "Security": "600000.XSHG",
+        "Shares": 1000,
+        "Market Value": 10_000.0,
+        "Cash": 990_000.0,
+        "Portfolio Value": 1_000_000.0,
+    }]).to_csv(positions_csv, index=False)
+    pd.DataFrame([{
+        "Date": "2026-07-01",
+        "Portfolio Value": 1_000_000.0,
+        "Daily PnL": 100.0,
+        "Turnover": 1.0,
+    }]).to_csv(pnl_csv, index=False)
+
+    fills = normalize_fills_csv(fills_csv, "run1")
+    assert list(fills.columns) == JOINQUANT_FILL_COLUMNS
+    assert fills.loc[0, "symbol_id"] == "sh600000"
+    assert fills.loc[0, "jq_symbol"] == "600000.XSHG"
+    assert fills.loc[0, "trade_cost"] == 0.0
+    assert fills.loc[0, "blocked"] == 0
+
+    paths = ingest_joinquant_outputs(
+        portfolio_name="run1",
+        fills_csv=fills_csv,
+        positions_csv=positions_csv,
+        pnl_csv=pnl_csv,
+        out_dir=tmp_path / "ingested",
+    )
+    assert list(pd.read_parquet(paths["fills"]).columns) == JOINQUANT_FILL_COLUMNS
+    assert list(pd.read_parquet(paths["positions"]).columns) == JOINQUANT_POSITION_COLUMNS
+    assert list(pd.read_parquet(paths["pnl"]).columns) == JOINQUANT_PNL_COLUMNS
+
+
+def _run_reconcile_case(
+    tmp_path: Path,
+    *,
+    positions: pd.DataFrame | None = None,
+    our_fills: pd.DataFrame | None = None,
+    jq_fills: pd.DataFrame | None = None,
+    jq_positions: pd.DataFrame | None = None,
+    our_pnl: pd.DataFrame | None = None,
+    jq_pnl: pd.DataFrame | None = None,
+) -> pd.DataFrame:
+    positions = _positions() if positions is None else positions
+    _, targets_dir = _export_targets_for(tmp_path, positions)
+    paths = _write_parquets(tmp_path, {
+        "our_fills": _our_fills() if our_fills is None else our_fills,
+        "our_positions": positions,
+        "our_pnl": _our_pnl() if our_pnl is None else our_pnl,
+        "jq_fills": _jq_fills() if jq_fills is None else jq_fills,
+        "jq_positions": _jq_positions() if jq_positions is None else jq_positions,
+        "jq_pnl": _jq_pnl() if jq_pnl is None else jq_pnl,
+    })
+    out_paths = reconcile_joinquant(
+        portfolio_name="run1",
+        targets_dir=targets_dir,
+        our_fills_path=paths["our_fills"],
+        our_positions_path=paths["our_positions"],
+        our_pnl_path=paths["our_pnl"],
+        jq_fills_path=paths["jq_fills"],
+        jq_positions_path=paths["jq_positions"],
+        jq_pnl_path=paths["jq_pnl"],
+        out_dir=tmp_path / "reconcile",
+    )
+    report = pd.read_parquet(out_paths["daily_reconcile"])
+    assert list(report.columns) == RECONCILE_COLUMNS
+    assert out_paths["summary_md"].exists()
+    assert out_paths["summary_csv"].exists()
+    return report
+
+
+def test_reconcile_exact_match(tmp_path):
+    report = _run_reconcile_case(tmp_path)
+    assert report.loc[0, "diff_reason"] == "MATCH"
+    assert report.loc[0, "filled_share_diff"] == 0
+    assert report.loc[0, "position_share_diff"] == 0
+
+
+def test_reconcile_price_mismatch(tmp_path):
+    report = _run_reconcile_case(tmp_path, jq_fills=_jq_fills(price=10.5))
+    assert report.loc[0, "diff_reason"] == "PRICE_MISMATCH"
+
+
+def test_reconcile_cost_mismatch(tmp_path):
+    report = _run_reconcile_case(
+        tmp_path,
+        jq_fills=_jq_fills(cost=8.0),
+        jq_pnl=_jq_pnl(cost=8.0),
+    )
+    assert report.loc[0, "diff_reason"] == "COST_MODEL"
+
+
+def test_reconcile_missing_symbol_in_joinquant(tmp_path):
+    empty_jq_fills = pd.DataFrame(columns=JOINQUANT_FILL_COLUMNS)
+    empty_jq_positions = pd.DataFrame(columns=JOINQUANT_POSITION_COLUMNS)
+    report = _run_reconcile_case(
+        tmp_path,
+        jq_fills=empty_jq_fills,
+        jq_positions=empty_jq_positions,
+    )
+    assert report.loc[0, "diff_reason"] == "MISSING_IN_JOINQUANT"
+
+
+def test_reconcile_short_target_with_long_only_joinquant_output(tmp_path):
+    positions = _positions(shares=-100, price=10.0)
+    our_fills = _our_fills(shares=-100, price=10.0)
+    jq_fills = _jq_fills(shares=0, price=10.0, cost=0.0, raw_status="short clipped")
+    jq_positions = _jq_positions(shares=0, price=10.0)
+
+    report = _run_reconcile_case(
+        tmp_path,
+        positions=positions,
+        our_fills=our_fills,
+        jq_fills=jq_fills,
+        jq_positions=jq_positions,
+    )
+    assert report.loc[0, "diff_reason"] == "SHORT_NOT_SUPPORTED"
+
+
+def test_joinquant_cli_smoke_export_ingest_reconcile_and_wrapper(tmp_path):
+    runner = CliRunner()
+    positions_path = tmp_path / "positions.pq"
+    _positions().to_parquet(positions_path, index=False)
+
+    result = runner.invoke(cli, [
+        "joinquant", "export-targets",
+        "--positions-path", str(positions_path),
+        "--portfolio-name", "run1",
+        "--mode", "target_shares",
+        "--out-dir", str(tmp_path / "targets"),
+    ])
+    assert result.exit_code == 0, result.output
+    assert "Exported JoinQuant targets" in result.output
+
+    fills_csv = tmp_path / "jq_fills.csv"
+    positions_csv = tmp_path / "jq_positions.csv"
+    pnl_csv = tmp_path / "jq_pnl.csv"
+    _jq_fills().to_csv(fills_csv, index=False)
+    _jq_positions().to_csv(positions_csv, index=False)
+    _jq_pnl().to_csv(pnl_csv, index=False)
+
+    result = runner.invoke(cli, [
+        "joinquant", "ingest",
+        "--portfolio-name", "run1",
+        "--fills-csv", str(fills_csv),
+        "--positions-csv", str(positions_csv),
+        "--pnl-csv", str(pnl_csv),
+        "--out-dir", str(tmp_path / "ingested"),
+    ])
+    assert result.exit_code == 0, result.output
+    assert "Saved JoinQuant fills" in result.output
+
+    paths = _write_parquets(tmp_path, {
+        "our_fills": _our_fills(),
+        "our_pnl": _our_pnl(),
+    })
+    result = runner.invoke(cli, [
+        "joinquant", "reconcile",
+        "--portfolio-name", "run1",
+        "--targets-dir", str(tmp_path / "targets" / "run1"),
+        "--our-fills-path", str(paths["our_fills"]),
+        "--our-positions-path", str(positions_path),
+        "--our-pnl-path", str(paths["our_pnl"]),
+        "--jq-fills-path", str(tmp_path / "ingested" / "run1" / "fills.pq"),
+        "--jq-positions-path", str(tmp_path / "ingested" / "run1" / "positions.pq"),
+        "--jq-pnl-path", str(tmp_path / "ingested" / "run1" / "pnl.pq"),
+        "--out-dir", str(tmp_path / "reconcile"),
+    ])
+    assert result.exit_code == 0, result.output
+    assert "Saved reconciliation parquet" in result.output
+
+    wrapper_path = tmp_path / "wrapper_strategy_run1.py"
+    result = runner.invoke(cli, [
+        "joinquant", "write-wrapper",
+        "--portfolio-name", "run1",
+        "--mode", "target_shares",
+        "--out-path", str(wrapper_path),
+    ])
+    assert result.exit_code == 0, result.output
+    assert "Saved JoinQuant wrapper strategy" in result.output
+    text = wrapper_path.read_text()
+    assert 'PORTFOLIO_NAME = "run1"' in text
+    assert 'TARGET_MODE = "target_shares"' in text
+    assert "ALLOW_SHORT = False" in text
+
+
+def test_wrapper_strategy_generation_smoke(tmp_path):
+    path = write_wrapper_strategy(
+        portfolio_name="run2",
+        mode="target_value",
+        out_path=tmp_path / "wrapper.py",
+    )
+    text = path.read_text()
+    assert 'PORTFOLIO_NAME = "run2"' in text
+    assert 'TARGET_MODE = "target_value"' in text
+    assert "order_target_value" in text
+
+
+def test_wrapper_strategy_can_embed_target_csvs(tmp_path):
+    targets_dir = tmp_path / "targets"
+    targets_dir.mkdir()
+    (targets_dir / "20260701.csv").write_text(
+        "date,portfolio_name,symbol_id,jq_symbol,target_shares,target_value,export_mode\n"
+        "2026-07-01,run1,sh600000,600000.XSHG,1000,10000,target_shares\n",
+        encoding="utf-8",
+    )
+
+    path = write_wrapper_strategy(
+        portfolio_name="run1",
+        mode="target_shares",
+        out_path=tmp_path / "wrapper_embedded.py",
+        embedded_targets_dir=targets_dir,
+    )
+
+    text = path.read_text()
+    assert '"20260701.csv"' in text
+    assert "600000.XSHG" in text
+    assert "if file_name in _EMBEDDED_TARGETS" in text
+
+
+def test_build_fixed_share_positions_excludes_final_executionless_date():
+    data = pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000", "sh600000"],
+        "date": pd.to_datetime(["2024-01-09", "2024-01-10", "2024-01-11"]),
+        "close": [10.0, 10.5, 11.0],
+    })
+
+    positions = build_fixed_share_positions(
+        data,
+        trade_symbol="sh600000",
+        portfolio_name="run1",
+        shares=1000,
+        booksize=1_000_000.0,
+    )
+
+    assert list(positions.columns) == POSITION_COLUMNS
+    assert positions["date"].dt.strftime("%Y-%m-%d").tolist() == [
+        "2024-01-09",
+        "2024-01-10",
+    ]
+    assert positions["position_shares"].tolist() == [1000, 1000]
+    assert positions["target_value"].tolist() == [10_000.0, 10_500.0]
+
+
+def test_browser_config_template_and_placeholder_resolution(tmp_path):
+    config_path = write_browser_config_template(
+        tmp_path / "browser_config.json",
+        strategy_url="https://www.joinquant.com/example",
+    )
+    config = json.loads(config_path.read_text())
+    assert config["strategy_url"] == "https://www.joinquant.com/example"
+    assert config["actions"][0]["type"] == "goto"
+
+    context = {
+        "wrapper_path": "/tmp/wrapper.py",
+        "target_csvs": ["/tmp/20240110.csv", "/tmp/20240111.csv"],
+        "expected_joinquant_csvs": {"fills": "/tmp/jq_fills.csv"},
+    }
+    assert resolve_template("{wrapper_path}", context) == "/tmp/wrapper.py"
+    assert resolve_template("{target_csvs}", context) == [
+        "/tmp/20240110.csv",
+        "/tmp/20240111.csv",
+    ]
+    assert resolve_template("save:{expected_joinquant_csvs.fills}", context) == "save:/tmp/jq_fills.csv"
+    script = "(arg) => { return new Event('change', {bubbles: true}); }"
+    assert resolve_template(script, context) == script
+
+
+def test_load_env_file_handles_quotes_without_shell_sourcing(tmp_path):
+    env_path = tmp_path / "joinquant.env"
+    env_path.write_text(
+        "JOINQUANT_USERNAME=alice\n"
+        "JOINQUANT_PASSWORD=\"secret\"\n"
+        "JOINQUANT_STRATEGY_URL='https://example.test/path?x=1&y=2'\n"
+    )
+
+    env = load_env_file(env_path)
+    assert env["JOINQUANT_USERNAME"] == "alice"
+    assert env["JOINQUANT_PASSWORD"] == "secret"
+    assert env["JOINQUANT_STRATEGY_URL"] == "https://example.test/path?x=1&y=2"
+
+
+def test_sim_trade_browser_config_template(tmp_path):
+    config_path = write_browser_config_template(
+        tmp_path / "sim_config.json",
+        strategy_url="https://www.joinquant.com/sim",
+        flow="sim-trade",
+    )
+    config = json.loads(config_path.read_text())
+
+    assert config["flow"] == "sim-trade"
+    selectors = " ".join(
+        action.get("selector", "") for action in config["actions"]
+    )
+    assert "模拟盘" in selectors
+    assert "模拟交易" in selectors
+    assert any(
+        action["type"] == "screenshot"
+        and action["path"] == "{run_artifact_dir}/sim_trade_final.png"
+        for action in config["actions"]
+    )
+
+
+def test_joinquant_cli_browser_config_smoke(tmp_path):
+    runner = CliRunner()
+    config_path = tmp_path / "browser_config.json"
+    result = runner.invoke(cli, [
+        "joinquant",
+        "write-browser-config",
+        "--out-path",
+        str(config_path),
+        "--strategy-url",
+        "https://www.joinquant.com/example",
+    ])
+
+    assert result.exit_code == 0, result.output
+    assert config_path.exists()
+    assert default_browser_config()["actions"]
+
+    sim_config_path = tmp_path / "sim_browser_config.json"
+    result = runner.invoke(cli, [
+        "joinquant",
+        "write-browser-config",
+        "--out-path",
+        str(sim_config_path),
+        "--strategy-url",
+        "https://www.joinquant.com/sim",
+        "--flow",
+        "sim-trade",
+    ])
+    assert result.exit_code == 0, result.output
+    assert json.loads(sim_config_path.read_text())["flow"] == "sim-trade"

tests/test_minute_downloader.py

@@ -0,0 +1,369 @@
+"""Tests for raw Baostock minute bar download plumbing."""
+
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+import pytest
+
+import data.downloader as low_level_downloader
+import pipeline.data.downloader as pipeline_downloader
+from data.downloader import download_minute_batch
+from pipeline.common.schema import MINUTE_BAR_COLUMNS
+from pipeline.data.downloader import download_minute_universe
+
+
+class _FakeResult:
+    def __init__(self, rows, error_code="0", error_msg=""):
+        self.rows = rows
+        self.error_code = error_code
+        self.error_msg = error_msg
+        self._idx = -1
+
+    def next(self):
+        self._idx += 1
+        return self._idx < len(self.rows)
+
+    def get_row_data(self):
+        return self.rows[self._idx]
+
+
+def test_download_minute_batch_maps_and_parses_baostock_rows(monkeypatch):
+    rows = [
+        [
+            "2024-01-02",
+            "20240102093500000",
+            "sh.600000",
+            "10",
+            "11",
+            "9",
+            "10.5",
+            "1000",
+            "10500",
+            "3",
+        ],
+        [
+            "2024-01-02",
+            "20240102094000000",
+            "sh.600000",
+            "10.5",
+            "12",
+            "10",
+            "11",
+            "2000",
+            "22000",
+            "3",
+        ],
+    ]
+    calls = []
+
+    def fake_query(**kwargs):
+        calls.append(kwargs)
+        return _FakeResult(rows)
+
+    monkeypatch.setattr(low_level_downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(low_level_downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        low_level_downloader.bs,
+        "query_history_k_data_plus",
+        fake_query,
+    )
+
+    [(symbol, df)] = list(
+        download_minute_batch(
+            ["sh600000"],
+            "2024-01-02",
+            "2024-01-02",
+            frequency=5,
+        )
+    )
+
+    assert symbol == "sh600000"
+    assert calls[0]["code"] == "sh.600000"
+    assert calls[0]["frequency"] == "5"
+    assert calls[0]["adjustflag"] == "3"
+    assert df is not None
+    assert df["datetime"].iloc[0] == pd.Timestamp("2024-01-02 09:35:00")
+    assert df["time"].tolist() == ["09:35:00", "09:40:00"]
+    assert (df["frequency"] == "5m").all()
+    assert np.isclose(df["open"].iloc[0], 10.0)
+    assert np.isclose(df["vwap"].iloc[0], 10.5)
+    assert pd.api.types.is_numeric_dtype(df["volume"])
+
+
+def test_minute_frequency_and_timestamp_parsing_edge_cases():
+    frequency, label = low_level_downloader._normalize_minute_frequency("15m")
+    assert (frequency, label) == ("15", "15m")
+
+    with pytest.raises(ValueError, match="Unsupported minute frequency"):
+        low_level_downloader._normalize_minute_frequency("1m")
+
+    parsed = low_level_downloader._parse_minute_datetime(
+        pd.Series(["2024-01-02", "2024-01-02"]),
+        pd.Series(["0935", "09:40:00"]),
+    )
+
+    assert parsed.tolist() == [
+        pd.Timestamp("2024-01-02 09:35:00"),
+        pd.Timestamp("2024-01-02 09:40:00"),
+    ]
+
+
+def test_download_minute_batch_empty_result_yields_none(monkeypatch):
+    monkeypatch.setattr(low_level_downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(low_level_downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        low_level_downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: _FakeResult([]),
+    )
+
+    assert list(download_minute_batch(["sh600000"], "2024-01-02", "2024-01-02")) == [
+        ("sh600000", None)
+    ]
+
+
+def test_download_minute_batch_non_login_error_and_periodic_relogin(monkeypatch):
+    responses = [
+        _FakeResult([], error_code="1", error_msg="bad symbol"),
+        _FakeResult([]),
+    ]
+    login_count = 0
+    logout_count = 0
+
+    def fake_login():
+        nonlocal login_count
+        login_count += 1
+
+    def fake_logout():
+        nonlocal logout_count
+        logout_count += 1
+
+    monkeypatch.setattr(low_level_downloader.bs, "login", fake_login)
+    monkeypatch.setattr(low_level_downloader.bs, "logout", fake_logout)
+    monkeypatch.setattr(
+        low_level_downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: responses.pop(0),
+    )
+
+    assert list(
+        download_minute_batch(
+            ["sh600000", "sz000001"],
+            "2024-01-02",
+            "2024-01-02",
+            relogin_every=1,
+        )
+    ) == [("sh600000", None), ("sz000001", None)]
+    assert login_count == 2
+    assert logout_count == 2
+
+
+def test_download_minute_batch_second_session_loss_yields_none(monkeypatch):
+    responses = [
+        _FakeResult([], error_code="10002007", error_msg="用户未登录"),
+        _FakeResult([], error_code="10002007", error_msg="用户未登录"),
+    ]
+
+    monkeypatch.setattr(low_level_downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(low_level_downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        low_level_downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: responses.pop(0),
+    )
+
+    assert list(download_minute_batch(["sh600000"], "2024-01-02", "2024-01-02")) == [
+        ("sh600000", None)
+    ]
+
+
+def test_download_minute_batch_ignores_relogin_and_final_logout_failures(monkeypatch):
+    responses = [
+        _FakeResult([], error_code="1", error_msg="bad symbol"),
+        _FakeResult([]),
+    ]
+    logout_count = 0
+
+    def fake_logout():
+        nonlocal logout_count
+        logout_count += 1
+        raise RuntimeError("logout failed")
+
+    monkeypatch.setattr(low_level_downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(low_level_downloader.bs, "logout", fake_logout)
+    monkeypatch.setattr(
+        low_level_downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: responses.pop(0),
+    )
+
+    assert list(
+        download_minute_batch(
+            ["sh600000", "sz000001"],
+            "2024-01-02",
+            "2024-01-02",
+            relogin_every=1,
+        )
+    ) == [("sh600000", None), ("sz000001", None)]
+    assert logout_count == 2
+
+
+def test_download_minute_batch_rejects_unparsed_timestamps(monkeypatch):
+    bad_rows = [[
+        "2024-01-02",
+        "not-a-time",
+        "sh.600000",
+        "10",
+        "11",
+        "9",
+        "10.5",
+        "1000",
+        "10500",
+        "3",
+    ]]
+    monkeypatch.setattr(low_level_downloader.bs, "login", lambda: None)
+    monkeypatch.setattr(low_level_downloader.bs, "logout", lambda: None)
+    monkeypatch.setattr(
+        low_level_downloader.bs,
+        "query_history_k_data_plus",
+        lambda **kwargs: _FakeResult(bad_rows),
+    )
+
+    assert list(download_minute_batch(["sh600000"], "2024-01-02", "2024-01-02")) == [
+        ("sh600000", None)
+    ]
+
+
+def test_download_minute_universe_writes_frequency_month_partitions(tmp_path, monkeypatch):
+    minute = pd.DataFrame({
+        "symbol": ["sh600000", "sh600000"],
+        "datetime": pd.to_datetime(["2024-01-02 09:35:00", "2024-01-02 09:40:00"]),
+        "date": pd.to_datetime(["2024-01-02", "2024-01-02"]),
+        "time": ["09:35:00", "09:40:00"],
+        "frequency": ["5m", "5m"],
+        "open": [10.0, 10.5],
+        "high": [11.0, 12.0],
+        "low": [9.0, 10.0],
+        "close": [10.5, 11.0],
+        "volume": [1000.0, 2000.0],
+        "amount": [10500.0, 22000.0],
+        "vwap": [10.5, 11.0],
+        "adjustflag": ["3", "3"],
+    })
+
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "_resolve_universe",
+        lambda universe, max_symbols=0: pd.DataFrame({
+            "symbol_id": ["sh600000"],
+            "symbol_name": ["PF Bank"],
+        }),
+    )
+
+    def fake_batch(symbols, start, end, frequency=5):
+        assert symbols == ["sh600000"]
+        assert frequency == "5"
+        yield "sh600000", minute
+
+    monkeypatch.setattr(pipeline_downloader, "download_minute_batch", fake_batch)
+
+    preserved = tmp_path / "toy" / "frequency=15m" / "month=2024-01" / "old.pq"
+    preserved.parent.mkdir(parents=True)
+    preserved_minute = minute.copy()
+    preserved_minute["frequency"] = "15m"
+    preserved_minute["symbol_id"] = "sh600000"
+    preserved_minute["symbol_name"] = "PF Bank"
+    preserved_minute[MINUTE_BAR_COLUMNS].to_parquet(preserved, index=False)
+    stale = tmp_path / "toy" / "frequency=5m" / "month=2024-01" / "stale.pq"
+    stale.parent.mkdir(parents=True)
+    preserved_minute.assign(frequency="5m")[MINUTE_BAR_COLUMNS].to_parquet(stale, index=False)
+
+    stats = download_minute_universe(
+        universe="toy",
+        start_date="2024-01-02",
+        end_date="2024-01-02",
+        output_dir=str(tmp_path),
+        chunk_size=1,
+        frequency="5",
+    )
+
+    dataset_path = Path(stats["dataset_path"])
+    assert (dataset_path / "frequency=5m" / "month=2024-01").is_dir()
+    assert preserved.exists()
+    assert not stale.exists()
+    out = pd.read_parquet(dataset_path / "frequency=5m")
+    assert (set(MINUTE_BAR_COLUMNS) - {"frequency"}) <= set(out.columns)
+    assert set(out["symbol_id"]) == {"sh600000"}
+    assert set(out["symbol_name"]) == {"PF Bank"}
+    assert stats["n_rows"] == 2
+
+
+def test_download_minute_universe_raises_when_all_symbols_empty(tmp_path, monkeypatch):
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "_resolve_universe",
+        lambda universe, max_symbols=0: pd.DataFrame({
+            "symbol_id": ["sh600000"],
+            "symbol_name": ["PF Bank"],
+        }),
+    )
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "download_minute_batch",
+        lambda symbols, start, end, frequency=5: iter([("sh600000", None)]),
+    )
+
+    with pytest.raises(RuntimeError, match="No minute data"):
+        download_minute_universe(
+            universe="toy",
+            start_date="2024-01-02",
+            end_date="2024-01-02",
+            output_dir=str(tmp_path),
+        )
+
+
+def test_download_minute_universe_progress_branch_at_100_symbols(tmp_path, monkeypatch):
+    symbols = [f"sh6{i:05d}" for i in range(100)]
+    minute = pd.DataFrame({
+        "symbol": ["sh600000"],
+        "datetime": [pd.Timestamp("2024-01-02 09:35:00")],
+        "date": [pd.Timestamp("2024-01-02")],
+        "time": ["09:35:00"],
+        "frequency": ["5m"],
+        "open": [10.0],
+        "high": [11.0],
+        "low": [9.0],
+        "close": [10.5],
+        "volume": [1000.0],
+        "amount": [10500.0],
+        "vwap": [10.5],
+        "adjustflag": ["3"],
+    })
+
+    monkeypatch.setattr(
+        pipeline_downloader,
+        "_resolve_universe",
+        lambda universe, max_symbols=0: pd.DataFrame({
+            "symbol_id": symbols,
+            "symbol_name": symbols,
+        }),
+    )
+
+    def fake_batch(requested_symbols, start, end, frequency=5):
+        assert requested_symbols == symbols
+        for symbol in requested_symbols:
+            yield symbol, minute.copy()
+
+    monkeypatch.setattr(pipeline_downloader, "download_minute_batch", fake_batch)
+
+    stats = download_minute_universe(
+        universe="toy100",
+        start_date="2024-01-02",
+        end_date="2024-01-02",
+        output_dir=str(tmp_path),
+        chunk_size=200,
+    )
+
+    assert stats["n_symbols"] == 100
+    assert stats["n_rows"] == 100

tests/test_pipeline_invariants.py

@@ -0,0 +1,305 @@
+"""End-to-end correctness invariants for the reversal_5d pipeline (no network).
+
+Each test maps 1:1 to one of the ten review checks. Naming convention: the
+*execution date* ``d`` is the market session on which a target is actually
+filled at the open; the *signal date* ``t`` is the session whose close formed
+that target. The documented convention is ``d = next(t)`` (see
+``docs/portfolio_trading_cost_model.md``), so ``close[d-1] == close[t]``.
+"""
+
+import numpy as np
+import pandas as pd
+
+from pipeline.alpha.compute import compute_alpha, investable_universe_mask
+from pipeline.portfolio.construct import construct_positions
+from pipeline.portfolio.discretize import repair_exposure, round_to_valid_lot
+from pipeline.portfolio.market_rules import MarketRule
+from pipeline.portfolio.costs import SimpleProportionalCostModel
+from pipeline.portfolio.constraints import (
+    SuspensionConstraint,
+    VolumeCapConstraint,
+)
+from pipeline.portfolio.simulator import ReferenceSimulator
+
+
+_SYMBOLS = ("sh600000", "sz000001", "sh688981", "sz300750")
+
+
+def _panel(n_days=12, symbols=_SYMBOLS, start="2024-01-01", seed=0,
+           distinct_open=True):
+    """Contiguous long-format DATA frame with all columns the pipeline needs.
+
+    Open and close differ (so overnight vs intraday PnL terms are separable),
+    and the calendar is gap-free so each session is the next session's ``d-1``.
+    """
+    dates = pd.date_range(start, periods=n_days)
+    rng = np.random.default_rng(seed)
+    frames = []
+    for i, sym in enumerate(symbols):
+        close = np.abs(50.0 + i * 10 + np.cumsum(rng.standard_normal(n_days))) + 5.0
+        open_ = close * (0.99 + 0.02 * rng.random(n_days)) if distinct_open else close.copy()
+        preclose = np.concatenate([[close[0]], close[:-1]])
+        frames.append(pd.DataFrame({
+            "symbol_id": sym,
+            "symbol_name": sym,
+            "date": dates,
+            "open": open_,
+            "high": np.maximum(open_, close),
+            "low": np.minimum(open_, close),
+            "close": close,
+            "preclose": preclose,
+            "volume": 1_000_000.0,
+            "amount": 1_000_000.0 * close,
+            "tradestatus": 1,
+            "isST": 0,
+        }))
+    return pd.concat(frames, ignore_index=True)
+
+
+# --- 1. reversal signal uses only close[d-1] and earlier ---------------------
+
+def test_reversal_signal_does_not_peek_at_future_closes():
+    data = _panel(n_days=12)
+    base = compute_alpha(data, "rev", "reversal_rank", lookback=5)
+
+    # Perturb every close strictly AFTER an interior signal date t; the weight
+    # dated t (executed at d = t+1) must be unchanged — it may use close[t]
+    # (== close[d-1]) and earlier only.
+    t = sorted(data["date"].unique())[6]
+    future = data.copy()
+    mask = future["date"] > t
+    future.loc[mask, ["open", "high", "low", "close"]] *= 1.5
+    perturbed = compute_alpha(future, "rev", "reversal_rank", lookback=5)
+
+    b = base[base["date"] <= t].set_index(["symbol_id", "date"])["weight"]
+    p = perturbed[perturbed["date"] <= t].set_index(["symbol_id", "date"])["weight"]
+    pd.testing.assert_series_equal(b.sort_index(), p.sort_index())
+
+
+# --- 2. the executed (fill/PnL) date is the open-execution date d = next(t) --
+
+def test_fill_date_is_next_session_open_execution_date():
+    data = _panel(n_days=8)
+    weights = compute_alpha(data, "c", "reversal_rank", lookback=3)
+    weights = weights.rename(columns={"alpha_name": "combo_name"})
+    weights["combo_name"] = "c"
+    pos = construct_positions(weights, data, booksize=1e6, portfolio_name="run1")
+    fills, pnl = ReferenceSimulator(cost_bps=5, slippage_bps=5).run(pos, data)
+
+    sessions = sorted(data["date"].unique())
+    nxt = {s: sessions[i + 1] for i, s in enumerate(sessions[:-1])}
+    # Every executed date equals the session AFTER some position (signal) date.
+    pos_dates = set(pos["date"].unique())
+    exec_dates = set(pnl["date"].unique())
+    assert exec_dates == {nxt[t] for t in pos_dates if t in nxt}
+
+    # Execution price is the open of the execution date, not the signal close.
+    opn = data.pivot_table(index="date", columns="symbol_id", values="open",
+                           aggfunc="first").sort_index()
+    d = sorted(exec_dates)[1]
+    row = fills[(fills["date"] == d) & (fills["traded_shares"] != 0)].iloc[0]
+    sym = row["symbol_id"]
+    expected_cost = abs(row["traded_shares"]) * opn.loc[d, sym] * (5 + 5) / 1e4
+    assert np.isclose(row["trade_cost"], expected_cost)
+
+
+# --- 3. PnL identity: overnight(old book) + intraday(new book) - cost --------
+
+def test_daily_pnl_matches_overnight_plus_intraday_minus_cost():
+    data = _panel(n_days=8)
+    weights = compute_alpha(data, "c", "reversal_rank", lookback=3)
+    weights = weights.rename(columns={"alpha_name": "combo_name"})
+    weights["combo_name"] = "c"
+    pos = construct_positions(weights, data, booksize=1e6, portfolio_name="run1")
+    fills, pnl = ReferenceSimulator(cost_bps=5, slippage_bps=5).run(pos, data)
+
+    opn = data.pivot_table(index="date", columns="symbol_id", values="open", aggfunc="first").sort_index()
+    cls = data.pivot_table(index="date", columns="symbol_id", values="close", aggfunc="first").sort_index()
+    sessions = list(cls.index)
+
+    prev_close_of = {sessions[i]: sessions[i - 1] for i in range(1, len(sessions))}
+
+    for d in sorted(pnl["date"].unique()):
+        day = fills[fills["date"] == d]
+        prev = day.set_index("symbol_id")["prev_shares"]
+        realized = day.set_index("symbol_id")["realized_shares"]
+        cost = day["trade_cost"].sum()
+
+        intraday = float((realized * (cls.loc[d] - opn.loc[d]).reindex(realized.index)).sum())
+        # Overnight gap on the OLD book is taken from the previous *executed*
+        # date's close. With a gap-free calendar and daily execution that is the
+        # immediately preceding session; the first executed date has no prior
+        # book so the term is naturally zero (prev_shares == 0 there).
+        pc = prev_close_of.get(d)
+        if pc is not None and (prev != 0).any():
+            overnight = float((prev * (opn.loc[d] - cls.loc[pc]).reindex(prev.index)).sum())
+        else:
+            overnight = 0.0
+        expected = overnight + intraday - cost
+        got = float(pnl[pnl["date"] == d]["pnl"].iloc[0])
+        assert np.isclose(got, expected, rtol=1e-6, atol=1e-3), (d, got, expected)
+
+
+# --- 4. realized shares (not target shares) are threaded into the next day ---
+
+def test_realized_not_target_threaded_forward():
+    data = _panel(n_days=6)
+    weights = compute_alpha(data, "c", "reversal_rank", lookback=2)
+    weights = weights.rename(columns={"alpha_name": "combo_name"})
+    weights["combo_name"] = "c"
+    pos = construct_positions(weights, data, booksize=1e8, portfolio_name="run1")
+
+    # A tight volume cap forces partial fills, so realized != target on most
+    # names — exactly the case where threading target vs realized diverges.
+    fills, _ = ReferenceSimulator(
+        constraints=[VolumeCapConstraint(max_frac=1e-6)]
+    ).run(pos, data)
+
+    wide_prev = fills.pivot_table(index="date", columns="symbol_id", values="prev_shares", aggfunc="first")
+    wide_real = fills.pivot_table(index="date", columns="symbol_id", values="realized_shares", aggfunc="first")
+    exec_dates = list(wide_prev.index)
+    assert len(exec_dates) >= 2
+    # Today's prev_shares == yesterday's realized_shares for every name.
+    for a, b in zip(exec_dates[:-1], exec_dates[1:]):
+        prev_today = wide_prev.loc[b].dropna()
+        real_yest = wide_real.loc[a].reindex(prev_today.index).fillna(0.0)
+        pd.testing.assert_series_equal(
+            prev_today.astype(float), real_yest.astype(float), check_names=False
+        )
+    # And realized actually diverged from target (cap bit), so the test is real.
+    assert (fills["realized_shares"] != fills["target_shares"]).any()
+
+
+# --- 5. blocked trades create zero traded_shares and zero trade_cost ---------
+
+def test_blocked_trade_has_zero_shares_and_zero_cost():
+    data = _panel(n_days=6)
+    # Suspend one name on every session so any attempt to trade it is blocked.
+    data.loc[data["symbol_id"] == "sz000001", "tradestatus"] = 0
+    weights = compute_alpha(data, "c", "reversal_rank", lookback=2)
+    weights = weights.rename(columns={"alpha_name": "combo_name"})
+    weights["combo_name"] = "c"
+    pos = construct_positions(weights, data, booksize=1e6, portfolio_name="run1")
+
+    fills, _ = ReferenceSimulator(
+        constraints=[SuspensionConstraint()], cost_bps=5, slippage_bps=5
+    ).run(pos, data)
+
+    blocked = fills[fills["blocked"] == 1]
+    assert (blocked["traded_shares"] == 0).all()
+    assert (blocked["trade_cost"] == 0.0).all()
+    # The suspended name never trades and never accrues cost.
+    susp = fills[fills["symbol_id"] == "sz000001"]
+    assert (susp["traded_shares"] == 0).all()
+    assert (susp["trade_cost"] == 0.0).all()
+
+
+# --- 6. liquid universe uses only information known before open[d] -----------
+
+def test_investable_universe_mask_is_causal():
+    data = _panel(n_days=14)
+    close = data.pivot_table(index="date", columns="symbol_id", values="close", aggfunc="first").sort_index()
+    full = investable_universe_mask(data, close, top_n=10, min_history=3)
+
+    t = sorted(data["date"].unique())[8]
+    # Recompute the mask from data truncated at the signal date t: the mask row
+    # for t must be identical, proving it never reads dates > t (i.e. nothing
+    # from open[d=t+1] onward).
+    trunc = data[data["date"] <= t]
+    close_t = close.loc[:t]
+    mask_t = investable_universe_mask(trunc, close_t, top_n=10, min_history=3)
+    pd.testing.assert_series_equal(
+        full.loc[t].sort_index(), mask_t.loc[t].sort_index(), check_names=False
+    )
+
+
+# --- 7. cost bps is one-way per-trade (a round trip is charged twice) --------
+
+def test_cost_bps_is_one_way_per_trade():
+    model = SimpleProportionalCostModel(cost_bps=5, slippage_bps=5)
+    price = np.array([20.0])
+
+    buy = model.compute(np.array([1000]), price, np.array([1]), date=None)
+    sell = model.compute(np.array([-1000]), price, np.array([-1]), date=None)
+
+    one_way = 1000 * 20 * (5 + 5) / 1e4
+    assert np.isclose(buy[0], one_way)      # charged once on the buy leg
+    assert np.isclose(sell[0], one_way)     # charged again on the sell leg
+    # A full round trip (enter then exit) therefore costs ~2x the one-way rate.
+    assert np.isclose(buy[0] + sell[0], 2 * one_way)
+
+
+# --- 8. execution & PnL use raw tradable prices on the same scale as shares --
+
+def test_position_value_is_shares_times_raw_price():
+    data = _panel(n_days=10)
+    weights = compute_alpha(data, "c", "reversal_rank", lookback=3)
+    weights = weights.rename(columns={"alpha_name": "combo_name"})
+    weights["combo_name"] = "c"
+    pos = construct_positions(weights, data, booksize=1e6, portfolio_name="run1")
+
+    finite = pos["price"] > 0
+    # The stored value is exactly integer shares × the raw construction price —
+    # no adjusted-price factor is mixed into the share→value accounting.
+    expected = pos.loc[finite, "position_shares"] * pos.loc[finite, "price"]
+    pd.testing.assert_series_equal(
+        pos.loc[finite, "position_value"].astype(float),
+        expected.astype(float),
+        check_names=False,
+    )
+
+
+# --- 9. alpha is scale-free (adjusted prices ok); accounting uses raw units --
+
+def test_alpha_weights_invariant_to_per_symbol_price_scaling():
+    data = _panel(n_days=12)
+    base = compute_alpha(data, "rev", "reversal_rank", lookback=5)
+
+    # A qfq/hfq adjustment is (per symbol) a multiplicative rescaling of the
+    # price series; pct_change is scale-free, so the alpha weights must not move.
+    scaled = data.copy()
+    factor = {"sh600000": 2.0, "sz000001": 0.5, "sh688981": 3.0, "sz300750": 1.25}
+    for sym, f in factor.items():
+        m = scaled["symbol_id"] == sym
+        scaled.loc[m, ["open", "high", "low", "close"]] *= f
+    scaled_alpha = compute_alpha(scaled, "rev", "reversal_rank", lookback=5)
+
+    b = base.set_index(["symbol_id", "date"])["weight"].sort_index()
+    s = scaled_alpha.set_index(["symbol_id", "date"])["weight"].sort_index()
+    pd.testing.assert_series_equal(b, s)
+
+
+# --- 10. repaired book stays on valid A-share lot lattices -------------------
+
+def _on_lattice(q, min_open, increment):
+    q = np.abs(np.asarray(q, dtype=np.int64))
+    on = (q == 0) | ((q >= min_open) & ((q - min_open) % increment == 0))
+    return bool(on.all())
+
+
+def test_repair_output_stays_on_lot_lattice():
+    # Pre-2023 main board has a 100-share increment (the strongest lattice
+    # constraint); STAR uses min 200 / increment 1.
+    symbols = np.array(["sh600000", "sz000001", "sh688981", "sz300750"], dtype=object)
+    rule = MarketRule()
+    on = "2022-06-01"  # pre 2023-08-10 → main-board increment is 100
+    min_open, increment, odd_full, _ = rule.get_rules_vectorized(
+        symbols, on, np.zeros(len(symbols), dtype=bool)
+    )
+    assert min_open[0] == 100 and increment[0] == 100  # main board pre-2023
+
+    price = np.array([12.3, 8.7, 45.0, 230.0])
+    prev = np.zeros(len(symbols), dtype=np.int64)
+    q_target = np.array([3251.0, -7777.0, 640.0, -415.0])
+
+    q_round = round_to_valid_lot(q_target, prev, min_open, increment, odd_full)
+    assert _on_lattice(q_round, min_open, increment)
+
+    repaired = repair_exposure(
+        q_round, q_target, price, increment, min_open, prev, odd_full,
+        booksize=float(np.abs(q_target * price).sum()),
+    )
+    assert _on_lattice(repaired, min_open, increment)
+    # Repair never flips a name's sign relative to the rounded book.
+    nz = q_round != 0
+    assert np.all(np.sign(repaired[nz]) * np.sign(q_round[nz]) >= 0)

tests/test_portfolio.py

@@ -13,15 +13,21 @@ from pipeline.portfolio.market_rules import (
     Board,
     LimitStatus,
     MarketRule,
+    _to_date,
     compute_limit_status,
     detect_board,
 )
 from pipeline.portfolio.research import evaluate_portfolio
 from pipeline.portfolio.constraints import (
+    TradeConstraint,
+    available_constraints,
+    get_constraint,
     PriceLimitConstraint,
+    register_constraint,
     SuspensionConstraint,
     VolumeCapConstraint,
 )
+from pipeline.portfolio.costs import SimpleProportionalCostModel
 from pipeline.portfolio.simulator import (
     MarketSlice,
     ReferenceSimulator,
@@ -64,7 +70,7 @@ def _make_data(n_days: int = 40, symbols=_SYMBOLS, start="2024-01-01",
 def _make_weights(data: pd.DataFrame, name="combo") -> pd.DataFrame:
     """Demeaned per-date signed weights so the cross-section is dollar-neutral."""
     close = data.pivot_table(index="date", columns="symbol_id", values="close").sort_index()
-    raw = -close.pct_change(5)
+    raw = -close.pct_change(5, fill_method=None)
     demeaned = raw.sub(raw.mean(axis=1), axis=0)
     long = demeaned.reset_index().melt(id_vars="date", var_name="symbol_id",
                                        value_name="weight").dropna()
@@ -81,6 +87,7 @@ def test_detect_board():
     assert detect_board("sh688981") == Board.STAR
     assert detect_board("sz300750") == Board.CHINEXT
     assert detect_board("bj830000") == Board.UNKNOWN
+    assert detect_board("sh") == Board.UNKNOWN
 
 
 # --- MarketRule date transitions ---------------------------------------------
@@ -119,6 +126,26 @@ def test_get_rules_vectorized():
     assert list(limit) == [0.10, 0.20, 0.20]
 
 
+def test_market_rule_date_coercion_unknown_board_and_st_vector_override():
+    rules = MarketRule()
+
+    assert _to_date(dt.datetime(2024, 1, 2, 15, 0)) == dt.date(2024, 1, 2)
+    assert _to_date(pd.Timestamp("2024-01-03 09:30")) == dt.date(2024, 1, 3)
+    assert _to_date("2024-01-04 10:00:00") == dt.date(2024, 1, 4)
+
+    unknown_rule = rules.get_rule("xx999999", "2024-01-02")
+    assert unknown_rule.minimum_open_size == 100
+    assert unknown_rule.share_increment == 100
+    assert unknown_rule.price_limit_pct == 0.10
+
+    _, _, _, limit = rules.get_rules_vectorized(
+        np.array(["sh600000", "sh600000"], dtype=object),
+        "2024-01-02",
+        np.array([0, 1]),
+    )
+    assert limit.tolist() == [0.10, 0.05]
+
+
 def test_compute_limit_status():
     price = np.array([110.0, 90.0, 100.0])
     preclose = np.array([100.0, 100.0, 100.0])
@@ -129,6 +156,20 @@ def test_compute_limit_status():
     assert status[2] == LimitStatus.NORMAL.value
 
 
+def test_compute_limit_status_treats_bad_preclose_as_normal():
+    status = compute_limit_status(
+        price=np.array([np.nan, 110.0, 90.0]),
+        preclose=np.array([100.0, np.nan, 0.0]),
+        limit_pct=np.array([0.10, 0.10, 0.10]),
+    )
+
+    assert status.tolist() == [
+        LimitStatus.NORMAL.value,
+        LimitStatus.NORMAL.value,
+        LimitStatus.NORMAL.value,
+    ]
+
+
 # --- continuous targets ------------------------------------------------------
 
 def test_continuous_targets_normalization():
@@ -294,6 +335,70 @@ def test_repair_scales_to_4000_names():
     assert abs(gross - B) <= 0.03 * B
 
 
+def test_repair_handles_empty_input():
+    result = repair_exposure(
+        np.array([], dtype=np.int64),
+        np.array([], dtype=float),
+        np.array([], dtype=float),
+        np.array([], dtype=np.int64),
+        np.array([], dtype=np.int64),
+        np.array([], dtype=np.int64),
+    )
+
+    assert result.dtype == np.int64
+    assert result.tolist() == []
+
+
+def test_repair_respects_max_iters_and_zero_increment_noop():
+    q_round = np.array([100, -100], dtype=np.int64)
+    q_target = np.array([300.0, -300.0])
+    price = np.array([10.0, 10.0])
+    min_open = np.array([100, 100])
+    prev = np.zeros(2, dtype=np.int64)
+
+    capped = repair_exposure(
+        q_round,
+        q_target,
+        price,
+        increment=np.array([1, 1]),
+        min_open=min_open,
+        prev_shares=prev,
+        booksize=10_000.0,
+        gross_tol=0.0,
+        max_iters=0,
+    )
+    zero_increment = repair_exposure(
+        q_round,
+        q_target,
+        price,
+        increment=np.array([0, 0]),
+        min_open=min_open,
+        prev_shares=prev,
+        booksize=0.0,
+        net_tol=0.0,
+        gross_tol=0.0,
+    )
+
+    assert capped.tolist() == [100, -100]
+    assert zero_increment.tolist() == [100, -100]
+
+
+def test_repair_gross_growth_obeys_net_band():
+    pos = repair_exposure(
+        q_round=np.array([100], dtype=np.int64),
+        q_target=np.array([300.0]),
+        price=np.array([10.0]),
+        increment=np.array([1]),
+        min_open=np.array([100]),
+        prev_shares=np.array([0], dtype=np.int64),
+        booksize=2_000.0,
+        net_tol=0.5,
+        gross_tol=0.0,
+    )
+
+    assert pos.tolist() == [100]
+
+
 # --- construct_positions -----------------------------------------------------
 
 def test_construct_positions_schema():
@@ -305,6 +410,47 @@ def test_construct_positions_schema():
     assert pos["position_shares"].dtype == np.int64
 
 
+def test_construct_positions_empty_weights_returns_schema():
+    data = _make_data(n_days=3)
+    empty_weights = pd.DataFrame(columns=["symbol_id", "date", "combo_name", "weight"])
+
+    pos = construct_positions(
+        empty_weights,
+        data,
+        booksize=1e6,
+        portfolio_name="empty",
+    )
+
+    assert list(pos.columns) == POSITION_COLUMNS
+    assert pos.empty
+
+
+def test_construct_positions_ignores_absent_or_bad_prices():
+    dates = pd.to_datetime(["2024-01-01", "2024-01-02"])
+    data = pd.DataFrame([
+        {"symbol_id": "sh600000", "date": dates[0], "close": np.nan, "isST": 0},
+        {"symbol_id": "sz000001", "date": dates[0], "close": 20.0, "isST": 0},
+        {"symbol_id": "sh600000", "date": dates[1], "close": 10.0, "isST": 0},
+        {"symbol_id": "sz000001", "date": dates[1], "close": 20.0, "isST": 0},
+    ])
+    weights = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000", "sz000001"],
+        "date": [dates[0], dates[0], dates[1], dates[1]],
+        "combo_name": ["combo"] * 4,
+        "weight": [1.0, -1.0, 1.0, -1.0],
+    })
+
+    pos = construct_positions(weights, data, booksize=10000.0, portfolio_name="bad_price")
+
+    bad_price_rows = pos[
+        (pos["date"] == dates[0])
+        & (pos["symbol_id"] == "sh600000")
+    ]
+    assert bad_price_rows.empty or (bad_price_rows["target_weight"] == 0.0).all()
+    assert np.isfinite(pos["target_value"]).all()
+    assert np.isfinite(pos["position_value"]).all()
+
+
 def test_construct_positions_threads_state_and_closes_absent():
     data = _make_data()
     weights = _make_weights(data)
@@ -320,6 +466,34 @@ def test_construct_positions_threads_state_and_closes_absent():
     assert final.empty or (final["position_shares"] == 0).all()
 
 
+def test_construct_positions_closes_absent_short_position():
+    dates = pd.to_datetime(["2024-01-02", "2024-01-03"])
+    data = pd.DataFrame([
+        {"symbol_id": sym, "date": d, "close": price, "isST": 0}
+        for d in dates
+        for sym, price in (("sh600000", 10.0), ("sz000001", 20.0))
+    ])
+    weights = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sz000001"],
+        "date": [dates[0], dates[0], dates[1]],
+        "combo_name": ["combo", "combo", "combo"],
+        "weight": [-1.0, 1.0, 1.0],
+    })
+
+    pos = construct_positions(weights, data, booksize=20000.0, portfolio_name="absent_short")
+
+    first_day_short = pos[
+        (pos["date"] == dates[0])
+        & (pos["symbol_id"] == "sh600000")
+    ]
+    final_day_short = pos[
+        (pos["date"] == dates[1])
+        & (pos["symbol_id"] == "sh600000")
+    ]
+    assert (first_day_short["position_shares"] < 0).all()
+    assert final_day_short.empty or (final_day_short["position_shares"] == 0).all()
+
+
 def test_construct_positions_carries_book_on_zero_gross(caplog):
     dates = pd.to_datetime(["2024-01-01", "2024-01-02"])
     symbols = ["sh600000", "sz000001"]
@@ -391,8 +565,117 @@ def test_volume_cap_uses_traded_value():
     assert low[0] == -10000.0
 
 
+def test_constraints_compose_repeatably_regardless_of_order():
+    n = 1
+    sl = _slice(
+        n,
+        tradestatus=np.array([0.0]),
+        limit_status=np.array([LimitStatus.UP_LIMIT.value], dtype=np.int8),
+        amount=np.array([1_000.0]),
+        price=np.array([10.0]),
+    )
+    ctx = TradeContext(np.zeros(n, np.int64), np.array([500]), sl, 1e6)
+    first_order = ReferenceSimulator(
+        constraints=[
+            SuspensionConstraint(),
+            PriceLimitConstraint(),
+            VolumeCapConstraint(max_frac=0.1),
+        ],
+        cost_bps=10,
+    ).fill(ctx)
+    reversed_order = ReferenceSimulator(
+        constraints=[
+            VolumeCapConstraint(max_frac=0.1),
+            PriceLimitConstraint(),
+            SuspensionConstraint(),
+        ],
+        cost_bps=10,
+    ).fill(ctx)
+
+    assert first_order.traded_shares.tolist() == [0]
+    assert first_order.realized_shares.tolist() == [0]
+    assert first_order.blocked.tolist() == [1]
+    assert np.array_equal(first_order.traded_shares, reversed_order.traded_shares)
+    assert np.array_equal(first_order.realized_shares, reversed_order.realized_shares)
+    assert np.array_equal(first_order.blocked, reversed_order.blocked)
+    assert np.array_equal(first_order.cost, reversed_order.cost)
+
+
+def test_constraint_registry_and_default_adjust_targets():
+    class _NoopConstraint(TradeConstraint):
+        name = "_coverage_noop_constraint"
+
+        def delta_bounds(self, ctx):
+            return np.zeros(1), np.ones(1)
+
+    registered = register_constraint(_NoopConstraint)
+
+    assert registered is _NoopConstraint
+    assert "_coverage_noop_constraint" in available_constraints()
+    assert isinstance(get_constraint("_coverage_noop_constraint"), _NoopConstraint)
+    assert get_constraint("_coverage_noop_constraint").adjust_targets(object()) is None
+    with np.testing.assert_raises(KeyError):
+        get_constraint("_missing_constraint")
+    with np.testing.assert_raises(TypeError):
+        register_constraint(object)  # type: ignore[arg-type]
+    with np.testing.assert_raises(ValueError):
+        class _NoNameConstraint(TradeConstraint):
+            def delta_bounds(self, ctx):
+                return np.zeros(1), np.ones(1)
+
+        register_constraint(_NoNameConstraint)
+    with np.testing.assert_raises(ValueError):
+        class _DuplicateConstraint(TradeConstraint):
+            name = "_coverage_noop_constraint"
+
+            def delta_bounds(self, ctx):
+                return np.zeros(1), np.ones(1)
+
+        register_constraint(_DuplicateConstraint)
+
+
 # --- ReferenceSimulator ------------------------------------------------------
 
+def test_simulator_applies_constraint_target_adjustment():
+    class _HalveTarget(TradeConstraint):
+        name = "_halve_target"
+
+        def adjust_targets(self, ctx):
+            return ctx.target_shares // 2
+
+        def delta_bounds(self, ctx):
+            return np.full(len(ctx.target_shares), -np.inf), np.full(len(ctx.target_shares), np.inf)
+
+    sl = _slice(1, price=np.array([10.0]))
+    ctx = TradeContext(np.array([0], np.int64), np.array([100], np.int64), sl, 1e6)
+
+    result = ReferenceSimulator(constraints=[_HalveTarget()]).fill(ctx)
+
+    assert result.traded_shares.tolist() == [50]
+    assert result.realized_shares.tolist() == [50]
+
+
+def test_simulator_empty_positions_uses_default_booksize():
+    data = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [pd.Timestamp("2024-01-02")],
+        "open": [10.0],
+        "close": [10.0],
+        "preclose": [10.0],
+        "amount": [1e9],
+        "tradestatus": [1],
+        "isST": [0],
+    })
+    positions = pd.DataFrame(columns=POSITION_COLUMNS)
+
+    fills, pnl = ReferenceSimulator().run(positions, data)
+
+    assert list(fills.columns) == FILL_COLUMNS
+    assert list(pnl.columns) == PNL_COLUMNS
+    assert fills.empty
+    assert pnl.empty
+
+
 def test_simulator_next_open_and_blocked_buy_holds_prev():
     data = _make_data(n_days=15)
     weights = _make_weights(data)
@@ -445,6 +728,7 @@ def test_simulator_blocked_buy_when_suspended():
     assert res.traded_shares[0] == 0
     assert res.realized_shares[0] == 0
     assert res.blocked[0] == 1
+    assert res.cost[0] == 0.0
 
 
 def test_simulator_cost_is_positive_when_trading():
@@ -458,6 +742,154 @@ def test_simulator_cost_is_positive_when_trading():
     assert np.isclose(res.cost[0], 1000 * 20 * 15 / 1e4)
 
 
+def test_simulator_cost_only_on_nonzero_realized_trades():
+    n = 2
+    sim = ReferenceSimulator(constraints=[], cost_bps=10)
+    sl = _slice(n, price=np.array([10.0, 20.0]))
+    ctx = TradeContext(np.array([100, 100], np.int64),
+                       np.array([100, 150], np.int64), sl, 1e6)
+
+    res = sim.fill(ctx)
+
+    assert res.traded_shares.tolist() == [0, 50]
+    assert res.cost[0] == 0.0
+    assert np.isclose(res.cost[1], 50 * 20 * 10 / 1e4)
+
+
+def test_simulator_short_to_long_flip_trades_full_delta():
+    dates = pd.to_datetime(["2024-01-01", "2024-01-02", "2024-01-03"])
+    positions = pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000"],
+        "date": [dates[0], dates[1]],
+        "portfolio_name": ["flip", "flip"],
+        "target_weight": [-1.0, 1.0],
+        "target_value": [-1000.0, 1000.0],
+        "target_shares": [-100.0, 100.0],
+        "position_shares": [-100, 100],
+        "position_value": [-1000.0, 1000.0],
+        "price": [10.0, 10.0],
+    })
+    data = pd.DataFrame({
+        "symbol_id": ["sh600000"] * 3,
+        "date": dates,
+        "open": [10.0, 10.0, 10.0],
+        "close": [10.0, 10.0, 10.0],
+        "preclose": [10.0, 10.0, 10.0],
+        "amount": [1e9, 1e9, 1e9],
+        "tradestatus": [1, 1, 1],
+        "isST": [0, 0, 0],
+    })
+
+    fills, _ = ReferenceSimulator().run(positions, data)
+
+    by_date = fills.set_index("date")
+    assert by_date.loc[dates[1], "traded_shares"] == -100
+    assert by_date.loc[dates[1], "realized_shares"] == -100
+    assert by_date.loc[dates[2], "prev_shares"] == -100
+    assert by_date.loc[dates[2], "traded_shares"] == 200
+    assert by_date.loc[dates[2], "realized_shares"] == 100
+
+
+def test_simulator_volume_cap_partially_fills_sell():
+    sl = _slice(1, amount=np.array([10_000.0]), price=np.array([10.0]))
+    ctx = TradeContext(
+        np.array([1000], np.int64),
+        np.array([0], np.int64),
+        sl,
+        1_000_000.0,
+    )
+
+    result = ReferenceSimulator(
+        constraints=[VolumeCapConstraint(max_frac=0.10)]
+    ).fill(ctx)
+
+    assert result.traded_shares.tolist() == [-100]
+    assert result.realized_shares.tolist() == [900]
+    assert result.blocked.tolist() == [1]
+
+
+def test_simulator_missing_next_open_has_zero_cost_and_turnover():
+    dates = pd.to_datetime(["2024-01-01", "2024-01-02"])
+    positions = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [dates[0]],
+        "portfolio_name": ["missing_open"],
+        "target_weight": [1.0],
+        "target_value": [1000.0],
+        "target_shares": [100.0],
+        "position_shares": [100],
+        "position_value": [1000.0],
+        "price": [10.0],
+    })
+    data = pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000"],
+        "date": dates,
+        "open": [10.0, np.nan],
+        "close": [10.0, 10.0],
+        "preclose": [10.0, 10.0],
+        "amount": [1e9, 1e9],
+        "tradestatus": [1, 1],
+        "isST": [0, 0],
+    })
+
+    fills, pnl = ReferenceSimulator(cost_bps=10, slippage_bps=5).run(positions, data)
+
+    assert fills["traded_shares"].iloc[0] == 100
+    assert fills["trade_cost"].iloc[0] == 0.0
+    assert pnl["cost"].iloc[0] == 0.0
+    assert pnl["turnover"].iloc[0] == 0.0
+    assert pnl["gross_exposure"].iloc[0] == 1000.0
+
+
+def test_simple_cost_model_adds_cost_and_slippage_without_price_adjustment():
+    model = SimpleProportionalCostModel(cost_bps=10, slippage_bps=5)
+
+    cost = model.compute(
+        traded_shares=np.array([1000, -1000]),
+        execution_price=np.array([20.0, 20.0]),
+        side=np.array([1, -1]),
+        date=dt.date(2024, 1, 2),
+    )
+
+    assert np.allclose(cost, np.array([30.0, 30.0]))
+
+
+def test_daily_pnl_cost_matches_fill_trade_cost_sum():
+    dates = pd.to_datetime(["2024-01-01", "2024-01-02"])
+    positions = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001"],
+        "date": [dates[0], dates[0]],
+        "portfolio_name": ["run1", "run1"],
+        "target_weight": [0.5, -0.5],
+        "target_value": [1000.0, -1000.0],
+        "target_shares": [100.0, -50.0],
+        "position_shares": [100, -50],
+        "position_value": [1000.0, -1000.0],
+        "price": [10.0, 20.0],
+    })
+    data = pd.DataFrame([
+        {
+            "symbol_id": sym,
+            "date": d,
+            "open": price,
+            "close": price,
+            "preclose": price,
+            "amount": 1e9,
+            "tradestatus": 1,
+            "isST": 0,
+        }
+        for d in dates
+        for sym, price in (("sh600000", 10.0), ("sz000001", 20.0))
+    ])
+
+    fills, pnl = ReferenceSimulator(cost_bps=10, slippage_bps=5).run(positions, data)
+
+    total_fill_cost = fills["trade_cost"].sum()
+    assert np.isclose(total_fill_cost, 3.0)
+    assert np.isclose(pnl["cost"].iloc[0], total_fill_cost)
+    assert np.isclose(pnl["pnl"].iloc[0], -total_fill_cost)
+
+
 # --- evaluate_portfolio ------------------------------------------------------
 
 def test_evaluate_portfolio_keys_no_ic():
@@ -470,3 +902,58 @@ def test_evaluate_portfolio_keys_no_ic():
         assert key in metrics
     assert "ic" not in metrics
     assert "rank_ic" not in metrics
+
+
+def test_evaluate_portfolio_excludes_signal_without_forward_return():
+    dates = pd.date_range("2024-01-01", periods=3)
+    data = pd.DataFrame([
+        {"symbol_id": sym, "date": d, "open": price, "close": price}
+        for d, prices in zip(dates, [(100.0, 100.0), (100.0, 100.0), (200.0, 100.0)])
+        for sym, price in zip(("sh600000", "sz000001"), prices)
+    ])
+    positions = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000", "sz000001"],
+        "date": [dates[0], dates[0], dates[1], dates[1]],
+        "portfolio_name": ["run1"] * 4,
+        "target_weight": [0.5, -0.5, -0.5, 0.5],
+        "target_value": [500.0, -500.0, -500.0, 500.0],
+        "target_shares": [5.0, -5.0, -2.5, 5.0],
+        "position_shares": [5, -5, -2, 5],
+        "position_value": [500.0, -500.0, -400.0, 500.0],
+        "price": [100.0, 100.0, 200.0, 100.0],
+    })
+
+    metrics = evaluate_portfolio(positions, data)
+
+    assert metrics["n_dates"] == 1
+
+
+def test_evaluate_portfolio_empty_and_single_return_paths():
+    empty_metrics = evaluate_portfolio(
+        pd.DataFrame(columns=POSITION_COLUMNS),
+        pd.DataFrame(columns=["symbol_id", "date", "open"]),
+    )
+    assert empty_metrics["n_dates"] == 0
+    assert empty_metrics["cumulative_return"] == 0.0
+
+    dates = pd.date_range("2024-01-01", periods=3)
+    data = pd.DataFrame([
+        {"symbol_id": "sh600000", "date": d, "open": price}
+        for d, price in zip(dates, [100.0, 100.0, 110.0])
+    ])
+    positions = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [dates[0]],
+        "portfolio_name": ["single"],
+        "target_weight": [1.0],
+        "target_value": [1000.0],
+        "target_shares": [10.0],
+        "position_shares": [10],
+        "position_value": [1000.0],
+        "price": [100.0],
+    })
+
+    single_metrics = evaluate_portfolio(positions, data)
+
+    assert single_metrics["n_dates"] == 1
+    assert single_metrics["cumulative_return"] == 0.0

tests/test_schema_failures.py

@@ -0,0 +1,98 @@
+"""Malformed parquet/input tests for phase boundary contracts."""
+
+from __future__ import annotations
+
+import pandas as pd
+import pytest
+
+from pipeline.alpha.compute import compute_alpha
+from pipeline.combo.combine import combine_alphas
+from pipeline.derived.compute import validate_derived_frame
+from pipeline.portfolio.construct import construct_positions
+from pipeline.portfolio.simulator import ReferenceSimulator
+from tests.helpers import make_generated_daily_bars
+
+
+def test_alpha_compute_rejects_daily_data_without_close():
+    daily = make_generated_daily_bars().drop(columns=["close"])
+
+    with pytest.raises(KeyError, match="close"):
+        compute_alpha(daily, "bad", "reversal", lookback=3)
+
+
+def test_alpha_feature_path_rejects_duplicate_symbol_dates(tmp_path):
+    daily = make_generated_daily_bars()
+    feature = pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000"],
+        "date": ["2024-01-02 09:30:00", "2024-01-02 15:00:00"],
+        "toy_feature": [1.0, 2.0],
+    })
+    feature_path = tmp_path / "duplicate_feature_keys.pq"
+    feature.to_parquet(feature_path, index=False)
+
+    with pytest.raises(ValueError, match="duplicate symbol_id,date"):
+        compute_alpha(
+            daily,
+            "bad_features",
+            "reversal",
+            lookback=3,
+            feature_paths=[str(feature_path)],
+        )
+
+
+def test_derived_validation_rejects_bool_value_columns():
+    derived = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [pd.Timestamp("2024-01-02")],
+        "is_good": [True],
+    })
+
+    with pytest.raises(ValueError, match="numeric"):
+        validate_derived_frame(derived)
+
+
+def test_combo_combine_rejects_missing_weight_column(tmp_path):
+    bad_alpha = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [pd.Timestamp("2024-01-02")],
+        "alpha_name": ["bad"],
+    })
+    bad_alpha_path = tmp_path / "bad_alpha.pq"
+    bad_alpha.to_parquet(bad_alpha_path, index=False)
+
+    with pytest.raises(KeyError, match="weight"):
+        combine_alphas([str(bad_alpha_path)], "bad_combo")
+
+
+def test_portfolio_build_rejects_weights_without_symbol_id():
+    daily = make_generated_daily_bars()
+    bad_weights = pd.DataFrame({
+        "date": [pd.Timestamp("2024-01-02")],
+        "combo_name": ["bad"],
+        "weight": [1.0],
+    })
+
+    with pytest.raises(KeyError, match="symbol_id"):
+        construct_positions(
+            bad_weights,
+            daily,
+            booksize=1_000_000.0,
+            portfolio_name="bad_portfolio",
+        )
+
+
+def test_portfolio_simulate_rejects_positions_without_position_shares():
+    daily = make_generated_daily_bars()
+    bad_positions = pd.DataFrame({
+        "symbol_id": ["sh600000"],
+        "date": [pd.Timestamp("2024-01-02")],
+        "portfolio_name": ["bad"],
+        "target_weight": [1.0],
+        "target_value": [1000.0],
+        "target_shares": [100.0],
+        "position_value": [1000.0],
+        "price": [10.0],
+    })
+
+    with pytest.raises(KeyError, match="position_shares"):
+        ReferenceSimulator().run(bad_positions, daily)

tests/test_universe.py

@@ -0,0 +1,92 @@
+"""Offline tests for baostock-backed universe helpers."""
+
+from __future__ import annotations
+
+import pandas as pd
+
+import data.universe as universe
+
+
+class _FakeResult:
+    def __init__(self, rows, fields=None):
+        self.rows = rows
+        self.fields = fields or ["code", "name", "date"]
+        self._idx = -1
+
+    def next(self):
+        self._idx += 1
+        return self._idx < len(self.rows)
+
+    def get_row_data(self):
+        return self.rows[self._idx]
+
+
+def test_index_constituent_helpers_normalize_dotted_codes(monkeypatch):
+    calls: list[str] = []
+
+    monkeypatch.setattr(universe.bs, "login", lambda: calls.append("login"))
+    monkeypatch.setattr(universe.bs, "logout", lambda: calls.append("logout"))
+    monkeypatch.setattr(
+        universe.bs,
+        "query_hs300_stocks",
+        lambda: _FakeResult([
+            ["sh.600000", "浦发银行", "2024-01-12"],
+            ["sz.000001", "平安银行", "2024-01-12"],
+        ]),
+    )
+    monkeypatch.setattr(
+        universe.bs,
+        "query_zz500_stocks",
+        lambda: _FakeResult([
+            ["sh.600006", "东风汽车", "2024-01-12"],
+        ]),
+    )
+
+    hs300 = universe.get_hs300_stocks()
+    zz500 = universe.get_zz500_stocks()
+
+    assert calls == ["login", "logout", "login", "logout"]
+    assert hs300["code"].tolist() == ["sh600000", "sz000001"]
+    assert zz500["code"].tolist() == ["sh600006"]
+    assert hs300["name"].tolist() == ["浦发银行", "平安银行"]
+
+
+def test_get_all_stocks_walks_back_and_filters_to_listed_a_shares(monkeypatch):
+    fields = ["code", "tradeStatus", "code_name"]
+    responses = [
+        _FakeResult([], fields=fields),
+        _FakeResult(
+            [
+                ["sh.600000", "1", "浦发银行"],
+                ["sh.688001", "1", "华兴源创"],
+                ["sz.000001", "1", "平安银行"],
+                ["sz.300750", "1", "宁德时代"],
+                ["sz.399001", "1", "深证成指"],
+                ["sz.200001", "1", "深物业B"],
+                ["bj.430047", "1", "北交所样本"],
+            ],
+            fields=fields,
+        ),
+    ]
+    query_days: list[str] = []
+
+    monkeypatch.setattr(universe.bs, "login", lambda: None)
+    monkeypatch.setattr(universe.bs, "logout", lambda: None)
+
+    def fake_query_all_stock(day):
+        query_days.append(day)
+        return responses.pop(0)
+
+    monkeypatch.setattr(universe.bs, "query_all_stock", fake_query_all_stock)
+
+    result = universe.get_all_stocks("2024-01-07")
+
+    assert query_days == ["2024-01-07", "2024-01-06"]
+    assert result.columns.tolist() == ["code", "name"]
+    assert result["code"].tolist() == [
+        "sh600000",
+        "sh688001",
+        "sz000001",
+        "sz300750",
+    ]
+    assert result["name"].tolist() == ["浦发银行", "华兴源创", "平安银行", "宁德时代"]

tests/test_workflow_sanity.py

@@ -0,0 +1,508 @@
+"""Verbose offline checks for the daily research workflow."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+from pipeline.alpha.compute import compute_alpha
+from pipeline.combo.combine import combine_alphas
+from pipeline.common.schema import (
+    ALPHA_COLUMNS,
+    COMBO_COLUMNS,
+    FILL_COLUMNS,
+    PNL_COLUMNS,
+    POSITION_COLUMNS,
+)
+from pipeline.portfolio.constraints import (
+    PriceLimitConstraint,
+    SuspensionConstraint,
+    VolumeCapConstraint,
+)
+from pipeline.portfolio.construct import construct_positions
+from pipeline.portfolio.research import evaluate_portfolio
+from pipeline.portfolio.simulator import ReferenceSimulator
+from tests.helpers import (
+    GENERATED_SYMBOLS,
+    generated_sessions,
+    make_generated_alpha_weights,
+    make_generated_combo_weights,
+    make_generated_daily_bars,
+)
+
+
+FIXTURE_PATH = Path(__file__).parent / "fixtures" / "daily_bars_real_2024_01_sample.pq"
+
+
+def _assert_sorted_by_symbol_date(frame: pd.DataFrame) -> None:
+    expected = frame.sort_values(["symbol_id", "date"]).reset_index(drop=True)
+    pd.testing.assert_frame_equal(frame.reset_index(drop=True), expected)
+
+
+def _assert_metric_dict_is_finite(metrics: dict[str, float]) -> None:
+    for key in (
+        "cumulative_return",
+        "sharpe_annual",
+        "turnover_annual",
+        "max_drawdown",
+        "hit_rate",
+        "n_dates",
+    ):
+        assert key in metrics
+        assert np.isfinite(metrics[key])
+    assert "ic" not in metrics
+    assert "rank_ic" not in metrics
+    assert "ir" not in metrics
+
+
+def test_tiny_workflow_golden_outputs_are_stable(tmp_path):
+    dates = pd.to_datetime(["2024-01-02", "2024-01-03", "2024-01-04"])
+    daily_bars = pd.DataFrame([
+        {
+            "symbol_id": "sh600000",
+            "symbol_name": "A",
+            "date": dates[0],
+            "open": 10.0,
+            "high": 10.0,
+            "low": 10.0,
+            "close": 10.0,
+            "preclose": 10.0,
+            "volume": 1_000_000.0,
+            "amount": 10_000_000.0,
+            "vwap": 10.0,
+            "turn": 1.0,
+            "pctChg": 0.0,
+            "tradestatus": 1,
+            "isST": 0,
+            "peTTM": 1.0,
+            "pbMRQ": 1.0,
+            "psTTM": 1.0,
+            "pcfNcfTTM": 1.0,
+        },
+        {
+            "symbol_id": "sz000001",
+            "symbol_name": "B",
+            "date": dates[0],
+            "open": 20.0,
+            "high": 20.0,
+            "low": 20.0,
+            "close": 20.0,
+            "preclose": 20.0,
+            "volume": 1_000_000.0,
+            "amount": 20_000_000.0,
+            "vwap": 20.0,
+            "turn": 1.0,
+            "pctChg": 0.0,
+            "tradestatus": 1,
+            "isST": 0,
+            "peTTM": 1.0,
+            "pbMRQ": 1.0,
+            "psTTM": 1.0,
+            "pcfNcfTTM": 1.0,
+        },
+        {
+            "symbol_id": "sh600000",
+            "symbol_name": "A",
+            "date": dates[1],
+            "open": 10.0,
+            "high": 12.0,
+            "low": 10.0,
+            "close": 12.0,
+            "preclose": 10.0,
+            "volume": 1_000_000.0,
+            "amount": 10_000_000.0,
+            "vwap": 10.0,
+            "turn": 1.0,
+            "pctChg": 20.0,
+            "tradestatus": 1,
+            "isST": 0,
+            "peTTM": 1.0,
+            "pbMRQ": 1.0,
+            "psTTM": 1.0,
+            "pcfNcfTTM": 1.0,
+        },
+        {
+            "symbol_id": "sz000001",
+            "symbol_name": "B",
+            "date": dates[1],
+            "open": 20.0,
+            "high": 20.0,
+            "low": 18.0,
+            "close": 18.0,
+            "preclose": 20.0,
+            "volume": 1_000_000.0,
+            "amount": 20_000_000.0,
+            "vwap": 20.0,
+            "turn": 1.0,
+            "pctChg": -10.0,
+            "tradestatus": 1,
+            "isST": 0,
+            "peTTM": 1.0,
+            "pbMRQ": 1.0,
+            "psTTM": 1.0,
+            "pcfNcfTTM": 1.0,
+        },
+        {
+            "symbol_id": "sh600000",
+            "symbol_name": "A",
+            "date": dates[2],
+            "open": 12.0,
+            "high": 13.0,
+            "low": 12.0,
+            "close": 13.0,
+            "preclose": 12.0,
+            "volume": 1_000_000.0,
+            "amount": 12_000_000.0,
+            "vwap": 12.0,
+            "turn": 1.0,
+            "pctChg": 8.33,
+            "tradestatus": 1,
+            "isST": 0,
+            "peTTM": 1.0,
+            "pbMRQ": 1.0,
+            "psTTM": 1.0,
+            "pcfNcfTTM": 1.0,
+        },
+        {
+            "symbol_id": "sz000001",
+            "symbol_name": "B",
+            "date": dates[2],
+            "open": 18.0,
+            "high": 21.0,
+            "low": 18.0,
+            "close": 21.0,
+            "preclose": 18.0,
+            "volume": 1_000_000.0,
+            "amount": 18_000_000.0,
+            "vwap": 18.0,
+            "turn": 1.0,
+            "pctChg": 16.67,
+            "tradestatus": 1,
+            "isST": 0,
+            "peTTM": 1.0,
+            "pbMRQ": 1.0,
+            "psTTM": 1.0,
+            "pcfNcfTTM": 1.0,
+        },
+    ])
+    alpha = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000", "sz000001"],
+        "date": [dates[0], dates[0], dates[1], dates[1]],
+        "alpha_name": ["gold_alpha"] * 4,
+        "weight": [1.0, -1.0, -1.0, 1.0],
+    })
+    alpha_path = tmp_path / "gold_alpha.pq"
+    alpha.to_parquet(alpha_path, index=False)
+
+    combo = combine_alphas([str(alpha_path)], "gold_combo")
+    positions = construct_positions(combo, daily_bars, booksize=20_000.0, portfolio_name="gold_port")
+    fills, pnl = ReferenceSimulator().run(positions, daily_bars)
+
+    expected_combo = pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000", "sz000001", "sz000001"],
+        "date": [dates[0], dates[1], dates[0], dates[1]],
+        "combo_name": ["gold_combo"] * 4,
+        "weight": [1.0, -1.0, -1.0, 1.0],
+    })
+    expected_positions = pd.DataFrame({
+        "symbol_id": ["sh600000", "sh600000", "sz000001", "sz000001"],
+        "date": [dates[0], dates[1], dates[0], dates[1]],
+        "portfolio_name": ["gold_port"] * 4,
+        "target_weight": [0.5, -0.5, -0.5, 0.5],
+        "target_value": [10000.0, -10000.0, -10000.0, 10000.0],
+        "target_shares": [1000.0, -10000.0 / 12.0, -500.0, 10000.0 / 18.0],
+        "position_shares": [1000, -833, -500, 556],
+        "position_value": [10000.0, -9996.0, -10000.0, 10008.0],
+        "price": [10.0, 12.0, 20.0, 18.0],
+    })
+    expected_fills = pd.DataFrame({
+        "symbol_id": ["sh600000", "sz000001", "sh600000", "sz000001"],
+        "date": [dates[1], dates[1], dates[2], dates[2]],
+        "portfolio_name": ["gold_port"] * 4,
+        "prev_shares": [0, 0, 1000, -500],
+        "target_shares": [1000, -500, -833, 556],
+        "traded_shares": [1000, -500, -1833, 1056],
+        "realized_shares": [1000, -500, -833, 556],
+        "blocked": [0, 0, 0, 0],
+        "trade_cost": [0.0, 0.0, 0.0, 0.0],
+    })
+    expected_pnl = pd.DataFrame({
+        "date": [dates[1], dates[2]],
+        "portfolio_name": ["gold_port", "gold_port"],
+        "gross_exposure": [21000.0, 22505.0],
+        "net_exposure": [3000.0, 847.0],
+        "pnl": [3000.0, 835.0],
+        "cost": [0.0, 0.0],
+        "turnover": [1.0, 2.0502],
+        "n_positions": [2, 2],
+    })
+
+    pd.testing.assert_frame_equal(combo, expected_combo)
+    pd.testing.assert_frame_equal(positions, expected_positions)
+    pd.testing.assert_frame_equal(fills, expected_fills)
+    pd.testing.assert_frame_equal(pnl, expected_pnl)
+
+
+def test_generated_alpha_combo_portfolio_execution_workflow(tmp_path):
+    daily_bars = make_generated_daily_bars()
+
+    computed_alpha = compute_alpha(
+        data=daily_bars,
+        alpha_name="generated_reversal_3d",
+        alpha_type="reversal",
+        lookback=3,
+    )
+
+    assert list(computed_alpha.columns) == ALPHA_COLUMNS
+    assert not computed_alpha.empty
+    assert set(computed_alpha["symbol_id"]).issubset(set(GENERATED_SYMBOLS))
+    assert computed_alpha["date"].min() > daily_bars["date"].min()
+    assert computed_alpha["weight"].notna().all()
+    assert computed_alpha["weight"].abs().sum() > 0.0
+    assert {"ic", "rank_ic", "ir"}.isdisjoint(computed_alpha.columns)
+    _assert_sorted_by_symbol_date(computed_alpha)
+
+    alpha_a = make_generated_alpha_weights("alpha_a", zero_date_index=2)
+    alpha_b = make_generated_alpha_weights(
+        "alpha_b",
+        scale=0.5,
+        offset=0.25,
+        zero_date_index=2,
+    )
+    alpha_a_path = tmp_path / "alpha_a.pq"
+    alpha_b_path = tmp_path / "alpha_b.pq"
+    alpha_a.to_parquet(alpha_a_path, index=False)
+    alpha_b.to_parquet(alpha_b_path, index=False)
+
+    identity_combo = combine_alphas([str(alpha_a_path)], "identity_combo")
+    assert list(identity_combo.columns) == COMBO_COLUMNS
+    assert (identity_combo["combo_name"] == "identity_combo").all()
+    pd.testing.assert_frame_equal(
+        identity_combo[["symbol_id", "date", "weight"]],
+        alpha_a[["symbol_id", "date", "weight"]],
+    )
+
+    equal_combo = combine_alphas([str(alpha_a_path), str(alpha_b_path)], "equal_combo")
+    expected_equal_weights = (
+        pd.concat([alpha_a, alpha_b], ignore_index=True)
+        .groupby(["symbol_id", "date"], as_index=False)["weight"]
+        .mean()
+        .sort_values(["symbol_id", "date"])
+        .reset_index(drop=True)
+    )
+    pd.testing.assert_frame_equal(
+        equal_combo[["symbol_id", "date", "weight"]],
+        expected_equal_weights,
+    )
+
+    portfolio_weights = make_generated_combo_weights("workflow_combo", zero_date_index=2)
+    positions = construct_positions(
+        weights_df=portfolio_weights,
+        data_df=daily_bars,
+        booksize=2_000_000.0,
+        portfolio_name="workflow_portfolio",
+    )
+
+    assert list(positions.columns) == POSITION_COLUMNS
+    assert not positions.empty
+    assert (positions["portfolio_name"] == "workflow_portfolio").all()
+    assert pd.api.types.is_integer_dtype(positions["position_shares"])
+    assert np.allclose(
+        positions["position_value"],
+        positions["position_shares"].astype(float) * positions["price"].fillna(0.0),
+    )
+
+    target_gross_by_date = positions.groupby("date")["target_weight"].apply(lambda s: s.abs().sum())
+    nonzero_target_dates = target_gross_by_date[target_gross_by_date > 0.0]
+    assert np.allclose(nonzero_target_dates, 1.0)
+
+    nonzero_share_counts = positions.loc[positions["position_shares"] != 0, "position_shares"].abs()
+    assert (nonzero_share_counts >= 100).all()
+
+    zero_gross_date = generated_sessions(10)[2]
+    previous_date = generated_sessions(10)[1]
+    zero_gross_positions = positions[positions["date"] == zero_gross_date].set_index("symbol_id")
+    previous_positions = positions[positions["date"] == previous_date].set_index("symbol_id")
+    common_symbols = zero_gross_positions.index.intersection(previous_positions.index)
+    assert not common_symbols.empty
+    assert (zero_gross_positions.loc[common_symbols, "target_weight"] == 0.0).all()
+    pd.testing.assert_series_equal(
+        zero_gross_positions.loc[common_symbols, "position_shares"],
+        previous_positions.loc[common_symbols, "position_shares"],
+        check_names=False,
+    )
+
+    simulator = ReferenceSimulator(
+        constraints=[
+            SuspensionConstraint(),
+            PriceLimitConstraint(),
+            VolumeCapConstraint(max_frac=0.02),
+        ],
+        cost_bps=5,
+        slippage_bps=5,
+    )
+    fills, pnl = simulator.run(positions, daily_bars)
+
+    assert list(fills.columns) == FILL_COLUMNS
+    assert list(pnl.columns) == PNL_COLUMNS
+    assert not fills.empty
+    assert not pnl.empty
+    assert (fills["realized_shares"] == fills["prev_shares"] + fills["traded_shares"]).all()
+    assert fills["blocked"].sum() > 0
+
+    fill_prices = fills.merge(
+        daily_bars[["symbol_id", "date", "open"]],
+        on=["symbol_id", "date"],
+        how="left",
+        validate="many_to_one",
+    )
+    expected_trade_cost = (
+        fill_prices["traded_shares"].abs()
+        * fill_prices["open"].fillna(0.0)
+        * 10
+        / 10_000
+    )
+    assert np.allclose(fill_prices["trade_cost"], expected_trade_cost)
+
+    cost_by_date = fills.groupby("date")["trade_cost"].sum()
+    assert np.allclose(
+        pnl.set_index("date")["cost"],
+        cost_by_date.reindex(pnl["date"], fill_value=0.0),
+    )
+
+    booksize_used_by_simulator = positions.groupby("date")["target_value"].apply(lambda s: s.abs().sum()).max()
+    traded_value_by_date = (
+        fill_prices.assign(traded_value=fill_prices["traded_shares"].abs() * fill_prices["open"])
+        .groupby("date")["traded_value"]
+        .sum()
+    )
+    assert np.allclose(
+        pnl.set_index("date")["turnover"],
+        traded_value_by_date.reindex(pnl["date"], fill_value=0.0) / booksize_used_by_simulator,
+    )
+
+    metrics = evaluate_portfolio(positions, daily_bars)
+    _assert_metric_dict_is_finite(metrics)
+
+
+def test_generated_workflow_outputs_keep_parquet_schema_contracts(tmp_path):
+    daily_bars = make_generated_daily_bars(n_sessions=10, include_missing=False)
+    alpha = compute_alpha(
+        data=daily_bars,
+        alpha_name="schema_reversal",
+        alpha_type="reversal",
+        lookback=3,
+    )
+    alpha_path = tmp_path / "schema_reversal.pq"
+    alpha.to_parquet(alpha_path, index=False)
+
+    combo = combine_alphas([str(alpha_path)], "schema_combo")
+    positions = construct_positions(
+        weights_df=combo,
+        data_df=daily_bars,
+        booksize=1_000_000.0,
+        portfolio_name="schema_portfolio",
+    )
+    fills, pnl = ReferenceSimulator(cost_bps=5, slippage_bps=5).run(positions, daily_bars)
+
+    assert list(alpha.columns) == ALPHA_COLUMNS
+    assert pd.api.types.is_object_dtype(alpha["symbol_id"])
+    assert pd.api.types.is_datetime64_any_dtype(alpha["date"])
+    assert pd.api.types.is_object_dtype(alpha["alpha_name"])
+    assert pd.api.types.is_float_dtype(alpha["weight"])
+    assert not alpha.isna().any().any()
+    assert np.isfinite(alpha["weight"]).all()
+
+    assert list(combo.columns) == COMBO_COLUMNS
+    assert pd.api.types.is_object_dtype(combo["symbol_id"])
+    assert pd.api.types.is_datetime64_any_dtype(combo["date"])
+    assert pd.api.types.is_object_dtype(combo["combo_name"])
+    assert pd.api.types.is_float_dtype(combo["weight"])
+    assert not combo.isna().any().any()
+    assert np.isfinite(combo["weight"]).all()
+
+    assert list(positions.columns) == POSITION_COLUMNS
+    assert pd.api.types.is_integer_dtype(positions["position_shares"])
+    assert pd.api.types.is_datetime64_any_dtype(positions["date"])
+    assert not positions.isna().any().any()
+    position_numeric_columns = [
+        "target_weight",
+        "target_value",
+        "target_shares",
+        "position_value",
+        "price",
+    ]
+    assert np.isfinite(positions[position_numeric_columns]).all().all()
+
+    assert list(fills.columns) == FILL_COLUMNS
+    assert pd.api.types.is_integer_dtype(fills["prev_shares"])
+    assert pd.api.types.is_integer_dtype(fills["target_shares"])
+    assert pd.api.types.is_integer_dtype(fills["traded_shares"])
+    assert pd.api.types.is_integer_dtype(fills["realized_shares"])
+    assert pd.api.types.is_integer_dtype(fills["blocked"])
+    assert not fills.isna().any().any()
+    assert np.isfinite(fills["trade_cost"]).all()
+
+    assert list(pnl.columns) == PNL_COLUMNS
+    assert pd.api.types.is_integer_dtype(pnl["n_positions"])
+    assert not pnl.isna().any().any()
+    pnl_numeric_columns = [
+        "gross_exposure",
+        "net_exposure",
+        "pnl",
+        "cost",
+        "turnover",
+    ]
+    assert np.isfinite(pnl[pnl_numeric_columns]).all().all()
+
+
+def test_frozen_real_fixture_runs_high_level_workflow(tmp_path):
+    real_daily_bars = pd.read_parquet(FIXTURE_PATH)
+
+    assert real_daily_bars.shape == (36, 19)
+    assert set(real_daily_bars["symbol_id"]) == set(GENERATED_SYMBOLS)
+    assert real_daily_bars["date"].min() == pd.Timestamp("2024-01-02")
+    assert real_daily_bars["date"].max() == pd.Timestamp("2024-01-12")
+    assert real_daily_bars.groupby("date")["symbol_id"].nunique().eq(4).all()
+
+    reversal_alpha = compute_alpha(
+        data=real_daily_bars,
+        alpha_name="real_reversal_3d",
+        alpha_type="reversal",
+        lookback=3,
+    )
+    reversal_vol_alpha = compute_alpha(
+        data=real_daily_bars,
+        alpha_name="real_reversal_vol_3d",
+        alpha_type="reversal_vol",
+        lookback=3,
+        vol_window=3,
+    )
+
+    reversal_path = tmp_path / "real_reversal.pq"
+    reversal_vol_path = tmp_path / "real_reversal_vol.pq"
+    reversal_alpha.to_parquet(reversal_path, index=False)
+    reversal_vol_alpha.to_parquet(reversal_vol_path, index=False)
+
+    combo = combine_alphas([str(reversal_path), str(reversal_vol_path)], "real_equal_combo")
+    positions = construct_positions(
+        weights_df=combo,
+        data_df=real_daily_bars,
+        booksize=1_000_000.0,
+        portfolio_name="real_fixture_portfolio",
+    )
+    fills, pnl = ReferenceSimulator(cost_bps=5, slippage_bps=5).run(positions, real_daily_bars)
+    metrics = evaluate_portfolio(positions, real_daily_bars)
+
+    assert not reversal_alpha.empty
+    assert not reversal_vol_alpha.empty
+    assert not combo.empty
+    assert not positions.empty
+    assert not fills.empty
+    assert not pnl.empty
+    assert np.isfinite(combo["weight"]).all()
+    assert np.isfinite(positions["target_weight"]).all()
+    assert np.isfinite(pnl[["gross_exposure", "net_exposure", "pnl", "cost", "turnover"]]).all().all()
+    _assert_metric_dict_is_finite(metrics)

uv.lock

@@ -284,7 +284,6 @@ version = "0.1.0"
 source = { virtual = "." }
 dependencies = [
     { name = "akshare" },
-    { name = "backtrader" },
     { name = "baostock" },
     { name = "click" },
     { name = "matplotlib" },
@@ -293,24 +292,38 @@ dependencies = [
     { name = "pyarrow" },
 ]
 
+[package.optional-dependencies]
+backtrader = [
+    { name = "backtrader" },
+]
+joinquant-browser = [
+    { name = "playwright" },
+]
+
 [package.dev-dependencies]
 dev = [
+    { name = "coverage" },
     { name = "pytest" },
 ]
 
 [package.metadata]
 requires-dist = [
     { name = "akshare", specifier = ">=1.14.0" },
-    { name = "backtrader", specifier = ">=1.9.76.123" },
+    { name = "backtrader", marker = "extra == 'backtrader'", specifier = ">=1.9.76.123" },
     { name = "baostock", specifier = ">=0.8.8" },
     { name = "click", specifier = ">=8.0.0" },
     { name = "matplotlib", specifier = ">=3.7.0" },
     { name = "pandas", specifier = ">=2.0.0" },
+    { name = "playwright", marker = "extra == 'joinquant-browser'", specifier = ">=1.61.0" },
     { name = "pyarrow", specifier = ">=14.0.0" },
 ]
+provides-extras = ["backtrader", "joinquant-browser"]
 
 [package.metadata.requires-dev]
-dev = [{ name = "pytest", specifier = ">=7.0.0" }]
+dev = [
+    { name = "coverage", specifier = ">=7.14.1" },
+    { name = "pytest", specifier = ">=7.0.0" },
+]
 
 [[package]]
 name = "click"
@@ -493,6 +506,119 @@ wheels = [
     { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/0c/58/bd257695f39d05594ca4ad60df5bcb7e32247f9951fd09a9b8edb82d1daa/contourpy-1.3.3-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:3d1a3799d62d45c18bafd41c5fa05120b96a28079f2393af559b843d1a966a77", size = 225315, upload-time = "2025-07-26T12:02:58.801Z" },
 ]
 
+[[package]]
+name = "coverage"
+version = "7.14.1"
+source = { registry = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple/" }
+sdist = { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/54/fd/0ab2772530e946e1be1abd0bc09e647ec9b02e88f0867857601fefca8953/coverage-7.14.1.tar.gz", hash = "sha256:30c08f7d90415aa98b3c990385dea2939b0da55f38515e5b369b83655f8523be", size = 920132, upload-time = "2026-05-26T20:41:36.783Z" }
+wheels = [
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/92/69/0d2ef01ff4b8fcecd4cba920d11e92fa4f96ae412441d3b56a90a258e69b/coverage-7.14.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:3e3680291c4a1d0dadfa84a2c459576a4af5133abb617905714339a0c73138cf", size = 219722, upload-time = "2026-05-26T20:38:14.002Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f8/ae/9afdeaa31b9d9ce98124b6abf8bb49119bf71aecae04f8567c189d91299f/coverage-7.14.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a5274669f37f2343635a347b91a60777621341ab3378e9c6ac9335eee704bddf", size = 220240, upload-time = "2026-05-26T20:38:17.424Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/51/69/c998589871df7ea7dba865cc5ee32b5a3e1d47ba6c68ef91104c7c46fa5e/coverage-7.14.1-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:cfe5a5fec635799ef33428f1e5e61bafa45a92a96190ba731561ba558ccc214d", size = 246981, upload-time = "2026-05-26T20:38:19.266Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fc/10/1c7d04c13040dac531d21b712bbe08f902e6dd9b58f5d77875c4d030f8f2/coverage-7.14.1-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:62a9f70b52e0b5a95cfef4a5c5641b06983cadc5e538a3feeb5c00211f523ac2", size = 248812, upload-time = "2026-05-26T20:38:20.75Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c1/65/2a38a4607ef27cadcfbcee034dba5830ae2569f90144a0f4c7dbf47d30b0/coverage-7.14.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3c18ebc343e15be53049b3a2dce38fe82d58f37e20ab9094b3a39c0aa4f6bb47", size = 250675, upload-time = "2026-05-26T20:38:22.159Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c9/a2/a446ed9752a4a59b79e0fb6cbb319f6facb2183045c0725462625e66f87e/coverage-7.14.1-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b84ffdf877644e7096aa936991efeed873f7f3df57b9cd001312b7668ab08550", size = 252590, upload-time = "2026-05-26T20:38:23.63Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/9e/fd/e81fbd7ba752365546e9842b1cbdaad3d6919d2a522c590aef16a281ec5e/coverage-7.14.1-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:e854312c4103f2ad4c0dc023b69b77ebfd2c89db5f86c4c94dc2353f9a92167e", size = 247691, upload-time = "2026-05-26T20:38:25.057Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/53/35/f3c26fdaae9ea937d154ca4d372e5ea0a4167ff70d36c6074ac2eacb2f83/coverage-7.14.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:c643734307300234fafa36bf2a040a7235f8f177ea1fd6ec1423aea6fb7b929f", size = 248716, upload-time = "2026-05-26T20:38:26.406Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/2e/14/940b6c49551fd343e8507ee2b0ba7af5d0aa04ed5bf768285cb7c72a9884/coverage-7.14.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:84ac9499e48700399a5dd0ea7085b5091961fec52c68d66b4ec0d3cf7f4441b1", size = 246721, upload-time = "2026-05-26T20:38:28.282Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/aa/2c/40fc0634186c28292a662dff578866b3913983d6c375a3c2a74020938719/coverage-7.14.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:7f02d09f70776579b926d889a4c9c235070a1f47c40458aeaca563fae5acfdb5", size = 250533, upload-time = "2026-05-26T20:38:29.753Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/de/e3/2c26bf1e811f9df991ff2a9bdddebdd13ee0665d564df7d05979f9146297/coverage-7.14.1-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:ce66d8e46da2bb5ee313a745cbd2e391d319176c1f7a9451bfcd3a2fb920859b", size = 246990, upload-time = "2026-05-26T20:38:31.516Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a8/b0/060260ef56bd92363ebdce0c7095ce422b06e69aae71828efeca473ab1ca/coverage-7.14.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:c912c259304cfb5ee584481cfb7ce1ff932b4d61e6c9140b8f19cb7b5ed82332", size = 247593, upload-time = "2026-05-26T20:38:33.065Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/63/f3/501502046efeb0d6d94b5ca54941d95f1184183dd6bdb7f283985783bb4a/coverage-7.14.1-cp310-cp310-win32.whl", hash = "sha256:1238cb94638e610e972c60dac68e813f868dc7d6e982535270558443058d9d59", size = 222330, upload-time = "2026-05-26T20:38:35.36Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a0/5d/1bf99f2c558f128faf7906817ccbdb576ba815d3b41ce2ac1719b70a3663/coverage-7.14.1-cp310-cp310-win_amd64.whl", hash = "sha256:fc459e5d73be2d6332fcfe8dbf3d8994671fe33c700f4565988ecfa511547253", size = 223261, upload-time = "2026-05-26T20:38:37.196Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7d/d7/477ad149490e6cb849f28abea1dabb9c823cea72e7500c81b4240ce619c0/coverage-7.14.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:478b5bcd63c2e1357c5c7e16c070690df7b07f676b1c114d7b93e533c664309f", size = 219848, upload-time = "2026-05-26T20:38:38.715Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/91/82/a5eb47257c50601bb7b9a9d2857c67b7a3a85ad74180eb2c98bb1fbe0ce5/coverage-7.14.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a24a81f9715ee42ef59a316cc11611c98fe23920f7c81861315c9f3ff4a230f4", size = 220354, upload-time = "2026-05-26T20:38:40.232Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/43/8b/78419b5391a5cb706b6544390507e469d83ffc9a8248b02c4011aceb9365/coverage-7.14.1-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:196a13319ad88d6d8ef5ab489ec4f44ddde2143c0c7d5b27786f6c3ffd56a7e1", size = 250771, upload-time = "2026-05-26T20:38:41.782Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/77/63/e77aaacd491182210d639636b7a8bba23ffffa9b82aa3762da9431855fa9/coverage-7.14.1-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:3d452fd08b5c72c5167c93e6867b5c08500bd40f2a21e1e854a500550b6cc36f", size = 252683, upload-time = "2026-05-26T20:38:43.305Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/65/1c/a022e3cfbec2ac241640003cb3a817e161d9c7f5aa9b49173756cdc03204/coverage-7.14.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:23bf7fa51ac02e07fc7c96849b82946da47ae862dc8f86d183b2a4864fc38129", size = 254791, upload-time = "2026-05-26T20:38:45.361Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/61/d6/967e408aca4c1ceb88cb0cc677169110ae7f5995fb5eaf5fb1f5a1bb8f5d/coverage-7.14.1-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:bcaa50684dcaadfa599ac48f81103c756d791cfd85c97203d2217c593d48b860", size = 256748, upload-time = "2026-05-26T20:38:46.91Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b8/be/869188f7fe28638078ec479331ace6dc5f7b40b7153eb616f47ab79404d8/coverage-7.14.1-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:4ea1c034f95c9b056e856b794630b17f9fa3d57e4800ff1e503d3be0f9c9078c", size = 250907, upload-time = "2026-05-26T20:38:48.493Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/07/aa/adb7d3b4278d690e68703abcd76ab1b948242e3668d921711551b78f9ddb/coverage-7.14.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c7e057326434e441306226fbeb5d1aaf14a2637efe97ba668306635835f32ad7", size = 252483, upload-time = "2026-05-26T20:38:50.074Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/43/61/331c74103c62dcb0c4b9b3a0de9a61aca016208b0a90f109592a9f9ecc28/coverage-7.14.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:59baf88468dbc8d63b1887afd92bda52e40bb1561696e5819670601403810cec", size = 250545, upload-time = "2026-05-26T20:38:51.613Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f6/b6/c5dae3c104d89be04828f61810e6b3473825482e4c288cc4ed04553e08ae/coverage-7.14.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:d34d75f892b3ab73ba11cab5442cce7b3e168fd64162b16f0e1e0d09c508edef", size = 254310, upload-time = "2026-05-26T20:38:53.503Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ad/a1/2b9d5863e3b83c01ad8199e3c597802fbb3a9dc90b058885804c20296d31/coverage-7.14.1-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:3a56abc20a472baf0304c455721bc601477440d28ecfde8a03dde79ede07e0df", size = 250266, upload-time = "2026-05-26T20:38:55.414Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7f/5e/0e511fbdb269359be26fe678a1c3fa1f2aa2a01573cc3f54268c8d6d4797/coverage-7.14.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:6a3cb83d1552c0cd1b4906655b6a33fd4a8473229633a901c6b73bf86914dee9", size = 251174, upload-time = "2026-05-26T20:38:57.141Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/85/10/e55307b622b3dd9671cb321824502dc10f93e72f2802b9946159a8edadeb/coverage-7.14.1-cp311-cp311-win32.whl", hash = "sha256:10274a1fbeb8ec5d72966e17bb198a3104257aca4ac09d98667c5f8aca8c8548", size = 222354, upload-time = "2026-05-26T20:38:58.727Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/71/cf/107421693cfb71e4f1ca5bf70443f64d4161878068d07a3e51c7ad21d17b/coverage-7.14.1-cp311-cp311-win_amd64.whl", hash = "sha256:87ebdf787d4888e3f3f2d523eadc6e18c6d18c6d0eb173801a189641627fb37e", size = 223290, upload-time = "2026-05-26T20:39:00.413Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b8/1d/3e3644585eb29e9dafefb19555078529a4d7cce12bd21929664eea989277/coverage-7.14.1-cp311-cp311-win_arm64.whl", hash = "sha256:dd34767fa19848d35659ffc0a75314f58c7af3f1cd87ec521e8292a1238398a3", size = 221953, upload-time = "2026-05-26T20:39:02.159Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/3d/b7/bdbb725ba02c5b42825b200c940f38b7a54fcad24627b7192f78f8110d76/coverage-7.14.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:a06c76364a9360e33d6d23769aefdf7f66f38e2ffb60ceb1baaa4989d83b695c", size = 220022, upload-time = "2026-05-26T20:39:03.702Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/72/81/fdc0898a55c6219223291ec1a1fe89966ef212ce82276aa0899df84b5de0/coverage-7.14.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:fad54e871165f6ec2f536063ac74c3104508a12963e64072ba44bd822de52b0c", size = 220379, upload-time = "2026-05-26T20:39:05.381Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/de/72/de048c4a25e13bce59ac6a339351c10bdf2515e07459afcdaf04dc3143a2/coverage-7.14.1-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:84b535f00655ecafe1d929d1fb00ed5d6fa3051ea643ab2c161a3887b86f294b", size = 251888, upload-time = "2026-05-26T20:39:07.367Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/28/30/300c343f68beb9d4cbb64ec81e58c5b6b80b56927f72d2b38654ac26e013/coverage-7.14.1-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:6b6b0853b895fe0e98cbfc580d1ec3393d9302b4b1e96a77b3f5c91fdab899e6", size = 254624, upload-time = "2026-05-26T20:39:09.037Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b1/ed/7b25642496e8170b6bac14adce00537c6e5fa2d586159401a4de3e8b49e6/coverage-7.14.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:442cc9c952b2df400cda54bb04ab87330cf2cd08a8692cbbea36773531eb6f37", size = 255739, upload-time = "2026-05-26T20:39:10.889Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7f/a2/abd210b8c4e29c24e4624916db97bb519097a91034aaeb767f937e7da794/coverage-7.14.1-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8270544c361ed405a27a060dbc9ed2c124b084d96dfdc2d9a2510482aef981ad", size = 257998, upload-time = "2026-05-26T20:39:12.722Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7f/24/7c50beed3792fe62f6ce0545c6686ce83379719e2c0276179333d97eae92/coverage-7.14.1-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:48b283b1dd6372e8de2a7a9a4c4d5dc06f4d4fd209b876f3c88a7a205a0c8f84", size = 252296, upload-time = "2026-05-26T20:39:14.259Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/15/05/0f874628ebcbfc77ead559ff210281ef06a97db08481832e7dd39274a135/coverage-7.14.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5b0c99ba93a07d56f6df340bb79be53202a082b2fdb81bfe6190b741a3470d54", size = 253658, upload-time = "2026-05-26T20:39:15.923Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/99/6f/ca6ad067364b337ef997802115e7ecad2abd2248b05471464b0dea02b4d4/coverage-7.14.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:e471bc5769ff073b058cfadb0d736b56ce067c8560eabeb0da88462df98c23e7", size = 251803, upload-time = "2026-05-26T20:39:17.537Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c0/30/b9b4d377cd9f40baf228068f5a81faf8450c6228503011bd499708483a50/coverage-7.14.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:f497a1ea81d4cd7c10ddcaa685135b9aabd291af3d55775a9ddf3cb7a364cdd9", size = 255873, upload-time = "2026-05-26T20:39:19.414Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/3c/21/7c721a9e5e6bb88547d30a787aefb97512d3f54c1324c7488d9b3743f7f9/coverage-7.14.1-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:2222be86d0b54f5dd5a38f45f17f315f737245e857bf0bdedc70734f84a13c02", size = 251372, upload-time = "2026-05-26T20:39:21.169Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/9d/8c/f8ae5a2200130e1503cd7661a6cd3b2b7bacef98277fbf3571fb13f8b766/coverage-7.14.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:85e85586565842f6932abebd4c18bcb1074223dc0b3576e7d173ca710622813a", size = 253245, upload-time = "2026-05-26T20:39:23.097Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/34/62/70a9024672a5f6910517d9628c52c9afbdd3cf8f46426af52bb148a56fff/coverage-7.14.1-cp312-cp312-win32.whl", hash = "sha256:4a28fd227808366b196a75476dced2eb35b351d6766ba9c858dc93319e87f4f1", size = 222567, upload-time = "2026-05-26T20:39:24.868Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f6/81/8b7cd386839b039ebe1855733b9f9449a8dec5d79564018234f185a7fa70/coverage-7.14.1-cp312-cp312-win_amd64.whl", hash = "sha256:54acdb6674a4661768d7bf7db32dfb9f46ab1d764f8aba6df75ce1a6a088724e", size = 223372, upload-time = "2026-05-26T20:39:26.603Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ae/ba/b44d472022f620d289d95fa830143235c0c36461c6f2437ea8d51e5481ed/coverage-7.14.1-cp312-cp312-win_arm64.whl", hash = "sha256:99cd41ff91afd94896fea3bc002706b6ae4ce95727d06e4a0f39c0a8d8bd8b1a", size = 221989, upload-time = "2026-05-26T20:39:28.242Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8a/9e/5f6d56327c62b185225d145191c607e07515294a0aa6338e58805cd4a5ac/coverage-7.14.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:be9f2c802dcfce3f71298303aa5dad0dce440a76c52f2f60dacd8656dab78793", size = 220044, upload-time = "2026-05-26T20:39:29.902Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/75/92/e82aca356744cbbc0f77a0b623e38918c1872361963413a3bab5d0340393/coverage-7.14.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6223a72fd0e4c7156353ec0f08a5f93623e1d3034d0e2683b9bb8ea674131b1d", size = 220412, upload-time = "2026-05-26T20:39:31.561Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/27/c9/385bde0bf7ed0f4bf3a7ee5367060a86b5d218718cfd6fb943c0f836b34f/coverage-7.14.1-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:7279d2110a28cebc738b6459ecda2771735a4c18465fbbd36b3288fe5ed92247", size = 251412, upload-time = "2026-05-26T20:39:33.337Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/51/8c/23faf6a2343a0d17f960a4bd56c43bc7eb4cf312f774dd6ceebd82c7d8fc/coverage-7.14.1-cp313-cp313-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:9eeb3fcbc13ba40dfbdb22d01d196a28e9cef9ed4c29b60061a1e0e823a9929d", size = 254008, upload-time = "2026-05-26T20:39:35.009Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/42/06/36f4aa9ca8a815e6036156e80706a67828bb97bd826948244f6996dda957/coverage-7.14.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5f0cfc27c539f07cf5c0a4cfe211d0b6cae039f8f40526dbaa71944e64b50a7b", size = 255241, upload-time = "2026-05-26T20:39:36.71Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ca/79/95266316352f90f6b1c6736bb413302edfde2453fb32422d3911642691b3/coverage-7.14.1-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:221c70f316241a78e77e607c227cefc8808d4e08f28d99c04f35694690e940be", size = 257373, upload-time = "2026-05-26T20:39:38.412Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e3/9c/58316d1f66c488b5fca8a0eb3e98348807813efa8a0d0833b9021be27488/coverage-7.14.1-cp313-cp313-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:da028256b04ec30e5e0114b6f76172938c313991f0a2d3d894271315cf5d5e43", size = 251635, upload-time = "2026-05-26T20:39:40.268Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ef/5a/ca2398a568e16fed7bb713e84ba3603a7164fb65779abe645c565ec890d5/coverage-7.14.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:76a085d7005236a767e3426148b2c407e53ad61695c562f8a81da2d373324901", size = 253373, upload-time = "2026-05-26T20:39:42.145Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/6e/2c/0396562c32deaebe7be51d865b3a41e9a87d7561acafe1a28f53b07e019a/coverage-7.14.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b553d04b5e778a8e56d57eb134aff42a92718ecba45e79c4764ecfa40efd92ff", size = 251341, upload-time = "2026-05-26T20:39:43.907Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fd/8f/a94f9221184c9cae1ee115820e3798e48b6b17777a9f19e46fb9a0c8dc74/coverage-7.14.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:46f714d2fb8ae2f4f29f23ada7f1e79b759fff5a70f94a1dac23af204c3ec9e4", size = 255497, upload-time = "2026-05-26T20:39:46.166Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/71/69/505d70e47db1eaebcd002c39759707621ef184cd6b1ae084d9f41293f323/coverage-7.14.1-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:1896f5e19ff3f0431c7ce2172adc54890fd97f86b59ced8ca1649145d9ffe35d", size = 251159, upload-time = "2026-05-26T20:39:48.03Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e0/aa/58681c383aa33a9d2ed40a02d7a22fbf780d1fa4d575396365777828198c/coverage-7.14.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:62fd185ef9df3c33d1c8178c5af105f762afbad96038de9a4ae100aa6297ca33", size = 252934, upload-time = "2026-05-26T20:39:49.872Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/eb/fd/11c928cd6bdffc7074bb5965c173d9ebf517fb00205e1da524b98d29ef92/coverage-7.14.1-cp313-cp313-win32.whl", hash = "sha256:ab4af6352741a604c431c6072fce5bee33bf0f20dc7a56618d6bf6bb89e9810c", size = 222584, upload-time = "2026-05-26T20:39:51.68Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/6f/92/fb416fc26d340dcba19518c418d6048e913186e17243982c5e435e41fa7a/coverage-7.14.1-cp313-cp313-win_amd64.whl", hash = "sha256:7af486dabe8954d03b087f0021540897afe084f04e16ff5579e08cc46f871416", size = 223394, upload-time = "2026-05-26T20:39:53.472Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/73/c6/02d56e3867972f77d5036de924643f26c056e848f00452cafb4dbc3c29b4/coverage-7.14.1-cp313-cp313-win_arm64.whl", hash = "sha256:2224f89ffd0c5605ccce1ed7a584da162bc7c55f601ab1c946bc9de31a486b42", size = 222015, upload-time = "2026-05-26T20:39:55.374Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/4d/9e/fcc77914050df73f7662fa1f00902774c79c075a8388ab334074574bf77e/coverage-7.14.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:de286598cc65d2b489411174b1faec2f5a7775fb3201fd925db2a76b4030f37d", size = 220733, upload-time = "2026-05-26T20:39:57.189Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f7/67/2963cbdaf5cbadec44efa3a1e39eaa1f02df4079585f05387607a221e126/coverage-7.14.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:042c46ded7c288aeb07cf14a28b6c1e10b78fcba40171c3fa1e939377eeef0b5", size = 221086, upload-time = "2026-05-26T20:39:59.019Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c8/c5/8701645574e11881f2f47d8930f98bc48b5d43b25eb5b4430dfc4a2f9f48/coverage-7.14.1-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:f4ddbe407477f04c45115d1a4e5bc480f753553b534d338d4c3358b1cdd0ea52", size = 262381, upload-time = "2026-05-26T20:40:00.822Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7c/28/7a64d73598263e0c5abd5084211a8474488d31b3c552ff531c719dfcff62/coverage-7.14.1-cp313-cp313t-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:d13e6725992e2d2fd7d81d4f5241952d13740121dfd501da09201be39b2c003a", size = 264458, upload-time = "2026-05-26T20:40:02.506Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fa/d8/4969179db9f7eb4df218e69540adf829d1c835f59452513d065d15446802/coverage-7.14.1-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f747dc8edcfe740130f28f32f3995e955494285717e86ee25af51db2219df08a", size = 266884, upload-time = "2026-05-26T20:40:04.421Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a6/78/a45d5794dbc9bafd97afc96a4377c86c7820d78b6cf51b89bc1d4e919275/coverage-7.14.1-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ced2f09ef276fd58611a1ef502164ad266d2b75174e5a40cabbdb4033f9f6cf2", size = 268022, upload-time = "2026-05-26T20:40:06.298Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/21/cb/4f5e354e9e3e67af96bd4e57113e6db6b22298c7168b13eec408a549903d/coverage-7.14.1-cp313-cp313t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:b84800013769a78ccb9ef4659402e26d06867e337b61ec365f77ad008adea80e", size = 261631, upload-time = "2026-05-26T20:40:08.226Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ec/49/eced49af4cb996d5d8b7e94e736175c513e4facd3398507b89892b4326d8/coverage-7.14.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:ea8cd6ca0ee9f616aaef3afc6882e32c2cbf18b00d96313ffd76af650574034d", size = 264443, upload-time = "2026-05-26T20:40:10.137Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f1/d8/5603a88a7c5913a6b54f6cb1a8c46f7b39cbb30f27cd3f492908da09b2d7/coverage-7.14.1-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:aa5e304a873fabddc11e484e9b6b738bd38bd7bed17b09aa84eecf5332e8b8bb", size = 262069, upload-time = "2026-05-26T20:40:11.999Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f0/59/2ae3cb79da554a06c8619d6c88ea19dd1e4aed4b834b6a83bb1fa243bdc5/coverage-7.14.1-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:5a1c5215be81035e629d5bc756650634d0bf31991038db7a0eccb90f025ce16d", size = 265780, upload-time = "2026-05-26T20:40:13.858Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/af/5f/b130c1dc999031f2648bd25317fbce505ad8d5562079b4ed81e736a84967/coverage-7.14.1-cp313-cp313t-musllinux_1_2_riscv64.whl", hash = "sha256:79058c47dae6788504b5effb319961bcd72d7240551464b91d474bc0ed186d69", size = 260970, upload-time = "2026-05-26T20:40:16.142Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/87/d1/ec13ccddeb48ec963bdfa72a11224bac2584bd045ba13beca82f8113e9c7/coverage-7.14.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:370c5afae3fa0658e11694a32b24c2778f6bc2d17718121f94ee185e69f26b54", size = 263157, upload-time = "2026-05-26T20:40:18.382Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/cf/c2/cd91ead503045161092d3845f7bb95ea2f25131ce96d3e314dd835d91b9c/coverage-7.14.1-cp313-cp313t-win32.whl", hash = "sha256:3758dd0a7f1fa57365ef2e781df0f0731d38b6e3772259d13dae4bd8a958d4b1", size = 223259, upload-time = "2026-05-26T20:40:20.381Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/71/9f/1e28d97e6bd2c76b07f38b7c02870f1371255ff6717f54eca578fcbbdd0e/coverage-7.14.1-cp313-cp313t-win_amd64.whl", hash = "sha256:6ff665fb023a77386fe11685190cee1f60a7d635994a30d9b0a061533d470fce", size = 224320, upload-time = "2026-05-26T20:40:22.316Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a9/e0/d936e908f0e1efa55e52b91e01b52f1055cef5e1ab2718493390ed8e2fb8/coverage-7.14.1-cp313-cp313t-win_arm64.whl", hash = "sha256:17a5a241e5997621a956a7f402a7433ef4221e5152809b785bec79e2323799f1", size = 222577, upload-time = "2026-05-26T20:40:24.894Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/d6/34/fc2f101b151af3799a101f0550b0454aa008afdc0add677394ec4aa8ea10/coverage-7.14.1-cp314-cp314-macosx_10_15_x86_64.whl", hash = "sha256:d5ed429d0b8edaac649e889b4ffcedb6c80b06629a3f93050e3dddfb99235bee", size = 220091, upload-time = "2026-05-26T20:40:27.249Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/3d/a7/1ebae2ab5b961b5c79bb09fe7b3ac99edb190d8be4a8c510b2cf66f46468/coverage-7.14.1-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:8011224a62280e50dab346960c03cf47aca1a1e09e608c0fb33fd6e0cc8e9500", size = 220421, upload-time = "2026-05-26T20:40:30.084Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/5e/90/92aca9cf0acc95123c96cd1eb1f08917897a7f5dee01e15738922971ec31/coverage-7.14.1-cp314-cp314-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:12c42ec1e14f553c4f817e989365982e646e27211f10a0f717855b94a79c8906", size = 251466, upload-time = "2026-05-26T20:40:32.542Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/26/2b/78048cbe3b999f6cbf9cc0d90abba6a88a3e0863a8c1c6cbc762f3f8802f/coverage-7.14.1-cp314-cp314-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:06144cd511cf2624873a035c5069cf297144f6e77a73ee3d7a55b605ec5efb42", size = 253973, upload-time = "2026-05-26T20:40:34.473Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8e/21/c2e33b29d1cfde484a19d437afc343c6cd30b08d78cbbf9f5aff14e57b2b/coverage-7.14.1-cp314-cp314-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a311d8e1da24be5c1ccf85cbfb06315dbaa1703d5a1eab3f6432c72b837917c8", size = 255318, upload-time = "2026-05-26T20:40:38.154Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8e/ee/aad2f108d63b769121005302f16bf66db8625c88ceaba466942e09a2607e/coverage-7.14.1-cp314-cp314-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c79cead5b5bc584d9c71451cb984d0e3a84e0c0937379c8efcbf27c8d661b851", size = 257633, upload-time = "2026-05-26T20:40:40.164Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c2/f8/11a2c29b4fd76d9849f81d0bb812ec0017a9396df3217214e38934a8c837/coverage-7.14.1-cp314-cp314-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:dcbf65f1f66a26cdd88c35cf68fb4729c5d1cd2e88added72420541dfb212034", size = 251488, upload-time = "2026-05-26T20:40:42.631Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c9/b8/9a5820de4b8ac2b71d85e3b5fb49108d7469c665f0e2ad0dd7569023e305/coverage-7.14.1-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:fd86572566fb40189a8260446158235159bc7a82dfbc87a3b39cf4fb57fcec1c", size = 253329, upload-time = "2026-05-26T20:40:45.208Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/6b/ff/f33e4823667e27548e8fd8df44217515303f9808d0ff29817db56f87d990/coverage-7.14.1-cp314-cp314-musllinux_1_2_i686.whl", hash = "sha256:7771b601718fdde84832c3a434ca9bbf4ae9adbc49d84198b4110700c3c77c36", size = 251291, upload-time = "2026-05-26T20:40:47.502Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/68/9b/489db0ebb209054766b90a9014a45f6d26eb724c02ec21311c3733b5a644/coverage-7.14.1-cp314-cp314-musllinux_1_2_ppc64le.whl", hash = "sha256:39b21e212c55af06fa375e3dbf90a8a8e38792f3a910c580066d23563830ddd5", size = 255564, upload-time = "2026-05-26T20:40:49.372Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/27/b5/16bc2d4c2409b23c7737edb68c83bc89e345f378050549fe1d75ac7d34d5/coverage-7.14.1-cp314-cp314-musllinux_1_2_riscv64.whl", hash = "sha256:f2302660e32562a532b442480121aef8aa61a5bdb20b30bf0adab29f10a5a4b4", size = 251107, upload-time = "2026-05-26T20:40:51.677Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7d/0c/2629997469a00cd069d588a41c9dc887610f2775ae89d250c4791e65272a/coverage-7.14.1-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:03a6f93c1ec3b7f2e77b5dbcc5573a2c21f12529a5c6bbe0f16f72303cc2fa4d", size = 252764, upload-time = "2026-05-26T20:40:54.267Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/d2/ee/f78d63c8f079e0d7211c7e2401fa17e311514534ba61bae03e4b287ce4ab/coverage-7.14.1-cp314-cp314-win32.whl", hash = "sha256:8a3ce026d73290f42f08dafecbd82c193a74df280461fbf97300fec51fd133ee", size = 222837, upload-time = "2026-05-26T20:40:56.496Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/dc/b9/be539854f93a70dfbeec69117f33ec70dc42ff0b65b5b07ab8d40d04228e/coverage-7.14.1-cp314-cp314-win_amd64.whl", hash = "sha256:114c95ef29302423b87d159075805f4ab973254a2638a5d7d046c94887cc87d7", size = 223650, upload-time = "2026-05-26T20:40:58.351Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fe/9e/24e2842fef40f35ac82ba3a7719c8023d011bf3bf652d0675316a9d088a1/coverage-7.14.1-cp314-cp314-win_arm64.whl", hash = "sha256:a07891c3f4805442b31b71e84ba3cf29ed1aa9a428284e06deeb4b23e5b46343", size = 222218, upload-time = "2026-05-26T20:41:00.321Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/0a/1d/ac0a9df5fe31c1e8bdd658074905fc12844a05c1a7e3fdb8417e97c31e23/coverage-7.14.1-cp314-cp314t-macosx_10_15_x86_64.whl", hash = "sha256:1101a5ebb083aecb625ebb6209d4105b58f647b093cb2dc8122d7b33f743cfe1", size = 220822, upload-time = "2026-05-26T20:41:02.281Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/32/cf/f964fd9aff20323f9f1a726c97135f8a76bcd87b92dad141a456a43f3c64/coverage-7.14.1-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:851b9e1e4e8a4608e77c79714b2e77c0970d2ed7202a05e92ae407817481887b", size = 221084, upload-time = "2026-05-26T20:41:04.593Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/d8/5e/7e5ef2aba844de2b80d678619fcf0841b42e3f37f16411226f3fe4c1016f/coverage-7.14.1-cp314-cp314t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d5b89cdfb2ee051b71e8c3c70bd81a9eff81100f736a269136fe1a68efe00474", size = 262454, upload-time = "2026-05-26T20:41:06.641Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/64/62/75809bded87015cc4935524218a2a8ed8dd1a8498bfed30a2f4f7a4b4d34/coverage-7.14.1-cp314-cp314t-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:0177614a0370f227888b4e436a7c55686d6a9f90eb1ade2b624ba685a1686e86", size = 264578, upload-time = "2026-05-26T20:41:08.556Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f3/42/d33392dc14633525012d2d504fa1a33b05538bf535f5c1d64675e5754b78/coverage-7.14.1-cp314-cp314t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2d69af5dea2de76fc485a83032a630523f985198b7e25be901ec60181587b01e", size = 266981, upload-time = "2026-05-26T20:41:10.824Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/2a/49/0157c4428c2aca7f1e09d5565930586fd5ae36f1655f08b0daa7cf1fcae1/coverage-7.14.1-cp314-cp314t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:35ab22d91de736e8966b980dc355cbcdd2c6dbbcfe275f9a2991bc8a91b3df65", size = 268112, upload-time = "2026-05-26T20:41:12.966Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/96/26/86b9ce71f4092b1ed325ce1421698081df1286b833400b6836912834d6e0/coverage-7.14.1-cp314-cp314t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:357d4e32935c36588aaba057d734fa32428c360c9fc2e4442afbf1b646beee6e", size = 261558, upload-time = "2026-05-26T20:41:15Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/20/4c/c311210c5472cf5401d8422b0d7812cdd520f24417673afabda6c323faca/coverage-7.14.1-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:51bd64741cc6fa065abd300ede1afe5a5291ece9c31da8b24884deda48bcc3f8", size = 264447, upload-time = "2026-05-26T20:41:17.369Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fb/71/59513f8710ed3e6b0ac0a050a5b7e977bb9c9e880354863b5d00d8809256/coverage-7.14.1-cp314-cp314t-musllinux_1_2_i686.whl", hash = "sha256:9132cd363a68a4c3daa7c8704a654b1e39d3360f6f5b8ddd470608a945236c07", size = 262048, upload-time = "2026-05-26T20:41:19.309Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/84/8d/bceed32dc494f5bbf50f775cd2e78ca814953942b5ea28d3c1c3ac316f14/coverage-7.14.1-cp314-cp314t-musllinux_1_2_ppc64le.whl", hash = "sha256:07c6290b1697b862c0478eab545eec949a0d0e4d6d03497f446d706da3b4f2de", size = 265781, upload-time = "2026-05-26T20:41:21.559Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e7/c5/9348fe40dbfd4991aaf78df2c6c3098bfb2cc834d1fd362a64b4efef855a/coverage-7.14.1-cp314-cp314t-musllinux_1_2_riscv64.whl", hash = "sha256:5ea0c297e27133853b4d8a3eb799bff5a2dbd9f2f41537a240d337ac9b4df890", size = 260896, upload-time = "2026-05-26T20:41:23.428Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ca/92/1ea0f03929da7cf87206b1fa24f4c8e9c158be0455481af29ec0a1f3503f/coverage-7.14.1-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:01b7733daad0237daa01ef80fe2dfceffc911e6a17fa7b55d14aa8214eaaaecd", size = 263214, upload-time = "2026-05-26T20:41:25.419Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f6/a9/b2493c054c0e01a643266742ab45e15744e60743f9260cd930c7142b1124/coverage-7.14.1-cp314-cp314t-win32.whl", hash = "sha256:6adc5a36984624a70bf11d7184e20fa0a49aa7c47ffab43804106a1a695ea22e", size = 223624, upload-time = "2026-05-26T20:41:27.795Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fc/bd/3e1e6a57fccd2d7c83fcdf338e93ba98eb85c6e877dd34731ac585375490/coverage-7.14.1-cp314-cp314t-win_amd64.whl", hash = "sha256:ddf799247318f34dbcd2efa8c95a8d0642674e926bb1774cf9b63dfd2a389d1c", size = 224728, upload-time = "2026-05-26T20:41:30.098Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/bb/d7/31066cf1d2f0c6c797fce911bcfa01dd35642dc6da992a950256097c5860/coverage-7.14.1-cp314-cp314t-win_arm64.whl", hash = "sha256:145986fe66647eb489f18d9a997567a3fd358584c4b5a808769113abc07466af", size = 222752, upload-time = "2026-05-26T20:41:32.123Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8a/3c/1a983b9a745d7f83d53f057bcc5bf79ba6a2bbc08266b3f0c7d6fe630c9b/coverage-7.14.1-py3-none-any.whl", hash = "sha256:a252f21c27e38347e60111a3266b03827422a7d5525951aceee313aa68bab1d2", size = 211815, upload-time = "2026-05-26T20:41:34.078Z" },
+]
+
 [[package]]
 name = "curl-cffi"
 version = "0.15.0"
@@ -622,6 +748,92 @@ wheels = [
     { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/2c/47/c99d5268f354002ce80f8d029cd9d7d872969da1de8b93d32de4dc56d6f4/fonttools-4.63.0-py3-none-any.whl", hash = "sha256:445af2eab030a16b9171ea8bdda7ebf7d96bda2df88ee182a464252f6e05e20d", size = 1164562, upload-time = "2026-05-14T12:04:29.092Z" },
 ]
 
+[[package]]
+name = "greenlet"
+version = "3.5.3"
+source = { registry = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple/" }
+sdist = { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e2/f1/fbbfef6af0bad0548f09bc28948ea3c275b4edb19e17fc5ca9900a6a634d/greenlet-3.5.3.tar.gz", hash = "sha256:a61efc018fd3eb317eeca31aba90ee9e7f26f22884a79b6c6ec715bf71bb62f1", size = 200270, upload-time = "2026-06-26T19:28:24.832Z" }
+wheels = [
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f5/a1/1f7f0c555f5858fd2906fe9f7b0a3554fddb85cb70df7a6aaec41dc292c2/greenlet-3.5.3-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:c180d22d325fb613956b443c3c6f4406eb70e6defc70d3974da2a7b59e06f48c", size = 285838, upload-time = "2026-06-26T18:21:05.167Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/0a/29/be9f43ed61677a5759b38c8a9389248133c8c731bbfc0574ecdff66c99fc/greenlet-3.5.3-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:483d08c11181c83a6ce1a7a61df0f624a208ec40817a3bb2302714592eee4f04", size = 602342, upload-time = "2026-06-26T19:07:06.908Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b9/42/ba41c97ec36aa4b3ec25e5aa691d79561254805fad7f2f826dd6770587e2/greenlet-3.5.3-cp310-cp310-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:1dae6e0091eae084317e411f047f0b7cb241c6db570f7c45fd6b900a274914ce", size = 615541, upload-time = "2026-06-26T19:10:04.909Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/2e/8c/231ca675b0df779816950ca66b40b1fa14dbff4a0ed9814a9a29ec399140/greenlet-3.5.3-cp310-cp310-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0f6ff50ff8dbd51fae9b37f4101648b04ea0df19b3f50ab2beb5061e7716a5c8", size = 622473, upload-time = "2026-06-26T19:24:12.786Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f5/c7/28747042e1df8a9cd120a1ebe15529fc4be3b486e13e8d551ff307a82412/greenlet-3.5.3-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9bcd2d72ccd70a1ec68ba6ef93e7fbb4420ef9997dabc7010d893bd4015e0bec", size = 615675, upload-time = "2026-06-26T18:32:14.444Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/81/fe/dd97c483a3ff82849196ccd07851600edd3ac9de74669ca8a6022ada9ea1/greenlet-3.5.3-cp310-cp310-manylinux_2_39_riscv64.whl", hash = "sha256:37bf9c538f5ae6e63d643f88dec37c0c83bdf0e2ebc62961dedcf458822f7b71", size = 418421, upload-time = "2026-06-26T19:25:34.503Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/cc/a8/b85525a6c8fba9f009a5f7c8df1545de8fb0f0bf3e0179194ef4e500317f/greenlet-3.5.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:73f152c895e09907e0dbe24f6c2db37beb085cd63db91c3825a0fcd0064124a8", size = 1575057, upload-time = "2026-06-26T19:09:00.264Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/03/79/fb76edb218fe6735ab0edeba176c7ab80df9618f7c02ce4208979f3ae7db/greenlet-3.5.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:8bdb43e1a1d1873721acab2be99c5befd4d2044ddfd52e4d610801019880a702", size = 1641692, upload-time = "2026-06-26T18:31:41.454Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/6b/79/86fe3ee50ed55d9b3907eecd3208b5c3fe8a79515519aae98b4753c3fa1d/greenlet-3.5.3-cp310-cp310-win_amd64.whl", hash = "sha256:0909f9355a9f24845d3299f3112e266a06afb68302041989fd26bd68894933db", size = 238742, upload-time = "2026-06-26T18:20:40.758Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/51/58/5404031044f55afad7aad1aff8be3f22b1bed03e237cfeabbc7e5c8cfde0/greenlet-3.5.3-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:aca9b4ce85b152b5524ef7d88170efdff80dc0032aa8b75f9aaf7f3479ea95b4", size = 287424, upload-time = "2026-06-26T18:20:31.469Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b4/bf/1c65e9b94a54d547068fa5b5a8a06f221f3316b48908e08668d29c77cb50/greenlet-3.5.3-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0f71be4920368fe1fabeeaa53d1e3548337e2b223d9565f8ad5e392a75ba23fc", size = 606523, upload-time = "2026-06-26T19:07:08.859Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b8/c7/b66baacc95775ad511287acb0137b95574a9ce5491902372b7564799d790/greenlet-3.5.3-cp311-cp311-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4d77e67f65f98449e3fb83f795b5d0a8437aead2f874ca89c96576caf4be3af6", size = 618315, upload-time = "2026-06-26T19:10:06.055Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b0/a0/68afd1ebad40db87dac0a28ffa120726b98bf9c7c40c481b0f63c105d298/greenlet-3.5.3-cp311-cp311-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:e18619ba655ac05d78d80fc83cac4ba892bd6927b99e3b8237aee861aaacc8bb", size = 626155, upload-time = "2026-06-26T19:24:14.44Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/78/2b/28ed29463522fdbe4c15b1f63922041626a7478316b34ab4adda3f0a4aba/greenlet-3.5.3-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8540f1e6205bd13ca0ce685581037219ca54a1b41a0a15d228c6c9b8ad5903d7", size = 617381, upload-time = "2026-06-26T18:32:16.077Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/07/7f/e327d912239ec4b3b49999e3967389bcf1ee8722b9ee9194d2752ecd558a/greenlet-3.5.3-cp311-cp311-manylinux_2_39_riscv64.whl", hash = "sha256:d27c0c653a60d9535f690226474a5cc1036a8b0d7b57504d1c4f89c44a07a80c", size = 421083, upload-time = "2026-06-26T19:25:35.804Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/2a/7b/ad04e9d1337fc04965dc9fc616b6a72cb65a24b800a014c011ec812f5489/greenlet-3.5.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:7ef56fe650f50575bf843acde967b9c567687f3c22340941a899b7bc56e956a8", size = 1577771, upload-time = "2026-06-26T19:09:01.537Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/d8/33/6c87ab7ba663f70ca21f3022aad1ffe56d3f3e0521e836c2415e13abcc3c/greenlet-3.5.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:5121af01cf911e70056c00d4b46d5e9b5d1415550038573d744138bacb59e6b8", size = 1644048, upload-time = "2026-06-26T18:31:42.996Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/1c/35/f0d8ee998b422cf8693b270f098e55d8d4ec8006b061b333f54f177d28d9/greenlet-3.5.3-cp311-cp311-win_amd64.whl", hash = "sha256:0f41e4a05a3c0cb31b17023eff28dd111e1d16bf7d7d00406cd7df23f31398a7", size = 239137, upload-time = "2026-06-26T18:23:21.664Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fb/96/b9820295576ef18c9edc404f10e260ae7215ceaf3781a54b720ed2627862/greenlet-3.5.3-cp311-cp311-win_arm64.whl", hash = "sha256:ec6f1af59f6b5f3fc9678e2ea062d8377d22ac644f7844cb7a292910cf12ff44", size = 237630, upload-time = "2026-06-26T18:24:00.281Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/5d/6e/4c37d51a2b7f82d2ff11bb6b5f7d766d9a011726624af255e843727627a3/greenlet-3.5.3-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:719757059f5a53fd0dde23f78cffeafcdd97b21c850ddb7ca684a3c1a1f122e2", size = 288685, upload-time = "2026-06-26T18:22:08.977Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7a/73/815dd90131c1b71ebdf53dbc7c276cafec2a1173b97559f97aba72724a87/greenlet-3.5.3-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:efa9f765dd09f9d0cdac651ffdf631ee59ec5dc6ee7a73e0c012ba9c52fbdf5b", size = 604761, upload-time = "2026-06-26T19:07:10.114Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/9f/57/079cfe76bcef36b153b25607ee91c6fcb58f17f8b23c86bbbeabe0c88d72/greenlet-3.5.3-cp312-cp312-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7faba15ac005376e02a0384504e0243be3370ce010296a44a820feb342b505ab", size = 617044, upload-time = "2026-06-26T19:10:07.25Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/fb/fb/d97dc261209c80744b7c8132693a30d70ec6e7315e632cb0a10b3fec94dd/greenlet-3.5.3-cp312-cp312-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:5795cd1101371140551c645f2d408b8d3c01a5a29cf8a9bce6e759c983682d23", size = 622351, upload-time = "2026-06-26T19:24:16.32Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/37/87/b4d095775a3fb1bcafbb483fc206b27ebb785724c83051447737085dc54e/greenlet-3.5.3-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:87142215824be6ac05e2e8e2786eec307ccbc27c36723c3881959df654af6861", size = 614244, upload-time = "2026-06-26T18:32:17.594Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8e/ac/e5fee13cbbd0e8de312d9a146584b8a51891c68847330ef9dc8b5109d23f/greenlet-3.5.3-cp312-cp312-manylinux_2_39_riscv64.whl", hash = "sha256:af4923b3096e26a36d7e9cf24ab88083a20f97d191e3b97f253731ce9b41b28c", size = 425395, upload-time = "2026-06-26T19:25:37.144Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8a/70/7559b609683650fa2b95b8ab84b4ab0b26556a635d19675e12aa832d826d/greenlet-3.5.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:215275b1b49320987352e6c1b054acca0064f965a2c66992bed9a6f7d913f149", size = 1574210, upload-time = "2026-06-26T19:09:03.077Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ae/73/be55392074c60fc37655ca40fa6022457bfbf6718e9e342a7b0b41f96dd2/greenlet-3.5.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:6b1b0eed82364b0e32c4ea0f221452d33e6bb17ae094d9f72aed9851812747ea", size = 1638627, upload-time = "2026-06-26T18:31:44.748Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/14/40/c57489acf8e37d74e2913d4eff63aa0dba17acccc4bdeef874dde2dbbec9/greenlet-3.5.3-cp312-cp312-win_amd64.whl", hash = "sha256:cde8adafa2365676f74a979744629589999093bc86e2484214f58e61df08902c", size = 239882, upload-time = "2026-06-26T18:23:27.518Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/71/fd/6fea0e3d6600f785069481ee637e09378dd4118acdfd38ad88ae2db31c98/greenlet-3.5.3-cp312-cp312-win_arm64.whl", hash = "sha256:c4e7b79d83805475f0102008843f6eb45fd3bb0b2e88c774adab5fbaab27117d", size = 238211, upload-time = "2026-06-26T18:22:37.671Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/9b/ff/a620267401db30a50cc8450ee90730e2d4a85658c055c0e760d4ed47fb13/greenlet-3.5.3-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:c8d87c2134d871df96ecdea9cec7cbaab286dadab0f56476e57aaf9e8ac11550", size = 287609, upload-time = "2026-06-26T18:21:14.724Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/d6/fa/5401ac78021c826a25b6dde0c705e0a8f29b617509f9185a31dac15fbe1b/greenlet-3.5.3-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a2d185dd1621757e70c3861cceffd5317ab4e7ed7eb09c82994828468527ade5", size = 607435, upload-time = "2026-06-26T19:07:11.412Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e9/76/1dc144a2e56e65d36405078ed774224375ea520a1870a6e46e08bb4ac7bf/greenlet-3.5.3-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:1c514a468149bf8fbbab874188a3535cd8a48a3e353eb53a3d424296f8dbacd3", size = 619787, upload-time = "2026-06-26T19:10:08.396Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/57/61/2f5b1adf256d039f5dab8005de8d3d7ad2b0070a3219c0e036b3fbfeb440/greenlet-3.5.3-cp313-cp313-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:9ad04dd75458c6300b047c61b8639092433d205a25a14e310d6582a480efcca1", size = 625580, upload-time = "2026-06-26T19:24:18.344Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/bf/87/c298cee62df1de4ad7fec32abda73526cff347fd143a6ed4ac369246668a/greenlet-3.5.3-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:915f887cf2682b66419b879423a2e072634aa7b7dce6f3ada4957cfced3f1e9a", size = 616786, upload-time = "2026-06-26T18:32:19.128Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/3e/d9/ab7fc9e543e44d6879b0a6ef9a4b2188940fd180cc65d6f646883ddf7201/greenlet-3.5.3-cp313-cp313-manylinux_2_39_riscv64.whl", hash = "sha256:afaabdd554cd7ae9bbb3ca070b0d7fdfd207dbf1d16865f7233837709d354bda", size = 427933, upload-time = "2026-06-26T19:25:38.219Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/9e/2e/e6f009885ed0705ccf33fe0583c117cfd03cde77e31a596dd5785a30762b/greenlet-3.5.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:766cfd421c13e450feb340cd472a3ed9957d438727b7b4593ad7c76c5d2b0deb", size = 1574316, upload-time = "2026-06-26T19:09:04.273Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ef/fe/43fd110b01e40da0adb7c90ac7ea744bef2d43dca00de5095fd2351c2a68/greenlet-3.5.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:2ecda9ec22edf38fa389369eaed8c3d37c05f3c54e69f69438dbb2cc1de1458b", size = 1638614, upload-time = "2026-06-26T18:31:46.297Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/0f/7c/062447147a61f8b4337b156fe70d32a165fcf2f89d7ca6255e572806705c/greenlet-3.5.3-cp313-cp313-win_amd64.whl", hash = "sha256:c82304750f057167ff60d188df1d0cc1764ce9567eadf03e6a7443bcedd0b30b", size = 239850, upload-time = "2026-06-26T18:21:54.613Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c7/7e/220a7f5824a64a60443fc03b39dfac4ea63a7fb6d481efa27eafa928e7f4/greenlet-3.5.3-cp313-cp313-win_arm64.whl", hash = "sha256:dc133a1569ee667b2a6ef56ce551084aeefd87a5acbc4736d336d1e2edc6cfc4", size = 238141, upload-time = "2026-06-26T18:22:48.507Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c3/93/43e116ee114b28737ba7e12952a0d4e2f55944d0f84e42bc91ba7192a3c9/greenlet-3.5.3-cp314-cp314-macosx_11_0_universal2.whl", hash = "sha256:fd2e02fa07485778536a036222d616ab957b1d533f36b3ed98ce725d9c9d3117", size = 288202, upload-time = "2026-06-26T18:23:49.604Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/82/2f/146d218299046a43d1f029fd544b3d110d0f175a09c715c7e8da4a4a345d/greenlet-3.5.3-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:df0a0628d1597eb0897b62f55d1343f772405fd25f3b2a796c76874b0c2e22e8", size = 654096, upload-time = "2026-06-26T19:07:12.71Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a0/cc/04738cafb3f45fa991ea44f9de94c47dcec964f5a972300988a6751f49d9/greenlet-3.5.3-cp314-cp314-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ebd933a6adabc298bab47731a130fe6bfb888bd934eee37810f151159544540d", size = 666304, upload-time = "2026-06-26T19:10:09.503Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/86/a9/73fa62893d5b84b4205544e6b673c654cc43aa5b9899bac00f04d64af73d/greenlet-3.5.3-cp314-cp314-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:8d19fe6c39ebff9259f07bcc685d3290f8fa4ea2278e51dd0008e4d6b0f2d814", size = 670657, upload-time = "2026-06-26T19:24:19.967Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ce/aa/4e0dad5e605c270c784ab911c43da6adb136ccd4d81180f763ca429a723d/greenlet-3.5.3-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4b9d501b40e80b70e32323c799dd9b420a5577a9601469d362ae1ffb690f3a7c", size = 663635, upload-time = "2026-06-26T18:32:20.802Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/29/7e/2ffce64929fb3cab7b65d5a0b20aaf9764e227681d731b041077fc9a525a/greenlet-3.5.3-cp314-cp314-manylinux_2_39_riscv64.whl", hash = "sha256:962c5df2db8cb446da51edf1ca5296c389d93b99c9d8aa2ee4c7d0d8f1218260", size = 473497, upload-time = "2026-06-26T19:25:39.421Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/d1/50/13efdbea246fe3d3b735e191fec08fb50809f53cd2383ebe123d0809e44b/greenlet-3.5.3-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:a1fad1d11e7d6aab184107baa8e4ece11ccba3ec9599cd7efa5ff4d70d43256a", size = 1621252, upload-time = "2026-06-26T19:09:05.647Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f7/22/c0a336ae4a1410fd5f5121098e5bfbf1865f64c5ef80b4b5412886c4a332/greenlet-3.5.3-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:fad5aec764399f1b5cc347ad250a59660f20c8f8888ea6bae1f93b769cce1154", size = 1684824, upload-time = "2026-06-26T18:31:47.738Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7a/94/91aec0030bea75c4b3244251d0de60a1f3432d1ecb53ab6c437fb5c3ba61/greenlet-3.5.3-cp314-cp314-win_amd64.whl", hash = "sha256:7669aa24cf2a1041d6f7899575b494a3ab4cf68bfcc8609b1dc0be7272db835e", size = 240754, upload-time = "2026-06-26T18:22:15.669Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e5/06/68d0983e79e02138f64b4d303c500c27ddb48e5e77f3debb80888a921eae/greenlet-3.5.3-cp314-cp314-win_arm64.whl", hash = "sha256:5b4807c4082c9d1b6d9eed56fcd041863e37f2228106eef24c30ca096e238605", size = 239549, upload-time = "2026-06-26T18:22:42.996Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/91/95/3e161213d7f1d378d15aa9e792093e9bfe01844680d04b7fd6e0107c9098/greenlet-3.5.3-cp314-cp314t-macosx_11_0_universal2.whl", hash = "sha256:271a8ea7c1024e8a0d7dd2be66dd66dda8a07193f41a17b9e924f7600f5b62be", size = 296389, upload-time = "2026-06-26T18:22:20.657Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/00/92/715c44721abe2b4d1ae9abde4179411868a5bff312479f54e105d372f131/greenlet-3.5.3-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:19131729ae0ddc3c2e1ef85e650169b5e37ee32e400f215f78b94d7b0d567310", size = 653382, upload-time = "2026-06-26T19:07:14.209Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a0/83/37a10372a1090a6624cca8e74c12df1a36c2dc36429ed0255b7fb1aeee23/greenlet-3.5.3-cp314-cp314t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:1540dd8e5fc2a5aec40fbb98ef8e149fa47c89a4b4a1cf2575a14d3d1869d7a8", size = 659401, upload-time = "2026-06-26T19:10:10.876Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/cb/73/8faec206b851c22b1733545fda900829a1f3f5b1c78ae7e0fb3dba57d9f4/greenlet-3.5.3-cp314-cp314t-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:b897d97759425953f69a9c0fac67f8fe333ec0ce7377ef186fb2b0c3ad5e354d", size = 659582, upload-time = "2026-06-26T19:24:21.357Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/db/e2/d1509cad4207da559cc42986ecdd8fc67ad0d1bba2bf03023c467fd5e0f3/greenlet-3.5.3-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e81fa194a1d20967877bdf9c7794db2bc99063e5be36aee710c08f04c5bb087f", size = 656969, upload-time = "2026-06-26T18:32:22.272Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b4/55/50c19e49f8045834ada71ef12f8ad048eba8517c6aa41161bed676328fae/greenlet-3.5.3-cp314-cp314t-manylinux_2_39_riscv64.whl", hash = "sha256:3236754d423955ea08e9bb5f6c04a7895f9e22c290b66aa7653fcb922d839eb0", size = 491037, upload-time = "2026-06-26T19:25:40.672Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/86/7d/eaf70de20aadca3a5884aec58362861c64ce45e7b277f47ed026926a3b89/greenlet-3.5.3-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:55cf4d777485d43110e47133cbba6d74a8885a87ec1227ef0267f9ee80c5aa21", size = 1617822, upload-time = "2026-06-26T19:09:06.893Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8a/f9/414d38fc400ae4350d4185eaad1827676f7cf5287b9136e0ed1cbbe20a7f/greenlet-3.5.3-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:12a248ba75f6a9a236375f52296c498c89ff1d8badf32deb9eca7abd5853f7da", size = 1677983, upload-time = "2026-06-26T18:31:49.396Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e4/15/7edb977e08f9bff702fe42d6c902702786ff6b9694058b4e6a2a6ac90e57/greenlet-3.5.3-cp314-cp314t-win_amd64.whl", hash = "sha256:efc6bd60ea02e085862c74a3ef64b147ffc6f1a5ea7d9f26e7a939943f68c1e3", size = 243626, upload-time = "2026-06-26T18:24:41.485Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/2c/8a/93928dce91e6b3598b5e779e8d1fd6576a504640c58e78627077f6a7a91a/greenlet-3.5.3-cp315-cp315-macosx_11_0_universal2.whl", hash = "sha256:ea03f2f04367845d6b58eeed276e1e56e51f0b97d8ad5a88a7d20a91dc9056cc", size = 288860, upload-time = "2026-06-26T18:22:48.07Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/4f/ca/69db42d447a1378043e2c8f19c09cbbd1263371505053c496b49066d3d16/greenlet-3.5.3-cp315-cp315-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:78dbef602fda6d97d957eb7937f70c9ce9e9527330347f8f6b6f9e554a9e7a47", size = 659747, upload-time = "2026-06-26T19:07:15.565Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a8/0b/af7ac2ef8dd41e3da1a40dda6305c23b9a03e13ba975ec916357b50f8575/greenlet-3.5.3-cp315-cp315-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:6f73857adb8fee13fa56c172bd11262f888c0c648f9fea113e777bb2c7904a81", size = 670419, upload-time = "2026-06-26T19:10:12.293Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/25/aa/952cf28c2ff949a8c971134fb43854dd7eaa737218723aaef758f8c9aead/greenlet-3.5.3-cp315-cp315-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:cefa9cef4b371f9844c6053db71f1138bc6807bab1578b0dae5149c1f1141357", size = 674261, upload-time = "2026-06-26T19:24:22.79Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/51/1e/1d51640cacbfc455dbe9f9a9f594c49e4e244f63b9971a2f4764e46cc53d/greenlet-3.5.3-cp315-cp315-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:232fec92e823addaf02d9472cf7381e24a1d046a6ced1103c5caa4c21b9dfc1d", size = 668787, upload-time = "2026-06-26T18:32:24.298Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/dc/f2/b00d6f5e63e531a93562b2ec1a4c320fbee91f580fc42e6417af69d706e5/greenlet-3.5.3-cp315-cp315-manylinux_2_39_riscv64.whl", hash = "sha256:6219b6d04dbf6ba6084d77dc609e8473060dc55f759cbf626d512122781fa128", size = 480322, upload-time = "2026-06-26T19:25:41.852Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/21/66/4030d5b0b5894500023f003bb054d9bb354dfbd1e186c3a296759172f5f5/greenlet-3.5.3-cp315-cp315-musllinux_1_2_aarch64.whl", hash = "sha256:2421c3564da9429d5586d46ca31ebb26516b5498a802cf65c041a8e8a8980d34", size = 1626305, upload-time = "2026-06-26T19:09:08.281Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/0e/50/5221371c7550108dfa3c378debc41d032aa9c78e89abb01d8011cfc93289/greenlet-3.5.3-cp315-cp315-musllinux_1_2_x86_64.whl", hash = "sha256:e0f0d160f0b2e558e6c75f7930967183255dc9735e5f5b8cae58ee09c9576d8b", size = 1688631, upload-time = "2026-06-26T18:31:51.278Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/68/5d/00d469daae3c65d2bf620b10eee82eb022127d483c6bc8c69fae6f3fbf17/greenlet-3.5.3-cp315-cp315-win_amd64.whl", hash = "sha256:dd99329bbc15ca78dcc583dba05d0b1b0bae01ab6c2174989f5aaee3e41ac930", size = 241027, upload-time = "2026-06-26T18:22:38.203Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e7/e8/883785b44c5780ed71e83d3e4437e710470be17a2e181e8b601e2da0dc4a/greenlet-3.5.3-cp315-cp315-win_arm64.whl", hash = "sha256:499fef2acede88c1864a57bb586b4bf533c81e1b82df7ab93451cdb47dfec227", size = 240085, upload-time = "2026-06-26T18:23:54.217Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/1c/da/4f4a8450962fad137c1c8981a3f1b8919d06c829993d4d476f9c525d5173/greenlet-3.5.3-cp315-cp315t-macosx_11_0_universal2.whl", hash = "sha256:176bc16a721fa5fc294d70b87b4dfa5fbdd251b3da5d5372735ecef9bd7d6d0c", size = 297221, upload-time = "2026-06-26T18:23:27.176Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/57/66/b3bfae3e220a9b63ea539a0eea681800c69ab1aada757eae8789f183e7ce/greenlet-3.5.3-cp315-cp315t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:629b614d2b786e89c50440e246f33eea78f58a962d0bdbbcc809e6d13605903f", size = 657221, upload-time = "2026-06-26T19:07:16.973Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/7b/81/b6d4d73a709684fc77e7fa034d7c2fe82cffa9fc920fadcaa659c2626213/greenlet-3.5.3-cp315-cp315t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:2b2e857ae16f5f72142edf75f9f176fe7526ba19a2841df1420516f83831c9f2", size = 663226, upload-time = "2026-06-26T19:10:13.723Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/e9/39/0e0938a75115b939d42733a2a12e1d349653c9531fe6fe563e8a681f04e6/greenlet-3.5.3-cp315-cp315t-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:16d192579ed281051396dddd7f7754dac6259e6b1fb26378c87b66622f8e3f91", size = 663706, upload-time = "2026-06-26T19:24:24.312Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f5/07/e210b02b589f16e74ff48b730690e4a34ffe984219fce4f3c1a0e7ec8545/greenlet-3.5.3-cp315-cp315t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e515757e2e36bcbf1fad09a46e1557e8b1ae1797d4b44d09da7deed88ad28608", size = 660802, upload-time = "2026-06-26T18:32:26.081Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/5b/41/35d1c678cdb3c3b9e6bee691728e563cfb294202b23c7a4c3c2ccc343589/greenlet-3.5.3-cp315-cp315t-manylinux_2_39_riscv64.whl", hash = "sha256:4399eb8d041f20b68d943918bc55502a93d6fdc0a37c14da7881c04139acee9d", size = 498803, upload-time = "2026-06-26T19:25:43.063Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/eb/2e/5303eb3fa06bca089060f479707182a93e360683bc252acf846c3090d34e/greenlet-3.5.3-cp315-cp315t-musllinux_1_2_aarch64.whl", hash = "sha256:b363d46ed1ea431825fdb01471bb024fc08399bad1572a616e853c7684415adb", size = 1622157, upload-time = "2026-06-26T19:09:09.527Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/54/70/50de47a488f14df260b50ae34fb5d56016e308b098eab02c878b5223c26a/greenlet-3.5.3-cp315-cp315t-musllinux_1_2_x86_64.whl", hash = "sha256:e44da2f5bbdaabaf7d80b73dbb430c7035771e9f244e3c8b769715c9d8fa0a16", size = 1681159, upload-time = "2026-06-26T18:31:52.986Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a7/13/1055e1dda7882073eda533e2b96c62e55bbd2db7fda6d5ece992febc7071/greenlet-3.5.3-cp315-cp315t-win_amd64.whl", hash = "sha256:8ff8bed3e3baa20a3ea261ce00526f1898ad4801d4886fd2220580ee0ad8fadf", size = 244007, upload-time = "2026-06-26T18:22:04.353Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b4/0d/ca7d15afbdc397e3401134c9e1800d51d12b829661786187a4ad08fe484f/greenlet-3.5.3-cp315-cp315t-win_arm64.whl", hash = "sha256:b7068bd09f761f3f5b4d214c2bed063186b2a86148c740b3873e3f56d79bac31", size = 242586, upload-time = "2026-06-26T18:23:37.93Z" },
+]
+
 [[package]]
 name = "html5lib"
 version = "1.1"
@@ -1412,6 +1624,25 @@ wheels = [
     { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/bc/60/5382c03e1970de634027cee8e1b7d39776b778b81812aaf45b694dfe9e28/pillow-12.2.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:bfa9c230d2fe991bed5318a5f119bd6780cda2915cca595393649fc118ab895e", size = 7080946, upload-time = "2026-04-01T14:46:11.734Z" },
 ]
 
+[[package]]
+name = "playwright"
+version = "1.61.0"
+source = { registry = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple/" }
+dependencies = [
+    { name = "greenlet" },
+    { name = "pyee" },
+]
+wheels = [
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/44/ee/31e4e0db36588b817a10b299a0285082545fde7d36543c2abe498bb3d61a/playwright-1.61.0-py3-none-macosx_10_13_x86_64.whl", hash = "sha256:ff138c3a604f69911e9d42fd036e55c2a171e5616edf04c1e7f60a2a285540b0", size = 43421877, upload-time = "2026-06-29T10:32:48.428Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/42/35/71395dd3ecc798965be4a3ef8c443217d4abca168e7cb34536304f9489e6/playwright-1.61.0-py3-none-macosx_11_0_arm64.whl", hash = "sha256:009588c2a7e499bc5a8b425b61fa65490968bbda9cd69e0cf2cff10f8304659a", size = 42205016, upload-time = "2026-06-29T10:32:52.104Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f4/44/323164cf5cd1647bdefce76ffce27651aadb959d089b48f53ea40918276e/playwright-1.61.0-py3-none-macosx_11_0_universal2.whl", hash = "sha256:9f7de4536088d12037c13a52b7ea34b59270b78926bb56935070597ffac6b1af", size = 43421884, upload-time = "2026-06-29T10:32:55.773Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/ab/f8/a35bf179e4ba2522c1893635094a64e407572547bd61528820fc0abc87fe/playwright-1.61.0-py3-none-manylinux1_x86_64.whl", hash = "sha256:54f3b39f6eab832e33458c1dd7da0b5682aedab3b09ae731b5c59fa12fd2024e", size = 47421381, upload-time = "2026-06-29T10:32:59.903Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/b7/eb/e3f922348ec17c315f98c463f72faa1181a1c3de0bfe31a8d2edf6561723/playwright-1.61.0-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:93454322ade8c11d5d6c211bfd91bdfb9ffb4810e3e026371bcbc4bec1b7ee4c", size = 47120545, upload-time = "2026-06-29T10:33:03.574Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/c2/a6/5be4e52b40a9c0c8a073e7c5b0785c05cf5a9ea8f8a7b5b260e32d970342/playwright-1.61.0-py3-none-win32.whl", hash = "sha256:372d55a6f1248fa1dd47599686980cb8fb5bbe6fcda59eab793eb657c11d8a9b", size = 37844841, upload-time = "2026-06-29T10:33:07.361Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/6c/fd/2b78036e5fbe9d5f5645bbe08a1eac7160c51243c0093963edbcf67c35d9/playwright-1.61.0-py3-none-win_amd64.whl", hash = "sha256:35c6cc4589a5d00964a59d7b3e59641e0aac0c02f15479a7af77d20f6bc79597", size = 37844846, upload-time = "2026-06-29T10:33:10.637Z" },
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/27/0d/1b0f3c4ee4eb0514bc805b5c2f9a223e5b6de4f11a926f5235d51d0fc81b/playwright-1.61.0-py3-none-win_arm64.whl", hash = "sha256:e9fcbffcf557a8620fdedd92491eb59a32d18e23d6f3b4f6214b952be324fe51", size = 33955127, upload-time = "2026-06-29T10:33:14.008Z" },
+]
+
 [[package]]
 name = "pluggy"
 version = "1.6.0"
@@ -1496,6 +1727,18 @@ wheels = [
     { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/0c/c3/44f3fbbfa403ea2a7c779186dc20772604442dde72947e7d01069cbe98e3/pycparser-3.0-py3-none-any.whl", hash = "sha256:b727414169a36b7d524c1c3e31839a521725078d7b2ff038656844266160a992", size = 48172, upload-time = "2026-01-21T14:26:50.693Z" },
 ]
 
+[[package]]
+name = "pyee"
+version = "13.0.1"
+source = { registry = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple/" }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/8b/04/e7c1fe4dc78a6fdbfd6c337b1c3732ff543b8a397683ab38378447baa331/pyee-13.0.1.tar.gz", hash = "sha256:0b931f7c14535667ed4c7e0d531716368715e860b988770fc7eb8578d1f67fc8", size = 31655, upload-time = "2026-02-14T21:12:28.044Z" }
+wheels = [
+    { url = "https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/a0/c4/b4d4827c93ef43c01f599ef31453ccc1c132b353284fc6c87d535c233129/pyee-13.0.1-py3-none-any.whl", hash = "sha256:af2f8fede4171ef667dfded53f96e2ed0d6e6bd7ee3bb46437f77e3b57689228", size = 15659, upload-time = "2026-02-14T21:12:26.263Z" },
+]
+
 [[package]]
 name = "pygments"
 version = "2.20.0"