topstep-backtest 0.2.2__tar.gz → 0.2.3__tar.gz

{topstep_backtest-0.2.2 → topstep_backtest-0.2.3}/CHANGELOG.md

@@ -6,6 +6,35 @@ below 1.0, minor releases may contain breaking changes.
 
 ## [Unreleased]
 
+## [0.2.3] — 2026-08-16
+
+### Fixed
+
+- The `[data]` and `[dev]` extras now install `pyarrow`, so **Parquet input works out of the
+  box**. `examples/run_real_data.py` has always branched to `pd.read_parquet` on a `.parquet` /
+  `.pq` suffix, but pandas ships no Parquet engine of its own — so `[data]` alone met that
+  branch with `ImportError: Unable to find a usable engine`, on the format most vendors
+  (Databento included) actually export. Deliberately unbounded, unlike `topstep-sdk` and
+  `ta-lib`: a breaking `pyarrow` major fails loudly at read time instead of quietly changing a
+  number. Reading the same bars from CSV and from Parquet produces a byte-identical report.
+
+  Released as 0.2.3, not 0.2.2: the `v0.2.2` tag was cut from a commit three minutes before
+  this change merged, so **0.2.2 shipped without pyarrow** and its Parquet path fails exactly
+  as 0.2.1's did. A PyPI version can never be re-uploaded, so the fix moved forward a version
+  rather than the tag moving backward.
+
+## [0.2.2] — 2026-08-16
+
+### Fixed
+
+- Repository plumbing only; no user-facing change. `ruff format` also formats Python inside
+  markdown fences, so the tearsheet examples added to the README and `website/results.md` in
+  0.2.1 were unformatted code and CI had been red since — through two merges. And the docs
+  workflow's `actions/configure-pages` step called the Pages REST API under a `contents: read`
+  token, failing every deploy with "Resource not accessible by integration" while blaming the
+  repository's Pages settings; MkDocs takes its base URL from `site_url`, so the step was
+  removed rather than the token widened.
+
 ## [0.2.1] — 2026-08-16
 
 ### Fixed

{topstep_backtest-0.2.2 → topstep_backtest-0.2.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: topstep-backtest
-Version: 0.2.2
+Version: 0.2.3
 Summary: Event-driven backtesting framework for Topstep Trading Combine strategies, with backtest/live parity against topstep-sdk.
 Author-email: Tarric Sookdeo <tarricsookdeo@outlook.com>
 License-Expression: MIT
@@ -23,9 +23,11 @@ Requires-Dist: topstep-sdk<0.2,>=0.1.2
 Requires-Dist: tzdata>=2024.1; sys_platform == 'win32'
 Provides-Extra: data
 Requires-Dist: pandas>=2.2; extra == 'data'
+Requires-Dist: pyarrow>=17; extra == 'data'
 Provides-Extra: dev
 Requires-Dist: hypothesis>=6.100; extra == 'dev'
 Requires-Dist: pandas>=2.2; extra == 'dev'
+Requires-Dist: pyarrow>=17; extra == 'dev'
 Requires-Dist: pyright>=1.1.380; extra == 'dev'
 Requires-Dist: pytest-asyncio>=0.24; extra == 'dev'
 Requires-Dist: pytest-cov>=5.0; extra == 'dev'
@@ -80,7 +82,7 @@ framework helper that compares two TA-Lib outputs rather than computing anything
 
 ```bash
 pip install topstep-backtest
-pip install "topstep-backtest[data]"   # adds pandas, for DataFrame input
+pip install "topstep-backtest[data]"   # adds pandas + pyarrow: DataFrame and Parquet input
 ```
 
 Requires Python 3.12+. TA-Lib is a core dependency and ships wheels for common platforms;

{topstep_backtest-0.2.2 → topstep_backtest-0.2.3}/README.md

@@ -40,7 +40,7 @@ framework helper that compares two TA-Lib outputs rather than computing anything
 
 ```bash
 pip install topstep-backtest
-pip install "topstep-backtest[data]"   # adds pandas, for DataFrame input
+pip install "topstep-backtest[data]"   # adds pandas + pyarrow: DataFrame and Parquet input
 ```
 
 Requires Python 3.12+. TA-Lib is a core dependency and ships wheels for common platforms;

{topstep_backtest-0.2.2 → topstep_backtest-0.2.3}/docs/DESIGN.md

@@ -219,7 +219,8 @@ Violations are bugs, not style. Most are pinned by a named test.
   `ruff format --check` is the one people forget — which is also why `ruff` is pinned
   `>=0.16,<0.17`: format output changes between minors, so an unbounded pin would fail CI on
   correctly formatted code. Runtime deps are `topstep-sdk`, `msgspec`, `ta-lib` (the last is
-  core, not an extra); exactly two extras exist, `[data]` and `[dev]`.
+  core, not an extra); exactly three extras exist, `[data]` (pandas + pyarrow, so both
+  DataFrame and Parquet input work), `[dev]` and `[docs]`.
 
 ## 6. Testing idioms
 

{topstep_backtest-0.2.2 → topstep_backtest-0.2.3}/pyproject.toml

@@ -37,13 +37,24 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
-data = ["pandas>=2.2"]
+data = [
+    "pandas>=2.2",
+    # pandas reads Parquet only through a third-party engine and ships none, so
+    # `[data]` without this turns every .parquet export — the format Databento
+    # and most vendors hand you — into an ImportError at load time. No upper
+    # bound, unlike topstep-sdk and ta-lib above: a breaking pyarrow major fails
+    # LOUDLY at read time rather than quietly changing a number, so pinning it
+    # out would buy nothing and strand the extra on an old wheel.
+    "pyarrow>=17",
+]
 dev = [
     "pytest>=8.0",
     "pytest-asyncio>=0.24",
     "pytest-cov>=5.0",
     "hypothesis>=6.100",
     "pandas>=2.2",
+    # Mirrors `[data]`: the Parquet path needs an engine to be exercised at all.
+    "pyarrow>=17",
     # ruff is PINNED to a minor range on purpose: `ruff format --check` runs in
     # CI, and formatting output changes between ruff minors. Unbounded, CI would
     # spontaneously fail on code that is correctly formatted locally. Bump this

{topstep_backtest-0.2.2 → topstep_backtest-0.2.3}/tests/unit/test_wrangler.py

@@ -5,6 +5,7 @@ from __future__ import annotations
 from collections.abc import Sequence
 from datetime import UTC, datetime
 from decimal import Decimal
+from pathlib import Path
 from typing import Literal
 
 import pandas as pd  # pyright: ignore[reportMissingTypeStubs]
@@ -354,6 +355,41 @@ class TestDataFrame:
                 stamp="open",
             )
 
+    def test_parquet_and_csv_agree(self, tmp_path: Path) -> None:
+        """The `[data]` extra must be able to READ Parquet, not just accept it.
+
+        pandas ships no Parquet engine of its own, so this fails with
+        ``ImportError: Unable to find a usable engine`` if the extra ever loses
+        pyarrow again — which is exactly what shipped in 0.2.1, on the format
+        most vendors export. Reading the two formats through the documented path
+        (`examples/run_real_data.py` parses CSV timestamps itself, because the
+        wrangler takes ints or tz-aware datetimes, never `read_csv`'s strings)
+        must produce identical bars.
+        """
+        stamps = [datetime(2026, 3, 2, 14, 30 + i, tzinfo=UTC) for i in range(3)]
+        df = pd.DataFrame(
+            {
+                "timestamp": stamps,
+                "open": [100.0, 100.25, 100.5],
+                "high": [100.25, 100.5, 100.75],
+                "low": [99.75, 100.0, 100.25],
+                "close": [100.25, 100.5, 100.5],
+                "volume": [10, 20, 30],
+            }
+        )
+        csv_path, pq_path = tmp_path / "bars.csv", tmp_path / "bars.parquet"
+        df.to_csv(csv_path, index=False)  # pyright: ignore[reportUnknownMemberType]
+        # An ImportError here means the [data]/[dev] extras lost their Parquet engine.
+        df.to_parquet(pq_path)  # pyright: ignore[reportUnknownMemberType]
+
+        from_csv = pd.read_csv(csv_path)
+        from_csv["timestamp"] = pd.to_datetime(  # pyright: ignore[reportUnknownMemberType]
+            from_csv["timestamp"]  # pyright: ignore[reportUnknownArgumentType]
+        )
+        from_parquet = pd.read_parquet(pq_path)  # pyright: ignore[reportUnknownMemberType]
+
+        assert self._df_bars(from_csv) == self._df_bars(from_parquet) == self._df_bars(df)
+
 
 @given(
     ticks=st.integers(min_value=1, max_value=1_000_000),