ci: check the built sdist's contents in the release workflow

.github/workflows/releases.yml

@@ -53,15 +53,26 @@ jobs:
     needs: [build_artifacts]
     runs-on: ubuntu-latest
     steps:
+      - uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+
       - uses: actions/download-artifact@9000827ccba6bdab643e8b6fd33ac0654aef8333 # v8.0.2
         with:
           name: releases
           path: dist
 
-      - name: test
-        run: |
-          ls
-          ls dist
+      - name: Install uv
+        uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
+
+      - name: Install just
+        uses: extractions/setup-crate@7577c1bdf2d95e6d65d532788f35ed79d4b1dda2 # v2.0.1
+        with:
+          repo: casey/just
+          version: 1.58.0
+
+      - name: Check sdist contents
+        run: just check-sdist
 
   upload_pypi:
     name: Upload to PyPI

Justfile

@@ -131,6 +131,10 @@ lock *args:
 build *args:
     hatch build "$@"
 
+# Check the built sdist ships what it needs (default: the one in dist/)
+check-sdist sdist="dist":
+    uv run --no-project python ci/check_sdist_contents.py "$@"
+
 # Create a changelog fragment (interactive without arguments)
 changelog *args:
     {{ docs_run }} towncrier create "$@"

changes/4263.misc.md

@@ -0,0 +1 @@
+The release workflow now checks that the built sdist contains every `testpaths` entry, `examples/`, the generated `_version.py` and `py.typed`, and nothing from `packages/`. Run it locally with `just check-sdist` after `just build`.

ci/check_sdist_contents.py

@@ -0,0 +1,92 @@
+"""
+Check a built sdist ships the paths it needs and none it must not.
+
+Usage:
+    python check_sdist_contents.py SDIST
+
+SDIST is a path to a `.tar.gz`, or a directory containing exactly one.
+
+`[tool.hatch.build.targets.sdist]` in pyproject.toml is an allowlist, so the
+mistake to catch is a needed path that was never listed. Every `testpaths`
+entry must ship, read from the pyproject.toml inside the tarball so the check
+depends on the artifact alone. The paths below are needed but not covered by
+`testpaths`.
+"""
+
+import sys
+import tarfile
+import tomllib
+from pathlib import Path
+
+# `_version.py` is gitignored and written into the sdist by the hatch-vcs hook.
+REQUIRED_FILES = ["src/zarr/_version.py", "src/zarr/py.typed"]
+
+# `tests/test_examples.py` globs `examples/` and skips silently if it is empty.
+REQUIRED_DIRS = ["examples"]
+
+# Each subpackage under `packages/` is released as its own distribution.
+FORBIDDEN_PREFIXES = ["packages/"]
+
+
+def read_sdist(sdist: Path) -> tuple[set[str], dict[str, object]]:
+    """Return the sdist's paths, relative to its top-level directory, and its pyproject.toml."""
+    with tarfile.open(sdist) as tar:
+        names = tar.getnames()
+        roots = {name.split("/", 1)[0] for name in names}
+        if len(roots) != 1:
+            raise SystemExit(
+                f"Expected one top-level directory in {sdist.name}, got {sorted(roots)}"
+            )
+        root = roots.pop()
+        pyproject = tar.extractfile(f"{root}/pyproject.toml")
+        if pyproject is None:
+            raise SystemExit(f"pyproject.toml is missing from {sdist.name}")
+        config = tomllib.loads(pyproject.read().decode())
+    members = {name.split("/", 1)[1] for name in names if "/" in name}
+    return members, config
+
+
+def check(sdist: Path) -> int:
+    print(f"Checking {sdist.name}")
+    members, config = read_sdist(sdist)
+    testpaths = config["tool"]["pytest"]["ini_options"]["testpaths"]
+
+    missing_files = [p for p in REQUIRED_FILES if p not in members]
+    missing_dirs = [
+        p for p in [*testpaths, *REQUIRED_DIRS] if not any(m.startswith(f"{p}/") for m in members)
+    ]
+    forbidden = sorted(m for m in members if m.startswith(tuple(FORBIDDEN_PREFIXES)))
+
+    if not (missing_files or missing_dirs or forbidden):
+        print(f"OK ({len(members)} paths)")
+        return 0
+
+    for heading, paths in [
+        ("Missing files", missing_files),
+        ("Missing directories", missing_dirs),
+        ("Paths that must not ship", forbidden),
+    ]:
+        if paths:
+            print(f"\n{heading}:")
+            print("\n".join(paths[:20]))
+            if len(paths) > 20:
+                print(f"... and {len(paths) - 20} more")
+    print("\nFix the allowlist in `[tool.hatch.build.targets.sdist]` in pyproject.toml.")
+    return 1
+
+
+if __name__ == "__main__":
+    if len(sys.argv) != 2:
+        print(__doc__)
+        sys.exit(2)
+    target = Path(sys.argv[1])
+    if not target.exists():
+        print(f"{target} does not exist; build the sdist first with `just build`")
+        sys.exit(2)
+    if target.is_dir():
+        tarballs = sorted(target.glob("*.tar.gz"))
+        if len(tarballs) != 1:
+            print(f"Expected exactly one .tar.gz in {target}, found {len(tarballs)}")
+            sys.exit(2)
+        target = tarballs[0]
+    sys.exit(check(target))