Harden runtime distribution acceptance

tools/deployment/build-deployer-zipapp.py

@@ -6,11 +6,13 @@ from __future__ import annotations
 import argparse
 from hashlib import sha256
 from pathlib import Path
-import zipapp
+from zipfile import ZIP_DEFLATED, ZipFile, ZipInfo
 
 
 ROOT = Path(__file__).resolve().parent
 DEFAULT_OUTPUT = ROOT.parent.parent / "runtime" / "deployment" / "govoplan-deploy.pyz"
+ZIP_TIMESTAMP = (1980, 1, 1, 0, 0, 0)
+PYTHON_FILE_MODE = 0o100644
 
 
 def main(argv: list[str] | None = None) -> int:
@@ -25,13 +27,7 @@ def main(argv: list[str] | None = None) -> int:
     temporary = output.with_name(f".{output.name}.tmp")
     if temporary.exists():
         temporary.unlink()
-    zipapp.create_archive(
-        ROOT,
-        target=temporary,
-        interpreter="/usr/bin/env python3",
-        compressed=True,
-        filter=_include_source,
-    )
+    _write_reproducible_zipapp(temporary)
     temporary.chmod(0o755)
     temporary.replace(output)
     digest = sha256(output.read_bytes()).hexdigest()
@@ -39,6 +35,42 @@ def main(argv: list[str] | None = None) -> int:
     return 0
 
 
+def _write_reproducible_zipapp(target: Path) -> None:
+    sources = tuple(
+        path
+        for path in sorted(ROOT.rglob("*"), key=lambda item: item.as_posix())
+        if path.is_file() and _include_source(path.relative_to(ROOT))
+    )
+    if not any(path.relative_to(ROOT).as_posix() == "__main__.py" for path in sources):
+        raise ValueError("deployment source has no __main__.py")
+    for path in sources:
+        if path.is_symlink():
+            raise ValueError(f"deployment source must not contain symlinks: {path}")
+
+    with target.open("wb") as handle:
+        handle.write(b"#!/usr/bin/env python3\n")
+        with ZipFile(
+            handle,
+            mode="w",
+            compression=ZIP_DEFLATED,
+            compresslevel=9,
+            strict_timestamps=True,
+        ) as archive:
+            for source in sources:
+                relative = source.relative_to(ROOT).as_posix()
+                info = ZipInfo(relative, date_time=ZIP_TIMESTAMP)
+                info.compress_type = ZIP_DEFLATED
+                info.create_system = 3
+                info.external_attr = PYTHON_FILE_MODE << 16
+                info.flag_bits |= 0x800
+                archive.writestr(
+                    info,
+                    source.read_bytes(),
+                    compress_type=ZIP_DEFLATED,
+                    compresslevel=9,
+                )
+
+
 def _include_source(path: Path) -> bool:
     return (
         "__pycache__" not in path.parts