feat(attachments): fail-soft archive_extractor (zip/cbz/rar/7z)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

tests/test_archive_extractor.py

@@ -0,0 +1,64 @@
+import zipfile
+from pathlib import Path
+
+import pytest
+
+from backend.app.services.archive_extractor import extract_archive, is_archive
+
+
+def test_is_archive():
+    assert is_archive(Path("a.zip"))
+    assert is_archive(Path("a.CBZ"))
+    assert is_archive(Path("a.rar"))
+    assert is_archive(Path("a.7z"))
+    assert not is_archive(Path("a.jpg"))
+
+
+def _zip(path, entries):
+    with zipfile.ZipFile(path, "w") as zf:
+        for name, data in entries.items():
+            zf.writestr(name, data)
+
+
+def test_zip_members_enumerated(tmp_path):
+    z = tmp_path / "set.zip"
+    _zip(z, {"a.jpg": b"img", "sub/b.png": b"img2", "notes.txt": b"hi"})
+    with extract_archive(z) as members:
+        names = sorted(n for n, _ in members)
+        assert names == ["a.jpg", "notes.txt", "sub/b.png"]
+        for _, p in members:
+            assert p.is_file()
+    for _, p in members:
+        assert not p.exists()  # temp cleaned on exit
+
+
+def test_cbz_alias(tmp_path):
+    z = tmp_path / "set.cbz"
+    _zip(z, {"p1.jpg": b"i"})
+    with extract_archive(z) as members:
+        assert [n for n, _ in members] == ["p1.jpg"]
+
+
+def test_corrupt_archive_yields_nothing(tmp_path):
+    bad = tmp_path / "broken.zip"
+    bad.write_bytes(b"not a zip at all")
+    with extract_archive(bad) as members:
+        assert members == []
+
+
+def test_7z_roundtrip(tmp_path):
+    py7zr = pytest.importorskip("py7zr")
+    a = tmp_path / "s.7z"
+    (tmp_path / "x.jpg").write_bytes(b"img")
+    with py7zr.SevenZipFile(a, "w") as zf:
+        zf.write(tmp_path / "x.jpg", "x.jpg")
+    with extract_archive(a) as members:
+        assert [n for n, _ in members] == ["x.jpg"]
+
+
+def test_rar_corrupt_is_failsoft(tmp_path):
+    pytest.importorskip("rarfile")
+    bad = tmp_path / "b.rar"
+    bad.write_bytes(b"Rar! not really")
+    with extract_archive(bad) as members:
+        assert members == []  # invalid / no unrar → fail-soft, nothing lost