fix(node): skip indexing audio files under 50KB, likely-corrupt source

The "P.H. Theme" failure investigated earlier turned out to be a genuinely
corrupt 1256-byte source file with no audio stream at all, just an ID3
tag — a real, if rare, corruption pattern worth guarding against directly
rather than only handling gracefully at playback time. Scoped to audio
only, applied wherever a file actually gets hashed/typed (fresh scan and
the cache-miss rehash path alike) — a tiny file of any other type is still
indexed normally.

1 files changed, 36 insertions, 1 deletions

diff --git a/packages/meshbay-node/tests/test_indexer.py b/packages/meshbay-node/tests/test_indexer.py
index 56571a4..4bae620 100644
--- a/packages/meshbay-node/tests/test_indexer.py
+++ b/packages/meshbay-node/tests/test_indexer.py
@@ -28,7 +28,11 @@ def shared_dir(tmp_path):
     d = tmp_path / "shared"
     d.mkdir()
     (d / "video.mkv").write_bytes(os.urandom(1024))
-    (d / "music.mp3").write_bytes(os.urandom(512))
+    # Above MIN_AUDIO_SIZE_BYTES: these tests use this file as a generic
+    # "here's an audio entry" stand-in, not as a test of the tiny-file gate
+    # itself (see test_is_indexable_size_* / test_initial_scan_skips_a_
+    # corrupt_tiny_audio_file below for that).
+    (d / "music.mp3").write_bytes(os.urandom(indexer_mod.MIN_AUDIO_SIZE_BYTES + 1024))
     (d / "readme.md").write_bytes(b"# Hello MeshBay")
     subdir = d / "docs"
     subdir.mkdir()
@@ -157,6 +161,37 @@ def test_type_detection_covers_wma_and_musepack():
     assert indexer_mod._detect_type(Path("track.mpc")) == "audio"
 
 
+def test_is_indexable_size_rejects_tiny_audio(tmp_path):
+    assert indexer_mod._is_indexable_size(Path("track.mp3"), 1256) is False
+    assert indexer_mod._is_indexable_size(
+        Path("track.mp3"), indexer_mod.MIN_AUDIO_SIZE_BYTES) is True
+    assert indexer_mod._is_indexable_size(
+        Path("track.mp3"), indexer_mod.MIN_AUDIO_SIZE_BYTES - 1) is False
+
+
+def test_is_indexable_size_does_not_apply_to_other_types():
+    """A tiny document/image is still worth indexing — the rule exists
+    because a truncated *audio* file is a known corruption signature, not
+    because small files in general are suspect."""
+    assert indexer_mod._is_indexable_size(Path("readme.txt"), 10) is True
+    assert indexer_mod._is_indexable_size(Path("icon.png"), 10) is True
+
+
+@pytest.mark.asyncio
+async def test_initial_scan_skips_a_corrupt_tiny_audio_file(tmp_path, sk_node, gek):
+    d = tmp_path / "shared"
+    d.mkdir()
+    (d / "good.mp3").write_bytes(os.urandom(indexer_mod.MIN_AUDIO_SIZE_BYTES + 1))
+    (d / "corrupt.mp3").write_bytes(os.urandom(1256))
+
+    indexer = DirectoryIndexer(roots=one_root(d), group_id="g", sk_node=sk_node, gek=gek)
+    await indexer.initial_scan()
+
+    names = {e.name for e in indexer.index.entries}
+    assert "good.mp3" in names
+    assert "corrupt.mp3" not in names
+
+
 @pytest.mark.asyncio
 async def test_hidden_files_excluded(tmp_path, sk_node, gek):
     d = tmp_path / "dir"