Add scanner tests with hash-based relocation

- Add CC0 test MP3 fixtures from SoundSafari/CC0-1.0-Music
- Add comprehensive scanner tests including:
  - Hash computation consistency
  - Metadata extraction
  - File discovery in subdirectories
  - Hash-based relocation detection
  - Missing/recovered file handling
  - Multi-file relocation scenarios
- Add file_hash index migration for faster lookups
- Add pytest-asyncio dependency

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: jeffcrouse

backend/app/db/models.py

@@ -168,7 +168,7 @@ class Track(Base):
 
     id: Mapped[UUID] = mapped_column(primary_key=True, default=uuid4)
     file_path: Mapped[str] = mapped_column(String(1000), unique=True, nullable=False)
-    file_hash: Mapped[str] = mapped_column(String(64), nullable=False)
+    file_hash: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
 
     # Basic metadata from tags
     title: Mapped[str | None] = mapped_column(String(500))

backend/app/services/scanner.py

@@ -237,7 +237,15 @@ async def scan(self, library_path: Path, full_scan: bool = False) -> dict[str, A
         logger.info("Loading existing tracks from database...")
         existing_tracks = await self._get_existing_tracks()
         existing_paths = {t.file_path: t for t in existing_tracks}
-        logger.info(f"Found {len(existing_tracks)} existing tracks in database")
+
+        # Build hash lookup for detecting relocated files (same content, different path)
+        # First track wins in case of hash collisions (shouldn't happen normally)
+        existing_hashes: dict[str, Track] = {}
+        for t in existing_tracks:
+            if t.file_hash and t.file_hash not in existing_hashes:
+                existing_hashes[t.file_hash] = t
+
+        logger.info(f"Found {len(existing_tracks)} existing tracks in database ({len(existing_hashes)} unique hashes)")
 
         # Find all audio files (in thread pool to not block)
         logger.info(f"Discovering audio files in {library_path}...")
@@ -296,12 +304,29 @@ def discovery_progress(dirs_scanned: int, files_found: int):
             )
 
             if path_str not in existing_paths:
-                # New file
-                logger.info(f"NEW: {file_path.name}")
-                track = await self._create_track(file_path, file_hash, file_mtime)
-                pending_analysis_ids.append(str(track.id))
-                results["new"] += 1
-                results["queued"] += 1
+                # Path not found - check if same file exists at different path (by hash)
+                if file_hash in existing_hashes:
+                    # Found by hash - this is a relocated file, update its path
+                    existing = existing_hashes[file_hash]
+                    old_path = existing.file_path
+                    logger.info(f"RELOCATED (by hash): {Path(old_path).name} -> {path_str}")
+                    existing.file_path = path_str
+                    existing.status = TrackStatus.ACTIVE
+                    existing.missing_since = None
+                    results["relocated"] += 1
+                    # Update path lookup so we don't process old path as missing
+                    existing_paths[path_str] = existing
+                    if old_path in existing_paths:
+                        del existing_paths[old_path]
+                else:
+                    # Truly new file
+                    logger.info(f"NEW: {file_path.name}")
+                    track = await self._create_track(file_path, file_hash, file_mtime)
+                    pending_analysis_ids.append(str(track.id))
+                    results["new"] += 1
+                    results["queued"] += 1
+                    # Add to hash lookup so subsequent files with same hash are detected
+                    existing_hashes[file_hash] = track
             else:
                 existing = existing_paths[path_str]
 

backend/migrations/versions/20251231_145519_add_file_hash_index.py

@@ -0,0 +1,24 @@
+"""add file_hash index
+
+Revision ID: 54718ec827f6
+Revises: baseline
+Create Date: 2025-12-31 14:55:19.935331
+"""
+from typing import Sequence, Union
+
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = '54718ec827f6'
+down_revision: Union[str, None] = 'baseline'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # Add index on file_hash for hash-based track matching
+    op.create_index(op.f('ix_tracks_file_hash'), 'tracks', ['file_hash'], unique=False)
+
+
+def downgrade() -> None:
+    op.drop_index(op.f('ix_tracks_file_hash'), table_name='tracks')

backend/pyproject.toml

@@ -103,6 +103,7 @@ ignore_errors = true
 
 [tool.pytest.ini_options]
 asyncio_mode = "auto"
+asyncio_default_fixture_loop_scope = "function"
 testpaths = ["tests"]
 
 [tool.uv]