initial commit

backend/app/services/scanner.py

@@ -0,0 +1,91 @@
+import hashlib
+from pathlib import Path
+from datetime import datetime
+from sqlalchemy.ext.asyncio import AsyncSession
+from sqlalchemy import select
+from app.models import MediaItem
+
+IMAGE_EXTENSIONS = {".jpg", ".jpeg", ".png", ".gif", ".webp", ".bmp", ".tiff", ".avif", ".heic"}
+VIDEO_EXTENSIONS = {".mp4", ".mkv", ".mov", ".avi", ".webm", ".m4v", ".flv", ".wmv", ".ts"}
+
+
+def classify(path: Path) -> str | None:
+    ext = path.suffix.lower()
+    if ext in IMAGE_EXTENSIONS:
+        return "image"
+    if ext in VIDEO_EXTENSIONS:
+        return "video"
+    return None
+
+
+def hash_file(path: Path) -> str:
+    h = hashlib.sha256()
+    with path.open("rb") as f:
+        for chunk in iter(lambda: f.read(65536), b""):
+            h.update(chunk)
+    return h.hexdigest()
+
+
+async def scan_library(library_id: int, library_path: str, db: AsyncSession) -> None:
+    root = Path(library_path)
+    existing = await db.execute(
+        select(MediaItem).where(MediaItem.library_id == library_id)
+    )
+    db_items = {item.rel_path: item for item in existing.scalars().all()}
+
+    seen_paths: set[str] = set()
+
+    for file_path in root.rglob("*"):
+        if not file_path.is_file():
+            continue
+        media_type = classify(file_path)
+        if not media_type:
+            continue
+
+        rel = str(file_path.relative_to(root))
+        seen_paths.add(rel)
+
+        if rel in db_items:
+            item = db_items[rel]
+            if item.missing:
+                item.missing = False
+                item.updated_at = datetime.utcnow()
+        else:
+            file_hash = hash_file(file_path)
+            # Check if this hash matches an orphaned (missing) item — file was moved while offline
+            moved = await _find_by_hash(library_id, file_hash, db)
+            if moved:
+                moved.rel_path = rel
+                moved.filename = file_path.name
+                moved.missing = False
+                moved.updated_at = datetime.utcnow()
+            else:
+                item = MediaItem(
+                    library_id=library_id,
+                    rel_path=rel,
+                    filename=file_path.name,
+                    file_hash=file_hash,
+                    media_type=media_type,
+                    size_bytes=file_path.stat().st_size,
+                    missing=False,
+                )
+                db.add(item)
+
+    # Mark items no longer on disk as missing
+    for rel_path, item in db_items.items():
+        if rel_path not in seen_paths and not item.missing:
+            item.missing = True
+            item.updated_at = datetime.utcnow()
+
+    await db.commit()
+
+
+async def _find_by_hash(library_id: int, file_hash: str, db: AsyncSession) -> MediaItem | None:
+    result = await db.execute(
+        select(MediaItem).where(
+            MediaItem.library_id == library_id,
+            MediaItem.file_hash == file_hash,
+            MediaItem.missing == True,  # noqa: E712
+        )
+    )
+    return result.scalars().first()