feat: bootstrap NAS music deduper

am_deduper/scanner.py

@@ -0,0 +1,47 @@
+from __future__ import annotations
+
+import os
+from pathlib import Path
+from typing import Iterator
+
+from .constants import MUSIC_EXTENSIONS
+from .db import connect
+from .hashing import fast_hash
+
+
+def iter_music(root: Path) -> Iterator[Path]:
+    for path in root.rglob("*"):
+        if not path.is_file() or path.is_symlink():
+            continue
+        if path.name.startswith(".") or ".dedupe" in path.parts:
+            continue
+        if path.suffix.lower() in MUSIC_EXTENSIONS:
+            yield path
+
+
+def index_file(conn, path: Path) -> None:
+    st = path.stat()
+    fh = fast_hash(path)
+    conn.execute(
+        """
+        INSERT INTO audio_tracks(file_path, file_size, fast_hash, inode, device, status)
+        VALUES (?, ?, ?, ?, ?, 'active')
+        ON CONFLICT(file_path) DO UPDATE SET
+          file_size=excluded.file_size,
+          fast_hash=excluded.fast_hash,
+          inode=excluded.inode,
+          device=excluded.device,
+          updated_at=CURRENT_TIMESTAMP,
+          status='active'
+        """,
+        (str(path), st.st_size, fh, st.st_ino, st.st_dev),
+    )
+
+
+def scan(root: Path, db_path: Path) -> int:
+    count = 0
+    with connect(db_path) as conn:
+        for path in iter_music(root):
+            index_file(conn, path)
+            count += 1
+    return count