feat: bootstrap NAS music deduper

tests/test_exact.py

@@ -0,0 +1,22 @@
+from pathlib import Path
+
+from am_deduper.db import connect
+from am_deduper.exact import exact_dedupe
+from am_deduper.scanner import scan
+
+
+def test_exact_dedupe_creates_hardlink(tmp_path: Path):
+    root = tmp_path / "AllMusic"
+    (root / "admin").mkdir(parents=True)
+    (root / "alice").mkdir(parents=True)
+    a = root / "admin" / "song.flac"
+    b = root / "alice" / "song.flac"
+    data = b"fake flac content" * 10000
+    a.write_bytes(data)
+    b.write_bytes(data)
+    db = root / ".dedupe" / "dedupe.db"
+    assert scan(root, db) == 2
+    actions = exact_dedupe(db, root)
+    assert len(actions) == 1
+    assert a.stat().st_ino == b.stat().st_ino
+    assert a.read_bytes() == b.read_bytes()

tests/test_hashing.py

@@ -0,0 +1,21 @@
+from pathlib import Path
+
+from am_deduper.hashing import fast_hash, sha256
+
+
+def test_fast_hash_same_content(tmp_path: Path):
+    a = tmp_path / "a.bin"
+    b = tmp_path / "b.bin"
+    data = b"x" * 200_000
+    a.write_bytes(data)
+    b.write_bytes(data)
+    assert fast_hash(a) == fast_hash(b)
+    assert sha256(a) == sha256(b)
+
+
+def test_fast_hash_uses_size(tmp_path: Path):
+    a = tmp_path / "a.bin"
+    b = tmp_path / "b.bin"
+    a.write_bytes(b"x" * 100)
+    b.write_bytes(b"x" * 101)
+    assert fast_hash(a) != fast_hash(b)