Fix infinite loop in find_unprocessed: check sha256 content (not just path), record dup paths; Optional[int] for max_batches

photo_ingest.py

@@ -42,20 +42,28 @@ def find_unprocessed(base_dir: str, batch_size: int, source: str = None) -> list
     conn = db.get_db()
     batch = []
     for p_str in walk_files(base_dir):
-        # skip if already registered for this source (or any source)
-        row = conn.execute(
-            "SELECT 1 FROM image_hashes WHERE sha256=?",
-            (db.sha256_file(p_str),),
-        ).fetchone() if False else None
         # cheap check: path already known?
         known = conn.execute(
             "SELECT 1 FROM image_hashes WHERE path=?", (p_str,)
         ).fetchone()
         if known:
             continue
+        # content check: sha256 already registered (catches same photo at other paths)
+        sha = db.sha256_file(p_str)
+        sha_known = conn.execute(
+            "SELECT 1 FROM image_hashes WHERE sha256=?", (sha,)
+        ).fetchone()
+        if sha_known:
+            # record this path too, so we don't re-hash it every loop
+            conn.execute(
+                "INSERT OR IGNORE INTO image_hashes (sha256, phash, dhash, file_size, path, source) "
+                "SELECT sha256, phash, dhash, file_size, ?, source FROM image_hashes WHERE sha256=?",
+                (p_str, sha))
+            continue
         batch.append(p_str)
         if len(batch) >= batch_size:
             break
+    conn.commit()
     conn.close()
     print(f"find_unprocessed: {len(batch)} new files (batch_size={batch_size})")
     return batch
@@ -112,7 +120,7 @@ def check_and_register(image_paths: list[str], source: str) -> dict:
 
 @flow(name="photo-ingest")
 def photo_ingest(base_dir: str, source: str = "takeout", batch_size: int = 1000,
-                 max_batches: int = None):
+                 max_batches: int | None = None):
     """Hash + dedup-check a folder against the persistent library, in batches.
 
     Args: