From a776722d5383b8c6d7496cdf05fa23ca49af1d42 Mon Sep 17 00:00:00 2001
From: Sam Rolfe <samuelrolfe@gmail.com>
Date: Sat, 8 Aug 2026 14:34:30 +1000
Subject: [PATCH] Fix PIL import in check_and_register (real infinite-loop
 cause); known_paths table for dup-path tracking

---
 photo_ingest.py | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/photo_ingest.py b/photo_ingest.py
index 9d15528..079768f 100644
--- a/photo_ingest.py
+++ b/photo_ingest.py
@@ -15,6 +15,7 @@ use walk_files for a streaming generator when batch_size is set.
 from pathlib import Path
 
 from prefect import flow, task
+from PIL import Image
 
 import photo_db as db
 
@@ -40,11 +41,15 @@ def find_unprocessed(base_dir: str, batch_size: int, source: str = None) -> list
     """Find the next batch of files NOT yet in the fingerprint DB."""
     db.init_db()
     conn = db.get_db()
+    conn.execute(
+        "CREATE TABLE IF NOT EXISTS known_paths (path TEXT PRIMARY KEY, sha256 TEXT NOT NULL)"
+    )
     batch = []
     for p_str in walk_files(base_dir):
-        # cheap check: path already known?
+        # cheap check: path already seen (image_hashes OR known_paths)
         known = conn.execute(
-            "SELECT 1 FROM image_hashes WHERE path=?", (p_str,)
+            "SELECT 1 FROM image_hashes WHERE path=? UNION SELECT 1 FROM known_paths WHERE path=?",
+            (p_str, p_str),
         ).fetchone()
         if known:
             continue
@@ -54,10 +59,9 @@ def find_unprocessed(base_dir: str, batch_size: int, source: str = None) -> list
             "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
+            # record this path in known_paths 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=?",
+                "INSERT OR IGNORE INTO known_paths (path, sha256) VALUES (?,?)",
                 (p_str, sha))
             continue
         batch.append(p_str)
@@ -95,7 +99,7 @@ def check_and_register(image_paths: list[str], source: str) -> dict:
                 near_dups.append((p_str, near, dist))
                 continue
             # new — register
-            with __import__("PIL.Image", fromlist=["Image"]).Image.open(p) as im:
+            with Image.open(p) as im:
                 w, h = im.size
             conn.execute(
                 "INSERT INTO image_hashes (sha256, phash, dhash, file_size, width, height, path, source) "