Remove-from-catalog: a third durable curation store

Titles that aren't games (misread box art, out-of-scope items) can now
be removed outright: a danger button in the catalog's edit panel posts
/api/remove-title, which drops the line's matches rows (veto'd ones
too — removal is the human explicitly discarding the line), records the
decision photo-scoped in data/title_removals.json, and replays
titles.json. Every rebuild filters removed sightings after edits and
before dedupe, so re-extraction cannot resurrect them; undo by deleting
the record from the store. The three stores now share one scoped-record
parser and recorder.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016jXZFSTZQKzAC8fqpWSz9g

tests/test_extract.py

@@ -24,6 +24,7 @@ from bggpipe.extract import (
     replay_titles,
     run_extract,
 )
+from bggpipe.normalize import normalize_title
 
 
 def _write_image(path, size=(400, 300), fmt="JPEG", color=(200, 30, 30)):
@@ -234,6 +235,22 @@ def test_edits_chain_and_target_photos():
     assert entries[1] == _entry("Wiz-War", "b.jpg")  # untargeted copy untouched
 
 
+def test_removals_filter_rebuilds_photo_scoped():
+    from bggpipe.extract import apply_title_removals
+
+    entries = [
+        _entry("Not A Game", "a.jpg"),
+        _entry("Not A Game", "b.jpg"),
+        _entry("Catan", "a.jpg"),
+    ]
+    removals = [{"norm": normalize_title("Not A Game"), "photos": {"a.jpg"}}]
+    kept = apply_title_removals(entries, removals)
+    assert [(e["title_raw"], e["source_photos"]) for e in kept] == [
+        ("Not A Game", ["b.jpg"]),  # other sighting untouched
+        ("Catan", ["a.jpg"]),
+    ]
+
+
 def test_stores_roundtrip_and_replay_from_raw(tmp_path):
     cfg = Config(data_dir=tmp_path / "data", photos_dir=tmp_path / "photos")
     raw = cfg.extract_raw_dir

tests/test_webreview.py

@@ -726,3 +726,43 @@ def test_single_photo_rowless_entry_is_not_splittable(tmp_path):
         if c["title_raw"] == "Fresh Off The Shelf"
     )
     assert line["can_split"] is False
+
+
+def test_remove_title_is_durable_and_drops_vetoed_rows(tmp_path):
+    cfg = make_cfg(tmp_path)
+    rows = read_matches(cfg.matches_path)
+    rows.append(
+        _row(
+            title_raw="Citadels",
+            match_status="approved",
+            bgg_id="478",
+            source_photos="shelf.jpg",
+            dedupe_veto="1",
+        )
+    )
+    write_matches(cfg.matches_path, rows)
+    web = TestClient(create_app(cfg, client=unauthorized_client(tmp_path)))
+    res = web.post(
+        "/api/remove-title",
+        json={"title_raw": "Citadels", "source_photos": "shelf.jpg"},
+    )
+    assert res.status_code == 200
+    assert "Citadels" not in [c["title_raw"] for c in res.json()["catalog"]]
+    # removal is explicit: the veto'd row goes too (unlike edits)
+    assert not any(r["title_raw"] == "Citadels" for r in read_matches(cfg.matches_path))
+    (record,) = json.loads(cfg.title_removals_path.read_text())
+    assert record == {"title": "Citadels", "photos": ["shelf.jpg"]}
+    # durable: a fresh replay keeps it gone; the sibling title is untouched
+    from bggpipe.extract import replay_titles
+
+    replay_titles(cfg)
+    titles = {e["title_raw"] for e in json.loads(cfg.titles_path.read_text())}
+    assert titles == {"Fresh Off The Shelf"}
+    # unknown titles still 404
+    assert (
+        web.post(
+            "/api/remove-title",
+            json={"title_raw": "No Such Game", "source_photos": "x.jpg"},
+        ).status_code
+        == 404
+    )