feat(ebook): enhance phrase judgment logging with failure tracking

python/ebook_search/api/routes/admin.py

@@ -121,9 +121,11 @@ def judge_ngrams(request: Request, config: AppConfig, session: DbSession) -> HTM
         return templates.TemplateResponse(request, "partials/error.html", {"message": str(error)}, status_code=500)
 
     logger.info(
-        "ebook_admin_judge_ngrams_complete books_seen=%s books_judged=%s candidates_judged=%s protected=%s mentions=%s",
+        "ebook_admin_judge_ngrams_complete books_seen=%s books_judged=%s books_failed=%s candidates_judged=%s "
+        "protected=%s mentions=%s",
         result.books_seen,
         result.books_judged,
+        result.books_failed,
         result.candidates_judged,
         result.protected_phrases,
         result.phrase_mentions,
@@ -136,6 +138,7 @@ def judge_ngrams(request: Request, config: AppConfig, session: DbSession) -> HTM
                 f"Judged {result.candidates_judged} candidates across {result.books_judged} of "
                 f"{result.books_seen} books; {result.protected_phrases} protected phrases, "
                 f"{result.phrase_mentions} mentions"
+                + (f"; {result.books_failed} books failed" if result.books_failed else "")
             )
         },
     )

python/ebook_search/protected_phrases/lib.py

@@ -207,6 +207,7 @@ class PhraseJudgmentBackfillResult:
 
     books_seen: int
     books_judged: int
+    books_failed: int
     candidates_judged: int
     protected_phrases: int
     phrase_mentions: int
@@ -1208,6 +1209,7 @@ def judge_candidate_phrases_for_books(
     sources = list(session.scalars(select(EbookSource).order_by(EbookSource.id)))
     books_seen = len(sources)
     books_judged = 0
+    books_failed = 0
     candidates_judged = 0
     protected_count = 0
     mention_count = 0
@@ -1282,13 +1284,15 @@ def judge_candidate_phrases_for_books(
             session.commit()
         except Exception:
             session.rollback()
+            books_failed += 1
             logger.exception(
-                "ebook_candidate_phrase_judgment_book_failed source_id=%s book_number=%s books_seen=%s",
+                "ebook_candidate_phrase_judgment_book_failed source_id=%s book_number=%s books_seen=%s books_failed=%s",
                 source.id,
                 book_number,
                 books_seen,
+                books_failed,
             )
-            raise
+            continue
         books_judged += 1
         candidates_judged += judged
         protected_count += len(protected)
@@ -1307,10 +1311,11 @@ def judge_candidate_phrases_for_books(
         )
 
     logger.info(
-        "ebook_candidate_phrase_judgment_complete books_seen=%s books_judged=%s candidates_judged=%s "
-        "protected=%s mentions=%s",
+        "ebook_candidate_phrase_judgment_complete books_seen=%s books_judged=%s books_failed=%s "
+        "candidates_judged=%s protected=%s mentions=%s",
         books_seen,
         books_judged,
+        books_failed,
         candidates_judged,
         protected_count,
         mention_count,
@@ -1318,6 +1323,7 @@ def judge_candidate_phrases_for_books(
     return PhraseJudgmentBackfillResult(
         books_seen=books_seen,
         books_judged=books_judged,
+        books_failed=books_failed,
         candidates_judged=candidates_judged,
         protected_phrases=protected_count,
         phrase_mentions=mention_count,