Skip to content

fix(novel2movie): only resume retrieval when score chunk files exist#72

Open
santhreal wants to merge 2 commits into
HKUDS:mainfrom
santhreal:fix/novel2movie-resume-cache-miss
Open

fix(novel2movie): only resume retrieval when score chunk files exist#72
santhreal wants to merge 2 commits into
HKUDS:mainfrom
santhreal:fix/novel2movie-resume-cache-miss

Conversation

@santhreal

Copy link
Copy Markdown

Resume treated any non-empty event_* directory as a retrieval cache hit, then parsed -score_ out of every filename. A stray notes.txt raised IndexError. Skipping unmatched names alone still left a stray-only directory as an empty hit and skipped retrieval.

iter_score_chunk_files now selects only *-score_*.txt entries with a parseable score, and resume runs only when at least one such file exists.

Repro: a cache dir containing only notes.txt must not IndexError and must not count as a hit.

Stray notes.txt in event_* made resume IndexError; skipping names alone
still treated the dir as a hit. Require at least one score file.
Keep cache I/O in utils.novel2movie_cache and skip unparsable scores.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant