fix(sonarr_database): remove recall_jaccard

with recall jaccard title like "Der Strand" and "Surfing Hippos – Der
Strand der Dschungeltiere" give a false positive match
create match group dict for episode title
This commit is contained in:
lucas committed 2026-11-04 08:31:20 +01:00
1 parent e591602d18
commit d722fe42f0
2 files changed
+33 -398

No files matched your search

+33 -14
View File
@@ -448,22 +448,22 @@ def best_episode_title_match(filename, episodes):
search_words,
candidate_words,
)
recall_jaccard1 = score_match_recall_jaccard(
search_words,
candidate_words,
)
recall_jaccard2 = score_match_recall_jaccard(
candidate_words,
search_words,
)
score = combined_score(
fuzzy,
jaccard,
weight_fuzzy=weight_fuzzy,
weight_jaccard=weight_jaccard,
)
score = max(score, recall_jaccard1, recall_jaccard2)
# recall_jaccard1 = score_match_recall_jaccard(
# search_words,
# candidate_words,
# )
# recall_jaccard2 = score_match_recall_jaccard(
# candidate_words,
# search_words,
# )
# score = max(score, recall_jaccard1, recall_jaccard2)
# if "hyänen" in search_words and score > 0.6:
# logger.warning(
@@ -556,6 +556,7 @@ def translate_series_folder(episodes, series_path, media_data: dict, args):
logger.info("All files renamed")
return
def reference_extra_episodes(episodes, series_path: Path, media_data: dict, args):
logger.info(f"Execute reference_extra_episodes for {media_data['name']}")
extra_episodes_path = series_path / "Extra Episodes"
@@ -603,6 +604,7 @@ def reference_extra_episodes(episodes, series_path: Path, media_data: dict, args
# logger.info(f"{file.name} -> {target}, score={score:.3f}")
safe_rename(file, target, logger=logger)
def recheck_current_episodes(episodes, series_path: Path, media_data: dict, args):
logger.info(f"Execute reference_extra_episodes for {media_data['name']}")
for file in series_path.glob("Season*/*"):
@@ -610,12 +612,21 @@ def recheck_current_episodes(episodes, series_path: Path, media_data: dict, args
continue
logger.trace(f"Reference file: {file.name}")
match = re.search(r" - S\d+E\d+ - (.+)$", file.stem)
title = match.group(1) if match else None
match = re.match(
r"^(?P<show>.+?) - (?P<se>S\d+E\d+) - (?P<title>.+?)(?: (?P<quality>(?:\[[^\]]+\])+))?$",
file.stem,
)
if not match:
logger.warning(f"No match found for {file.name}")
return
title = match["title"]
episode, score = best_episode_title_match(
title,
episodes,
)
quality = match["quality"]
season_episode = match["se"]
if episode is None:
continue
@@ -630,12 +641,17 @@ def recheck_current_episodes(episodes, series_path: Path, media_data: dict, args
"",
episode["name"],
)
new_season_episode = f"S{episode['season']:02d}E{episode['episode']:02d}"
if new_season_episode != season_episode:
logger.warning(
f"Setting new season and episode number: {season_episode} --> {new_season_episode}"
)
new_name = (
f"{media_data['name']}"
f" - S{episode['season']:02d}"
f"E{episode['episode']:02d}"
f" - {safe_title}"
f"{quality}"
f"{file.suffix}"
)
@@ -646,11 +662,12 @@ def recheck_current_episodes(episodes, series_path: Path, media_data: dict, args
logger.trace(f"{file.name} -> {new_name}, score={score:.3f}")
if args.dry_run:
logger.debug(f"{file.name} -> {target}")
logger.debug(f"{file.name} -> {target.relative_to(series_path)}")
continue
# logger.info(f"{file.name} -> {target}, score={score:.3f}")
safe_rename(file, target, logger=logger)
def get_series_data(series_path: Path, database: str = "tvdb"):
series_name, _, tmdb_id = parse_series_name(series_path.name)
@@ -733,6 +750,7 @@ def get_series_data(series_path: Path, database: str = "tvdb"):
return series_data, episodes, media_data
def process_series(series_path: Path, args):
series_name = series_path.name
logger.debug(f"Pricess series with name : {series_name}")
@@ -748,6 +766,7 @@ def process_series(series_path: Path, args):
if args.translate:
translate_series_folder(episodes, series_path, media_data, args)
def main():
parser = argparse.ArgumentParser()
parser.add_argument(