From 80e6587888ae18032549e6be319c079218f6db40 Mon Sep 17 00:00:00 2001 From: Lucas Vivo Date: Sat, 31 Oct 2026 12:21:31 +0100 Subject: [PATCH] fix(audiobook_library): use mutagen mp4/mp3 split for write/read metadata --- .../metadata_audiobooks_library.py | 148 +++++++++--------- 1 file changed, 73 insertions(+), 75 deletions(-) diff --git a/audiobook_metadata/metadata_audiobooks_library.py b/audiobook_metadata/metadata_audiobooks_library.py index 3a09e49..84fa228 100755 --- a/audiobook_metadata/metadata_audiobooks_library.py +++ b/audiobook_metadata/metadata_audiobooks_library.py @@ -10,12 +10,12 @@ import tempfile import sqlite3 from collections import defaultdict from concurrent.futures import ThreadPoolExecutor, as_completed -from pathlib import Path - from serverlibrary import * from mutagen.easyid3 import EasyID3 -from mutagen.id3 import ID3, ID3NoHeaderError +from pathlib import Path +from mutagen.id3 import ID3NoHeaderError from mutagen.mp4 import MP4 +from mutagen import File logging.setLoggerClass(CustomLogger) logger: CustomLogger = logging.getLogger(__name__) # type: ignore[assignment] @@ -49,11 +49,9 @@ series_patterns = [ rf"^(.+?){SEP}(?:Book|Band|Volume)\s+(\d+)(?:{SEP}(.+))$", ] - def ffmpeg_escape(path: Path) -> str: return str(path.resolve()).replace("'", r"'\''") - def clean_genre(genre): if not genre: return "" @@ -101,7 +99,6 @@ def clean_genre(genre): logger.debug(f"Update genre from {genre_old} to {genre}") return genre - def ffmetadata_escape(value: str) -> str: return ( value.replace("\\", "\\\\") @@ -111,7 +108,6 @@ def ffmetadata_escape(value: str) -> str: .replace("\n", "\\\n") ) - def get_chapters(media: Path) -> list[dict]: result = subprocess.run( [ @@ -131,72 +127,45 @@ def get_chapters(media: Path) -> list[dict]: data = json.loads(result.stdout) return data.get("chapters", []) +def _to_str(value): + if isinstance(value, (list, tuple)): + return "; ".join(map(str, value)) + return str(value) def get_metadata_all(file: Path) -> dict[str, str]: - metadata = {} + metadata = dict.fromkeys(metadata_fields, "") try: - try: - tags = EasyID3(file) - except ID3NoHeaderError: - tags = {} + audio = File(file) - metadata = { - key: "; ".join(value) if isinstance(value, list) else str(value) - for key, value in tags.items() - } + if audio is None or audio.tags is None: + return metadata - except Exception as error: - print(f"Error getting metadata for {file}: {error}") + # MP4 / M4A / M4B + if isinstance(audio, MP4): + for field, mp4_key in MP4_MAP.items(): + value = audio.tags.get(mp4_key) + if value: + metadata[field] = _to_str(value) - for field in metadata_fields: - metadata.setdefault(field, "") + # Everything else (MP3, FLAC, Opus, Ogg...) + else: + for field in metadata_fields: + value = audio.tags.get(field) + if value: + metadata[field] = _to_str(value) + + except (ID3NoHeaderError, Exception) as error: + logger.exception(f"Error getting metadata for {file}: {error}") return metadata - -def get_metadata_fields(file: Path, fields) -> dict[str, str]: - metadata = dict.fromkeys(fields, "") - - try: - tags = EasyID3(file) - - for field in fields: - if field in tags: - metadata[field] = tags[field][0] - - except Exception as error: - print(f"Error getting metadata for {file}: {error}") - - return metadata - - -def get_duration_ms(media: Path) -> int: - duration = subprocess.check_output( - [ - "ffprobe", - "-v", - "error", - "-show_entries", - "format=duration", - "-of", - "default=noprint_wrappers=1:nokey=1", - str(media), - ], - text=True, - ).strip() - - return int(float(duration) * 1000) - - def write_metadata(file, fields, values, dry_run=False): - logger.debug(f"Writing metadata for {file}") changes = { field.lower().replace(" ", ""): str(value) for field, value in zip(fields, values) } - - logger.trace(f"Changes: {changes}") + logger.debug(f"Writing metadata for {file}: {changes}") if dry_run: return @@ -238,6 +207,22 @@ def write_metadata(file, fields, values, dry_run=False): except Exception as e: logger.exception(f"Error writing metadata for {file}: {e}") +def get_duration_ms(media: Path) -> int: + duration = subprocess.check_output( + [ + "ffprobe", + "-v", + "error", + "-show_entries", + "format=duration", + "-of", + "default=noprint_wrappers=1:nokey=1", + str(media), + ], + text=True, + ).strip() + + return int(float(duration) * 1000) def normalize_name(name: str, series: str) -> str: # Remove anything in (), [], {} @@ -246,7 +231,7 @@ def normalize_name(name: str, series: str) -> str: # name = re.sub( # rf"^(?:{re.escape(series)}){SEP}|{SEP}(?:{re.escape(series)})$", # "", name, flags=re.IGNORECASE, ).strip() - name = re.sub(r"\s*[\(\[\{].*?[\)\]\}]\s*", " ", name).strip() + name = re.sub(r"\s*[(\[{].*?[)\]}]\s*", " ", name).strip() name = re.sub(r"\s+", " ", name) for pattern in series_patterns: @@ -273,7 +258,6 @@ def normalize_name(name: str, series: str) -> str: return name_out1 return name_out - def cleanup(text: str) -> str: # Collapse repeated punctuation groups like ", ,", ". .", "! !", "? ?" while True: @@ -287,13 +271,11 @@ def cleanup(text: str) -> str: return text.strip() - def get_empty_keys(data: dict) -> str: empty_keys = [str(key) for key, value in data.items() if not value] return f"Empty: {', '.join(empty_keys)}" if empty_keys else "" - def get_codec_bitrate(file: Path) -> tuple[str, str]: info = json.loads( subprocess.check_output( @@ -319,7 +301,6 @@ def get_codec_bitrate(file: Path) -> tuple[str, str]: bitrate_k = f"{bitrate // 1000}k" return stream["codec_name"].lower(), bitrate_k # codec, bits/s - class Audiobook: def __init__(self, book_path: Path, library_root_path: Path) -> None: self.title: str | list = "" @@ -434,12 +415,12 @@ class Audiobook: ).strip() patterns = [ - # 01 Title, 02GLCK~1.WMA - r"^(\d{2,3})(?=\D|$)", + # 0099_Apollo, 01 Title, 001 Chapter, 02GLCK~1.WMA + r"^(\d+)(?=\D|$)", # CD1-01, CD 1 - 01, CD1 - 01. r"CD\s*(\d+)\s*-\s*(\d{1,3})", - # 018-01, 018 - 01, 138-01, 01-01 - r"^(\d{1,3})\s*-\s*(\d{1,3})(?:\D|$)", + # 018-01, 018 - 01, 138-01, 0001-0999 + r"^(\d+)\s*-\s*(\d+)(?:\D|$)" # Track 01 r"Track\s+(\d+)$", # 01 - Kapitel 01 @@ -560,6 +541,13 @@ class Audiobook: f"[Updated] artist {self.artist} found in path for {self.bookPath}" ) + def fix_composer_from_narrator(self): + if not self.metadata.get("narrator"): + logger.trace(f"No Narrator found in metadata for {self.name}") + return + if not self.composer or len(self.composer) < 5: + self.composer = self.metadata.get("narrator") + def fix_composer_from_comment(self): m = re.search( r"(?:read\s+by:?|" @@ -658,22 +646,25 @@ class Audiobook: def move(self, desired_path: Path, desired_name_stem: str, dry_run=False): logger.trace(f"Moving {self.bookPath} to {desired_path}") - if dry_run: - return desired_path.mkdir(parents=True, exist_ok=True) padding = len(str(len(self.files))) once = True for number, file in enumerate(self.files, start=1): if self.singleFile: - new_name = f"{desired_name_stem}{file.suffix.lower()}" + new_name = desired_path / f"{desired_name_stem}{file.suffix.lower()}" else: - new_name = f"{number:0{padding}d} - {desired_name_stem}{file.suffix}" + new_name = desired_path / f"{number:0{padding}d} - {desired_name_stem}{file.suffix}" + + if file == new_name: + continue if once: once = False logger.debug(f"Moving {file} to {new_name}") else: logger.trace(f"Moving {file} to {new_name}") - self.files[number - 1] = file.rename(desired_path / new_name) + if dry_run: + continue + self.files[number - 1] = file.rename(new_name) for file in self.files_extra: if file.suffix.lower() in {".cue", ".log"} or file in { self.epub_file, @@ -688,6 +679,8 @@ class Audiobook: continue logger.trace(f"Moving {file} to {new_file}") + if dry_run: + continue new_file = file.rename(new_file) if file == self.epub_file: self.epub_file = new_file @@ -697,7 +690,7 @@ class Audiobook: if self.bookPath != desired_path: logger.debug(f"Moving {self.bookPath} to {desired_path}") remaining = list(self.bookPath.iterdir()) - if not remaining: + if not remaining and not dry_run: self.bookPath.rmdir() else: logger.warning( @@ -714,7 +707,8 @@ class Audiobook: if self.composer: composer = f" (by {self.composer})" name = self.name + composer - desired_path = self.library_root_path / self.artist / name + artist = re.split(r"[,;]", self.artist)[0].strip().replace("/", "_") + desired_path = self.library_root_path / artist / name if self.series_and_num: series = self.series_and_num number = "" @@ -724,7 +718,7 @@ class Audiobook: series = series[0].rstrip() desired_path = ( self.library_root_path - / self.artist.replace("/", "_") + / artist / series.replace("/", "_") / f"{number}{name}".replace("/", "_") ) @@ -941,6 +935,8 @@ class Audiobook: self.title.append(f"{self.name} - {idx + 1:0{padding}d}") + + class AudiobookLibrary: def __init__( self, @@ -1195,6 +1191,7 @@ class AudiobookLibrary: def fix_composers(self): for book in self.audiobooks: + book.fix_composer_from_narrator() book.fix_composer_from_comment() book.fix_composer_from_path() @@ -1261,6 +1258,7 @@ def main(): m4b_group = parser.add_argument_group("M4B creation") m4b_group.add_argument( + "-cm", "--create-m4b", action="store_true", help="Create M4B files from multi-file audiobooks",