From 8c2541fb34c48f23f096854d2e15b110909f04f4 Mon Sep 17 00:00:00 2001 From: Lucas Media Date: Mon, 26 Oct 2026 13:00:58 +0100 Subject: [PATCH] fix(ffmpeg handling): add language tag to is russian identification --- media_editing/ffmpeg_handling.py | 150 ++++++++++++++++++------------- serverlibrary/core.py | 2 + 2 files changed, 89 insertions(+), 63 deletions(-) diff --git a/media_editing/ffmpeg_handling.py b/media_editing/ffmpeg_handling.py index 7f81c0b..212e56b 100755 --- a/media_editing/ffmpeg_handling.py +++ b/media_editing/ffmpeg_handling.py @@ -136,8 +136,11 @@ RUSSIAN_RE = re.compile( ) -def is_russian_track(description: str) -> bool: - return bool(RUSSIAN_RE.search(description)) +def is_russian_track(track: dict) -> bool: + descr = bool(RUSSIAN_RE.search(track["description"])) + language = bool(RUSSIAN_RE.search(track["language"])) + lang = bool(RUSSIAN_RE.search(track["lang"])) + return descr or lang or language class ArgsBuffer: @@ -147,7 +150,7 @@ class ArgsBuffer: loud: bool | None = False key: str | None = None value: str | None = None - audio_index: int | None = None + audio_index: list | None = None dry_run: bool | None = None audio_codec: str | None = None video_codec: str | None = None @@ -168,27 +171,28 @@ def translate_audio_tracks(tracks: list[dict]) -> list[dict]: translated = [] for track in tracks: - description = track.get("description", "") + new_track = track.copy() - for src, dst in AUDIO_TRANSLATIONS.items(): - description = re.sub( - rf"\b{re.escape(src)}\b", - dst, - description, - flags=re.IGNORECASE, - ) + for key, value in new_track.items(): + if not isinstance(value, str): + continue - if CYRILLIC_RE.search(description): - raise ValueError( - f"Contains unidentified Cyrillic characters: {description}" - ) + for src, dst in AUDIO_TRANSLATIONS.items(): + value = re.sub( + rf"\b{re.escape(src)}\b", + dst, + value, + flags=re.IGNORECASE, + ) - translated.append( - { - **track, - "description": description, - } - ) + if CYRILLIC_RE.search(value): + raise ValueError( + f"Contains unidentified Cyrillic characters in {key}: {value}" + ) + + new_track[key] = value + + translated.append(new_track) return translated @@ -210,17 +214,17 @@ def ensure(value, prompt): return value -def move_override_file(files_move_from, files_move_to, args): +def move_overwrite_file(files_move_from, files_move_to, args): backup_files = args.backup silent = not args.loud for file_to, file_from in zip(files_move_to, files_move_from): if not file_to.is_file(): - logger.error(f"Override Target File does not exist: {file_to}") + logger.error(f"overwrite Target File does not exist: {file_to}") return False if not file_from.is_file(): - logger.trace(f"Override File does not exist: {file_from}") + logger.trace(f"overwrite File does not exist: {file_from}") continue backup = file_to.parent / f"{file_to.stem}_backup{file_from.suffix}" @@ -229,7 +233,7 @@ def move_override_file(files_move_from, files_move_to, args): logger.trace(f"Move original to backup: {backup}") safe_rename(file_to, backup, silent=silent) - # move override -> original name + # move overwrite -> original name retstr, ret = safe_rename(file_from, file_to, silent=silent, overwrite=True) logger.trace(retstr) if not ret: @@ -584,13 +588,14 @@ def cmd_remove_audio_track(args): """ probe = ffmpeg.probe(str(args.input)) - - remove_idx = select_audio_stream(args.input, args.audio_index) + remove_indices = [] + for audio_index in args.audio_index: + remove_indices.append(select_audio_stream(args.input, audio_index)) remaining_audio = [ s for s in probe["streams"] - if s["codec_type"] == "audio" and s["index"] != remove_idx + if s["codec_type"] == "audio" and s["index"] not in remove_indices ] if not remaining_audio: logger.error("No audiotracks remaining") @@ -598,7 +603,7 @@ def cmd_remove_audio_track(args): remaining_audio = translate_audio_tracks(remaining_audio) removed_was_default = any( - s["index"] == remove_idx and s.get("disposition", {}).get("default", 0) == 1 + s["index"] in remove_indices and s.get("disposition", {}).get("default", 0) == 1 for s in probe["streams"] ) @@ -610,7 +615,7 @@ def cmd_remove_audio_track(args): ] for stream in probe["streams"]: - if stream["index"] != remove_idx: + if stream["index"] not in remove_indices: cmd.extend(["-map", f"0:{stream['index']}"]) cmd.extend( @@ -671,19 +676,24 @@ def cmd_extract_audio_track(args): output_audio = Path(args.output) output_audio = str(output_audio.with_name(f"{output_audio.stem}.mp3")) - remove_idx = select_audio_stream(args.input, args.audio_index) + outputs = [] + for index in args.audio_index: + remove_idx = select_audio_stream(args.input, index) - extracted_audio = None + extracted_audio = None - for s in probe["streams"]: - idx = s["index"] - if idx == remove_idx: - extracted_audio = inp[str(idx)] + for s in probe["streams"]: + idx = s["index"] + if idx == remove_idx: + extracted_audio = inp[str(idx)] - out = ffmpeg.output( - extracted_audio, output_audio, acodec="libmp3lame", audio_bitrate="192k" - ) - return True, out + outputs.append( + ffmpeg.output( + extracted_audio, output_audio, acodec="libmp3lame", audio_bitrate="192k" + ) + ) + + return True, outputs def cmd_downsize(args): @@ -736,8 +746,14 @@ def cmd_filter_audio(args): 5. alimiter: Hard limit at 0.95 to prevent digital clipping (distortion). 6. dynaudnorm: Smooths out loudness levels over time. """ + if len(args.audio_index) > 1: + logger.warning( + f"Filter Audio only filters the first element of list audio_index: {args.audio_index}" + ) - idx_filt = select_audio_stream(args.input, args.audio_index) + index = args.audio_index[0] + + idx_filt = select_audio_stream(args.input, index) inp = ffmpeg.input(args.input) probe = ffmpeg.probe(str(args.input)) @@ -888,7 +904,13 @@ def cmd_set_primary_audio(args): inp = ffmpeg.input(args.input) - primary_audio = select_audio_stream(args.input, args.audio_index) + if len(args.audio_index) > 1: + logger.warning( + f"Set primary can only set the first element of list audio_index: {args.audio_index}" + ) + index = args.audio_index[0] + + primary_audio = select_audio_stream(args.input, index) probe = ffmpeg.probe(str(args.input)) audio_streams = [ @@ -1111,17 +1133,17 @@ def cmd_remove_russian_from_db(args_in): logger.warning("%s", e) return False, "nooutputneeded" - logger.trace(f"File: {file.name}: Audio tracks: {audio_tracks}") + logger.trace(f"Remove Russian from File: {file.name}") + logger.trace(f"With Audio tracks: {audio_tracks}") remove_tracks = [] for track in audio_tracks: - if is_russian_track(track["description"]): + if is_russian_track(track): remove_tracks.append(track["number"]) logger.debug(f"Removing russian track {track['number']} from {file}") - for number in remove_tracks: - args = args_in - args.audio_index = number - cmd_remove_audio_track(args) + args = args_in + args.audio_index = remove_tracks if remove_tracks: + cmd_remove_audio_track(args) return True, "nooutputneeded" else: return False, "nooutputneeded" @@ -1428,7 +1450,7 @@ def cmd_test(args): cmd = commands["remove-audio"] if cmd["name"] in args.choices: args_test = ArgsBuffer() - args_test.audio_index = 1 + args_test.audio_index = [1] for test_file in test_files: video_streams, audio_streams, subtitle_streams, cover_image = get_streams( @@ -1454,7 +1476,7 @@ def cmd_test(args): cmd = commands["extract-audio"] if cmd["name"] in args.choices: args_test = ArgsBuffer() - args_test.audio_index = 1 + args_test.audio_index = [1] for test_file in test_files: args_test.input = test_file @@ -1587,7 +1609,7 @@ def cmd_test(args): Path(test_dir) / f"out_{str(cmd['name']).replace('-', '_')}_{Path(test_file).stem}{Path(test_file).suffix}" ) - args_test.audio_index = 1 + args_test.audio_index = [1] test_dict = [ { "video_count": len(video_streams), @@ -1640,7 +1662,7 @@ def cmd_test(args): Path(test_dir) / f"out_{str(cmd['name']).replace('-', '_')}_{Path(test_file).stem}{Path(test_file).suffix}" ) - args_test.audio_index = 2 + args_test.audio_index = [2] test_dict = [ { "video_count": len(video_streams), @@ -1771,6 +1793,7 @@ def add_audio_index(parser): "-a", "--audio-index", type=int, + nargs="+", help="Audio track index", ) @@ -1789,7 +1812,7 @@ def main(): base_p.add_argument("input", nargs="?", type=Path, help="Input file or directory") base_p.add_argument("-o", "--output", type=Path, help="Output file") - base_p.add_argument("--override", action="store_true", help="Overwrite input") + base_p.add_argument("--overwrite", action="store_true", help="Overwrite input") base_p.add_argument("--backup", action="store_true", help="Backup original") base_p.add_argument("--loud", action="store_true", help="Show ffmpeg output") base_p.add_argument( @@ -2103,9 +2126,9 @@ def main(): parser.error(f"Input file {args.input} does not exist!") if commands[args.command]["output"] and not getattr(args, "output", None): input_path = Path(args.input) - if args.override: + if args.overwrite: args.output = input_path.with_name( - f"{input_path.stem}_override{input_path.suffix}" + f"{input_path.stem}_overwrite{input_path.suffix}" ) else: args.output = ensure( @@ -2158,9 +2181,10 @@ def main(): "extract-audio", "set-primary-audio", ]: - args.audio_index = ensure(getattr(args, "audio_index", None), "Audio index") + index = str(ensure(getattr(args, "audio_index", None), "Audio index")) + args.audio_index = [int(x) for x in re.split(r"[,\s]+", index.strip()) if x] if args.audio_index is None: - args.audio_index = 1 + args.audio_index = [1] if args.command == "chapters": args.chapter_file = ensure( @@ -2212,8 +2236,8 @@ def main(): continue logger.trace(f"Processing File: {file}") args.input = Path(file) - if args.override: - out_path = file.with_name(f"{file.stem}_override{file.suffix}") + if args.overwrite: + out_path = file.with_name(f"{file.stem}_overwrite{file.suffix}") else: out_path = file.with_name( f"{file.stem}_{args.command}{file.suffix}" @@ -2223,13 +2247,13 @@ def main(): return_value, output_str = args.func(args) if output_str != "nooutputneeded": outputs.append(output_str) - if args.override and return_value: + if args.overwrite and return_value: files_move_from.append(out_path) files_move_to.append(file) counter += 1 # ---- SINGLE FILE ---- elif input_path.is_file(): - return_value,output_str = args.func(args) + return_value, output_str = args.func(args) if output_str != "nooutputneeded": outputs.append(output_str) else: @@ -2245,10 +2269,10 @@ def main(): run_ffmpeg( ffmpeg.merge_outputs(*outputs), overwrite=True, loud=args.loud ) - if args.override and not move_override_file( + if args.overwrite and not move_overwrite_file( files_move_from, files_move_to, args ): - logger.error("move_override failed") + logger.error("move_overwrite failed") sys.exit(1) except ffmpeg.Error as e: logger.error("FFmpeg failed.\n") diff --git a/serverlibrary/core.py b/serverlibrary/core.py index e34a5c6..7e426f6 100755 --- a/serverlibrary/core.py +++ b/serverlibrary/core.py @@ -223,6 +223,8 @@ def get_audio_tracks( or stream.get("tags", {}).get("language") or f"Track {stream['index']}" ), + "lang": stream.get("lang", ""), + "language": stream.get("tags", {}).get("language", ""), "primary": stream.get("codec_type") == "audio" and stream.get("disposition", {}).get("default") == 1, }