From b18cc9c633ae73ae69f1a16fdf682bd5ac58444c Mon Sep 17 00:00:00 2001 From: pukkandan Date: Tue, 20 Feb 2024 09:03:28 +0530 Subject: [PATCH] Rewrite --- README.md | 4 +- yt_dlp/options.py | 3 +- yt_dlp/postprocessor/embedthumbnail.py | 11 +-- yt_dlp/postprocessor/ffmpeg.py | 112 +++++++++++++------------ yt_dlp/utils/_utils.py | 2 + 5 files changed, 70 insertions(+), 62 deletions(-) diff --git a/README.md b/README.md index 979d10fbe..c621e425a 100644 --- a/README.md +++ b/README.md @@ -1002,9 +1002,7 @@ ## Post-Processing Options: --no-post-overwrites Do not overwrite post-processed files --embed-subs Embed subtitles in downloaded media. Available for video (mp4, webm, mkv) and - audio (m4a, mp3, ogg, flac). Use --convert- - subtitles lrc when embedding subtitles in - audio files + "lrc" in audio (m4a, mp3, ogg, flac) --no-embed-subs Do not embed subtitles (default) --embed-thumbnail Embed thumbnail in the video as cover art --no-embed-thumbnail Do not embed thumbnail (default) diff --git a/yt_dlp/options.py b/yt_dlp/options.py index 3fa52c8ca..ad4e2190c 100644 --- a/yt_dlp/options.py +++ b/yt_dlp/options.py @@ -1584,8 +1584,7 @@ def _alias_callback(option, opt_str, value, parser, opts, nargs): action='store_true', dest='embedsubtitles', default=False, help=( 'Embed subtitles in downloaded media. ' - 'Available for video (mp4, webm, mkv) and audio (m4a, mp3, ogg, flac). ' - 'Use --convert-subtitles lrc when embedding subtitles in audio files')) + 'Available for video (mp4, webm, mkv) and "lrc" in audio (m4a, mp3, ogg, flac)')) postproc.add_option( '--no-embed-subs', action='store_false', dest='embedsubtitles', diff --git a/yt_dlp/postprocessor/embedthumbnail.py b/yt_dlp/postprocessor/embedthumbnail.py index 40f924a8d..4abab40d3 100644 --- a/yt_dlp/postprocessor/embedthumbnail.py +++ b/yt_dlp/postprocessor/embedthumbnail.py @@ -91,8 +91,10 @@ def run(self, info): success = True if info['ext'] == 'mp3': # Method 1: Use mutagen - # Prioritize mutagen over ffmpeg since ffmpeg messes up the lyrics data - if mutagen: + if not mutagen: + self.to_screen('mutagen not was found. Falling back to ffmpeg. Lyrics may be corrupted') + success = False + else: try: self._report_run('mutagen', filename) audio = mutagen.id3.ID3(filename) @@ -101,12 +103,11 @@ def run(self, info): encoding=mutagen.id3.Encoding.UTF8, mime=f'image/{thumbnail_ext}', type=mutagen.id3.PictureType.COVER_FRONT, desc='Cover (front)', data=thumbfile.read()) audio.save() - temp_filename = filename # Mutagen saves to the original file + temp_filename = filename except Exception as err: self.report_warning(f'unable to embed using mutagen; {err}') success = False - else: - success = False + # Method 2: Use ffmpeg if not success: options = [ diff --git a/yt_dlp/postprocessor/ffmpeg.py b/yt_dlp/postprocessor/ffmpeg.py index 9a685c5a5..41dd1df9a 100644 --- a/yt_dlp/postprocessor/ffmpeg.py +++ b/yt_dlp/postprocessor/ffmpeg.py @@ -596,16 +596,16 @@ def __init__(self, downloader=None, already_have_subtitle=False): @PostProcessor._restrict_to(images=False) def run(self, info): - if info['ext'] not in self.SUPPORTED_EXTS: + filename, ext = info['filepath'], info['ext'] + if ext not in self.SUPPORTED_EXTS: self.to_screen(f'Subtitles can only be embedded in {", ".join(self.SUPPORTED_EXTS)} files') return [], info + subtitles = info.get('requested_subtitles') if not subtitles: self.to_screen('There aren\'t any subtitles to embed') return [], info - filename = info['filepath'] - # Disabled temporarily. There needs to be a way to override this # in case of duration actually mismatching in extractor # See: https://github.com/yt-dlp/yt-dlp/issues/1870, https://github.com/yt-dlp/yt-dlp/issues/1385 @@ -616,76 +616,84 @@ def run(self, info): return [], info ''' - ext = info['ext'] - sub_langs, sub_names, sub_filenames = [], [], [] - webm_vtt_warn = False - mp4_ass_warn = False + warnings = set() + def warn_once(msg): + if msg not in warnings: + warnings.add(msg) + self.report_warning(msg) + + subtitles_to_embed = {} for lang, sub_info in subtitles.items(): if not os.path.exists(sub_info.get('filepath', '')): self.report_warning(f'Skipping embedding {lang} subtitle because the file is missing') - continue - sub_ext = sub_info['ext'] - if sub_ext == 'json': - self.report_warning('JSON subtitles cannot be embedded') - elif ext != 'webm' or ext == 'webm' and sub_ext == 'vtt': - sub_langs.append(lang) - sub_names.append(sub_info.get('name')) - sub_filenames.append(sub_info['filepath']) + elif sub_info['ext'] == 'json': + warn_once('JSON subtitles cannot be embedded') + elif ext == 'webm' and sub_info['ext'] != 'vtt': + warn_once('Only WebVTT subtitles can be embedded in webm files') + elif ext in self.SUPPORTS_LYRICS and sub_info['ext'] != 'lrc': + warn_once(f'Only lrc subtitles can be embedded in {ext} files') + elif ext in self.SUPPORTS_LYRICS and not mutagen: + raise PostProcessingError( + f'[{self.PP_NAME}] module mutagen was not found. Please install using `python -m pip install mutagen`') else: - if not webm_vtt_warn and ext == 'webm' and sub_ext != 'vtt': - webm_vtt_warn = True - self.report_warning('Only WebVTT subtitles can be embedded in webm files') - if not mp4_ass_warn and ext == 'mp4' and sub_ext == 'ass': - mp4_ass_warn = True - self.report_warning('ASS subtitles cannot be properly embedded in mp4 files; expect issues') + if ext == 'mp4' and sub_info['ext'] == 'ass': + warn_once('ASS subtitles cannot be properly embedded in mp4 files; expect issues') + subtitles_to_embed[lang] = sub_info - if not sub_langs: + if not subtitles_to_embed: return [], info - input_files = [filename] + sub_filenames + sub_files = [sub['filepath'] for sub in subtitles_to_embed.values()] + files_to_delete = [] if self._already_have_subtitle else sub_files + + if ext in self.SUPPORTS_LYRICS: + self._embed_lyrics(subtitles_to_embed, info['filepath'], ext) + return files_to_delete, info opts = [ - *self.stream_copy_opts(ext=info['ext']), + *self.stream_copy_opts(ext=ext), # Don't copy the existing subtitles, we may be running the # postprocessor a second time '-map', '-0:s', ] - for i, (lang, name) in enumerate(zip(sub_langs, sub_names)): - opts.extend(['-map', '%d:0' % (i + 1)]) - lang_code = ISO639Utils.short2long(lang) or lang - opts.extend(['-metadata:s:s:%d' % i, 'language=%s' % lang_code]) - if name: - opts.extend(['-metadata:s:s:%d' % i, 'handler_name=%s' % name, - '-metadata:s:s:%d' % i, 'title=%s' % name]) + for i, (lang, sub) in enumerate(subtitles_to_embed.items()): + lang = ISO639Utils.short2long(lang) or lang + opts.extend(['-map', f'{i + 1}:0', f'-metadata:s:s:{i}', f'language={lang}']) + if name := sub['name']: + opts.extend([f'-metadata:s:s:{i}', f'handler_name={name}', + f'-metadata:s:s:{i}', f'title={name}']) temp_filename = prepend_extension(filename, 'temp') - self.to_screen('Embedding subtitles in "%s"' % filename) - if info['ext'] in self.SUPPORTS_LYRICS: - self.embed_lyrics(info['filepath'], sub_dict=info['requested_subtitles']) - else: - self.run_ffmpeg_multiple_files(input_files, temp_filename, opts) - os.replace(temp_filename, filename) + self.to_screen(f'Embedding subtitles in "{filename}"') + self.run_ffmpeg_multiple_files([filename, *sub_files], temp_filename, opts) + os.replace(temp_filename, filename) - files_to_delete = [] if self._already_have_subtitle else sub_filenames return files_to_delete, info - def embed_lyrics(self, audio_file, sub_dict): - if not all(sub['ext'] == 'lrc' for sub in sub_dict.values()): - raise PostProcessingError('LRC subtitles required. Use "--convert-subs lrc" to convert') - if len(sub_dict) > 1: - self.report_warning('More than one subtitle file found. Your media player will likely be unable to display all of them.') + def _embed_lyrics(self, subtitles, filename, ext): + assert mutagen and ext in self.SUPPORTS_LYRICS and all(sub['ext'] == 'lrc' for sub in subtitles.values()) + self.to_screen(f'Embedding lyrics in "{filename}"') + if len(subtitles) > 1: + self.report_warning( + f'Your media player may be unable to display multiple subtitles in {ext}') - lyrics_list = [sub['data'] for sub in sub_dict.values()] - if audio_file.endswith('.mp3'): - audio = mutagen.id3.ID3(audio_file) - for lyrics in lyrics_list: - audio.add(mutagen.id3.USLT(encoding=mutagen.id3.Encoding.UTF8, lang='und', text=lyrics)) - audio.save() + for lang, sub in subtitles.items(): + if not sub.get('data'): + with open(sub['filepath'], encoding='utf-8') as f: + sub['data'] = f.read() + + if ext == 'mp3': + metadata = mutagen.id3.ID3(filename) + for lang, sub in subtitles.items(): + metadata.add(mutagen.id3.USLT( + encoding=mutagen.id3.Encoding.UTF8, + lang=ISO639Utils.short2long(lang) or 'und', + text=sub['data'])) else: - metadata = mutagen.File(audio_file) - metadata['©lyr' if audio_file.endswith('.m4a') else 'lyrics'] = lyrics_list - metadata.save() + metadata = mutagen.File(filename) + metadata['©lyr' if ext == 'm4a' else 'lyrics'] = [sub['data'] for sub in subtitles.values()] + metadata.save() class FFmpegMetadataPP(FFmpegPostProcessor): diff --git a/yt_dlp/utils/_utils.py b/yt_dlp/utils/_utils.py index 361617c02..54cac7e6b 100644 --- a/yt_dlp/utils/_utils.py +++ b/yt_dlp/utils/_utils.py @@ -3733,6 +3733,8 @@ class ISO639Utils: @classmethod def short2long(cls, code): """Convert language code from ISO 639-1 to ISO 639-2/T""" + if code in cls._lang_map.values(): + return code return cls._lang_map.get(code[:2]) @classmethod