[utils, cleanup] Consolidate known media extensions

README.md

@@ -916,7 +916,7 @@ ## Post-Processing Options:
                                     (requires ffmpeg and ffprobe)
     --audio-format FORMAT           Format to convert the audio to when -x is
                                     used. (currently supported: best (default),
-                                    mp3, aac, m4a, opus, vorbis, flac, alac,
+                                    aac, alac, flac, m4a, mp3, opus, vorbis,
                                     wav). You can specify multiple rules using
                                     similar syntax as --remux-video
     --audio-quality QUALITY         Specify ffmpeg audio quality to use when
@@ -924,9 +924,9 @@ ## Post-Processing Options:
                                     between 0 (best) and 10 (worst) for VBR or a
                                     specific bitrate like 128K (default 5)
     --remux-video FORMAT            Remux the video into another container if
-                                    necessary (currently supported: mp4, mkv,
+                                    necessary (currently supported: avi, flv,
-                                    flv, webm, mov, avi, mka, ogg, mp3, aac,
+                                    mkv, mov, mp4, webm, aac, aiff, alac, flac,
-                                    m4a, opus, vorbis, flac, alac, wav). If
+                                    m4a, mka, mp3, ogg, opus, vorbis, wav). If
                                     target container does not support the
                                     video/audio codec, remuxing will fail. You
                                     can specify multiple rules; Eg.
@@ -1025,7 +1025,7 @@ ## Post-Processing Options:
                                     be used multiple times
     --no-exec                       Remove any previously defined --exec
     --convert-subs FORMAT           Convert the subtitles to another format
-                                    (currently supported: srt, vtt, ass, lrc)
+                                    (currently supported: ass, lrc, srt, vtt)
                                     (Alias: --convert-subtitles)
     --convert-thumbnails FORMAT     Convert the thumbnails to another format
                                     (currently supported: jpg, png, webp). You

yt_dlp/YoutubeDL.py

@@ -52,6 +52,7 @@
     DEFAULT_OUTTMPL,
     IDENTITY,
     LINK_TEMPLATES,
+    MEDIA_EXTENSIONS,
     NO_DEFAULT,
     NUMBER_RE,
     OUTTMPL_TYPES,
@@ -543,9 +544,9 @@ class YoutubeDL:
         'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
     }
     _format_selection_exts = {
-        'audio': {'m4a', 'mp3', 'ogg', 'aac'},
+        'audio': set(MEDIA_EXTENSIONS.common_audio),
-        'video': {'mp4', 'flv', 'webm', '3gp'},
+        'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
-        'storyboards': {'mhtml'},
+        'storyboards': set(MEDIA_EXTENSIONS.storyboards),
     }
 
     def __init__(self, params=None, auto_init=True):

yt_dlp/__init__.py

@@ -19,8 +19,8 @@
 from .extractor.common import InfoExtractor
 from .options import parseOpts
 from .postprocessor import (
-    FFmpegPostProcessor,
     FFmpegExtractAudioPP,
+    FFmpegPostProcessor,
     FFmpegSubtitlesConvertorPP,
     FFmpegThumbnailsConvertorPP,
     FFmpegVideoConvertorPP,

yt_dlp/options.py

@@ -972,7 +972,7 @@ def _alias_callback(option, opt_str, value, parser, opts, nargs):
         }, help=(
             'Name or path of the external downloader to use (optionally) prefixed by '
             'the protocols (http, ftp, m3u8, dash, rstp, rtmp, mms) to use it for. '
-            f'Currently supports native, {", ".join(list_external_downloaders())}. '
+            f'Currently supports native, {", ".join(sorted(list_external_downloaders()))}. '
             'You can use this option multiple times to set different downloaders for different protocols. '
             'For example, --downloader aria2c --downloader "dash,m3u8:native" will use '
             'aria2c for http/ftp downloads, and the native downloader for dash/m3u8 downloads '
@@ -1469,7 +1469,7 @@ def _alias_callback(option, opt_str, value, parser, opts, nargs):
         '--audio-format', metavar='FORMAT', dest='audioformat', default='best',
         help=(
             'Format to convert the audio to when -x is used. '
-            f'(currently supported: best (default), {", ".join(FFmpegExtractAudioPP.SUPPORTED_EXTS)}). '
+            f'(currently supported: best (default), {", ".join(sorted(FFmpegExtractAudioPP.SUPPORTED_EXTS))}). '
             'You can specify multiple rules using similar syntax as --remux-video'))
     postproc.add_option(
         '--audio-quality', metavar='QUALITY',
@@ -1652,13 +1652,13 @@ def _alias_callback(option, opt_str, value, parser, opts, nargs):
         metavar='FORMAT', dest='convertsubtitles', default=None,
         help=(
             'Convert the subtitles to another format (currently supported: %s) '
-            '(Alias: --convert-subtitles)' % ', '.join(FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)))
+            '(Alias: --convert-subtitles)' % ', '.join(sorted(FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS))))
     postproc.add_option(
         '--convert-thumbnails',
         metavar='FORMAT', dest='convertthumbnails', default=None,
         help=(
             'Convert the thumbnails to another format '
-            f'(currently supported: {", ".join(FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)}). '
+            f'(currently supported: {", ".join(sorted(FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS))}). '
             'You can specify multiple rules using similar syntax as --remux-video'))
     postproc.add_option(
         '--split-chapters', '--split-tracks',

yt_dlp/postprocessor/ffmpeg.py

@@ -10,6 +10,7 @@
 from .common import PostProcessor
 from ..compat import functools, imghdr
 from ..utils import (
+    MEDIA_EXTENSIONS,
     ISO639Utils,
     Popen,
     PostProcessingError,
@@ -424,7 +425,7 @@ def _concat_spec(cls, in_files, concat_opts=None):
 
 
 class FFmpegExtractAudioPP(FFmpegPostProcessor):
-    COMMON_AUDIO_EXTS = ('wav', 'flac', 'm4a', 'aiff', 'mp3', 'ogg', 'mka', 'opus', 'wma')
+    COMMON_AUDIO_EXTS = MEDIA_EXTENSIONS.common_audio + ('wma', )
     SUPPORTED_EXTS = tuple(ACODECS.keys())
     FORMAT_RE = create_mapping_re(('best', *SUPPORTED_EXTS))
 
@@ -531,7 +532,7 @@ def run(self, information):
 
 
 class FFmpegVideoConvertorPP(FFmpegPostProcessor):
-    SUPPORTED_EXTS = ('mp4', 'mkv', 'flv', 'webm', 'mov', 'avi', 'mka', 'ogg', *FFmpegExtractAudioPP.SUPPORTED_EXTS)
+    SUPPORTED_EXTS = (*MEDIA_EXTENSIONS.common_video, *sorted(MEDIA_EXTENSIONS.common_audio + ('aac', 'vorbis')))
     FORMAT_RE = create_mapping_re(SUPPORTED_EXTS)
     _ACTION = 'converting'
 
@@ -924,7 +925,7 @@ class FFmpegFixupDuplicateMoovPP(FFmpegCopyStreamPP):
 
 
 class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
-    SUPPORTED_EXTS = ('srt', 'vtt', 'ass', 'lrc')
+    SUPPORTED_EXTS = MEDIA_EXTENSIONS.subtitles
 
     def __init__(self, downloader=None, format=None):
         super().__init__(downloader)
@@ -1046,7 +1047,7 @@ def run(self, info):
 
 
 class FFmpegThumbnailsConvertorPP(FFmpegPostProcessor):
-    SUPPORTED_EXTS = ('jpg', 'png', 'webp')
+    SUPPORTED_EXTS = MEDIA_EXTENSIONS.thumbnails
     FORMAT_RE = create_mapping_re(SUPPORTED_EXTS)
 
     def __init__(self, downloader=None, format=None):

yt_dlp/utils.py

@@ -150,22 +150,6 @@ def random_user_agent():
         'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'],
 }
 
-KNOWN_EXTENSIONS = (
-    'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac',
-    'flv', 'f4v', 'f4a', 'f4b',
-    'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus',
-    'mkv', 'mka', 'mk3d',
-    'avi', 'divx',
-    'mov',
-    'asf', 'wmv', 'wma',
-    '3gp', '3g2',
-    'mp3',
-    'mpg',
-    'flac',
-    'ape',
-    'wav',
-    'f4f', 'f4m', 'm3u8', 'smil')
-
 # needed for sanitizing filenames in restricted mode
 ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
                         itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
@@ -5647,6 +5631,22 @@ def items_(self):
         return self.__dict__.items()
 
 
+MEDIA_EXTENSIONS = Namespace(
+    common_video=('avi', 'flv', 'mkv', 'mov', 'mp4', 'webm'),
+    video=('3g2', '3gp', 'f4v', 'mk3d', 'divx', 'mpg', 'ogv', 'm4v', 'wmv'),
+    common_audio=('aiff', 'alac', 'flac', 'm4a', 'mka', 'mp3', 'ogg', 'opus', 'wav'),
+    audio=('aac', 'ape', 'asf', 'f4a', 'f4b', 'm4b', 'm4p', 'm4r', 'oga', 'ogx', 'spx', 'vorbis', 'wma'),
+    thumbnails=('jpg', 'png', 'webp'),
+    storyboards=('mhtml', ),
+    subtitles=('srt', 'vtt', 'ass', 'lrc'),
+    manifests=('f4f', 'f4m', 'm3u8', 'smil', 'mpd'),
+)
+MEDIA_EXTENSIONS.video += MEDIA_EXTENSIONS.common_video
+MEDIA_EXTENSIONS.audio += MEDIA_EXTENSIONS.common_audio
+
+KNOWN_EXTENSIONS = (*MEDIA_EXTENSIONS.video, *MEDIA_EXTENSIONS.audio, *MEDIA_EXTENSIONS.manifests)
+
+
 # Deprecated
 has_certifi = bool(certifi)
 has_websockets = bool(websockets)