Merge 74bbdc079b1dae209faebb8790282d55a6c88e89 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

* thx seproDev, yt-dlp/yt-dlp/pull/12694/commits/ae69e3c
* also add documenting comment in `InfoExtractor`

youtube_dl/__init__.py

@@ -282,6 +282,7 @@ def _real_main(argv=None):
         postprocessors.append({
             'key': 'FFmpegSubtitlesConvertor',
             'format': opts.convertsubtitles,
+            'ttml_convert_style': opts.ttmlconvertstyle,
         })
     if opts.embedsubtitles:
         postprocessors.append({

youtube_dl/extractor/bokecc.py

@@ -32,7 +32,7 @@ class BokeCCBaseIE(InfoExtractor):
 
 
 class BokeCCIE(BokeCCBaseIE):
-    _IE_DESC = 'CC视频'
+    IE_DESC = 'CC视频'
     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
 
     _TESTS = [{

youtube_dl/extractor/cloudy.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec'
+    IE_DESC = 'cloudy.ec'
     _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',

youtube_dl/extractor/common.py

@@ -422,6 +422,8 @@ class InfoExtractor(object):
     _GEO_COUNTRIES = None
     _GEO_IP_BLOCKS = None
     _WORKING = True
+    # supply this in public subclasses: used in supported sites list, etc
+    # IE_DESC = 'short description of IE'
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""

youtube_dl/extractor/itv.py

@@ -35,15 +35,6 @@ from ..utils import (
 
 class ITVBaseIE(InfoExtractor):
 
-    def _search_nextjs_data(self, webpage, video_id, **kw):
-        transform_source = kw.pop('transform_source', None)
-        fatal = kw.pop('fatal', True)
-        return self._parse_json(
-            self._search_regex(
-                r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
-                webpage, 'next.js data', group='js', fatal=fatal, **kw),
-            video_id, transform_source=transform_source, fatal=fatal)
-
     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
         if errnote is False:
             return False
@@ -109,7 +100,9 @@ class ITVBaseIE(InfoExtractor):
 
 class ITVIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
-    _IE_DESC = 'ITVX'
+    IE_DESC = 'ITVX'
+    _WORKING = False
+
     _TESTS = [{
         'note': 'Hub URLs redirect to ITVX',
         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
@@ -270,7 +263,7 @@ class ITVIE(ITVBaseIE):
                 'ext': determine_ext(href, 'vtt'),
             })
 
-        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
+        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default={})
         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
         info = self._og_extract(webpage, require_title=not title)
@@ -323,7 +316,7 @@ class ITVIE(ITVBaseIE):
 
 class ITVBTCCIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
-    _IE_DESC = 'ITV articles: News, British Touring Car Championship'
+    IE_DESC = 'ITV articles: News, British Touring Car Championship'
     _TESTS = [{
         'note': 'British Touring Car Championship',
         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',

youtube_dl/extractor/senateisvp.py

@@ -47,7 +47,7 @@ class SenateISVPIE(InfoExtractor):
         ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
         ['arch', '', 'http://ussenate-f.akamaihd.net/']
     ]
-    _IE_NAME = 'senate.gov'
+    IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
         'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png',

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError:

youtube_dl/options.py

@@ -870,6 +870,10 @@ def parseOpts(overrideArguments=None):
         '--convert-subs', '--convert-subtitles',
         metavar='FORMAT', dest='convertsubtitles', default=None,
         help='Convert the subtitles to other format (currently supported: srt|ass|vtt|lrc)')
+    postproc.add_option(
+        '--ttml-convert-no-style',
+        action='store_false', dest='ttmlconvertstyle', default=True,
+        help='Discard all style information when converting subtitles from ttml/dfxp')
 
     parser.add_option_group(general)
     parser.add_option_group(network)

youtube_dl/postprocessor/ffmpeg.py

@@ -591,9 +591,10 @@ class FFmpegFixupM3u8PP(FFmpegPostProcessor):
 
 
 class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
-    def __init__(self, downloader=None, format=None):
+    def __init__(self, downloader=None, format=None, ttml_convert_style=True):
         super(FFmpegSubtitlesConvertorPP, self).__init__(downloader)
         self.format = format
+        self.ttml_convert_style = ttml_convert_style
 
     def run(self, info):
         subs = info.get('requested_subtitles')
@@ -626,7 +627,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
                 srt_file = subtitles_filename(filename, lang, 'srt', info.get('ext'))
 
                 with open(dfxp_file, 'rb') as f:
-                    srt_data = dfxp2srt(f.read())
+                    srt_data = dfxp2srt(f.read(), self.ttml_convert_style)
 
                 with open(srt_file, 'w', encoding='utf-8') as f:
                     f.write(srt_data)

youtube_dl/utils.py

@@ -4902,7 +4902,7 @@ def srt_subtitles_timecode(seconds):
     return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000)
 
 
-def dfxp2srt(dfxp_data):
+def dfxp2srt(dfxp_data, convert_style=True):
     '''
     @param dfxp_data A bytes-like object containing DFXP data
     @returns A unicode object containing converted SRT data
@@ -4944,7 +4944,7 @@ def dfxp2srt(dfxp_data):
         def start(self, tag, attrib):
             if tag in (_x('ttml:br'), 'br'):
                 self._out += '\n'
-            else:
+            elif convert_style:
                 unclosed_elements = []
                 style = {}
                 element_style_id = attrib.get('style')
@@ -4987,7 +4987,7 @@ def dfxp2srt(dfxp_data):
                 self._unclosed_elements.append(unclosed_elements)
 
         def end(self, tag):
-            if tag not in (_x('ttml:br'), 'br'):
+            if convert_style and tag not in (_x('ttml:br'), 'br'):
                 unclosed_elements = self._unclosed_elements.pop()
                 for element in reversed(unclosed_elements):
                     self._out += '</%s>' % element