Merge 7a3c8386578511a4d20ad5822cd86178f0819872 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

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

youtube_dl/extractor/bilibili.py

@@ -38,7 +38,21 @@ class BiliBiliIE(InfoExtractor):
                     '''
 
     _TESTS = [{
-        'url': 'http://www.bilibili.tv/video/av1074402/',
+        'url': 'https://www.bilibili.com/video/BV1N7411A7WU/',
+        'info_dict': {
+            'id': '1N7411A7WU',
+            'ext': 'flv',
+            'title': '[经验分享]观看B站视频时如何开启CC字幕？(B站小白必看)',
+            'description': 'md5:b5b03414a659c29c7f5bce58c725b8ee',
+            'duration': 119.267,
+            'timestamp': 1582321654,
+            'upload_date': '20200221',
+            'uploader': 'Lynda教程和字幕',
+            'uploader_id': '389168097',
+            'subtitles': {'zh-CN': [{'url': 'http:\\u002F\\u002Fi0.hdslb.com\\u002Fbfs\\u002Fsubtitle\\u002F3c648ce057eea5b4e619c5dd9f0f23fbcad728f5.json', 'data': 'md5:ece63ef7a3a45fe2f7183f95f96d3181', 'ext': 'srt'}]}
+        },
+    }, {
+        'url': 'http://www.bilibili.com/video/av1074402/',
         'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
         'info_dict': {
             'id': '1074402',
@@ -112,6 +126,12 @@ class BiliBiliIE(InfoExtractor):
     _APP_KEY = 'iVGUTjsxvpLeuDCf'
     _BILIBILI_KEY = 'aHRmhWMLkdeMuILqORnYZocwMBpMEOdt'
 
+    def convert_timestamp_format(self, cc_timestamp):
+        ss, mm = "{:.3f}".format(float(cc_timestamp)).split(".")
+        m, s = divmod(int(ss), 60)
+        h, m = divmod(m, 60)
+        return "%02d:%02d:%02d,%s" % (h, m, s, mm)
+
     def _report_error(self, result):
         if 'message' in result:
             raise ExtractorError('%s said: %s' % (self.IE_NAME, result['message']), expected=True)
@@ -155,6 +175,21 @@ class BiliBiliIE(InfoExtractor):
                 self._report_error(js)
             cid = js['result']['cid']
 
+        subtitles = {}
+        for sub in re.findall(r'"lan":\s*"([^"]+)"[^}]+"subtitle_url":\s*"([^"]+)"', webpage, re.DOTALL):
+            subtitles[sub[0]] = [{'url': sub[1]}]
+            j = self._download_json(sub[1].replace("\\u002F", "/"), video_id, note='Downloading raw subtitle: %s' % (sub[0]))
+            srt = []
+            for idx, line in enumerate(j["body"]):
+                from_timestamp = self.convert_timestamp_format(line["from"])
+                to_timestamp = self.convert_timestamp_format(line["to"])
+                srt_tmp = (
+                    """%d\n%s --> %s\n%s\n""" % (idx, from_timestamp, to_timestamp, line["content"])
+                )
+                srt.append(srt_tmp)
+            subtitles[sub[0]][0]["data"] = "\n".join(srt)
+            subtitles[sub[0]][0]["ext"] = "srt"
+
         headers = {
             'Accept': 'application/json',
             'Referer': url
@@ -204,6 +239,7 @@ class BiliBiliIE(InfoExtractor):
                     'id': '%s_part%s' % (video_id, idx),
                     'duration': float_or_none(durl.get('length'), 1000),
                     'formats': formats,
+                    'subtitles': subtitles
                 })
             break
 

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: