Merge 96a0ad4778da7f30ed5be627f2c10df6d0af3ca8 into 4e714f9df1ed2cccd51df60d45ff5504abe827b7

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

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/extractors.py

@@ -1078,6 +1078,10 @@ from .rutube import (
     RutubePersonIE,
     RutubePlaylistIE,
 )
+from .megatvcom import (
+    MegaTVComIE,
+    MegaTVComEmbedIE,
+)
 from .rutv import RUTVIE
 from .ruutu import RuutuIE
 from .ruv import RuvIE

youtube_dl/extractor/generic.py

@@ -102,6 +102,7 @@ from .ustream import UstreamIE
 from .arte import ArteTVEmbedIE
 from .videopress import VideoPressIE
 from .rutube import RutubeIE
+from .megatvcom import MegaTVComEmbedIE
 from .limelight import LimelightBaseIE
 from .anvato import AnvatoIE
 from .washingtonpost import WashingtonPostIE
@@ -3400,6 +3401,12 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 rutube_urls, video_id, video_title, ie=RutubeIE.ie_key())
 
+        # Look for megatv.com embeds
+        megatvcom_urls = list(MegaTVComEmbedIE._extract_urls(webpage, url))
+        if megatvcom_urls:
+            return self.playlist_from_matches(
+                megatvcom_urls, video_id, video_title, ie=MegaTVComEmbedIE.ie_key())
+
         # Look for WashingtonPost embeds
         wapo_urls = WashingtonPostIE._extract_urls(webpage)
         if wapo_urls:

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/megatvcom.py

@@ -0,0 +1,203 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import hashlib
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_str,
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+)
+from ..utils import (
+    HEADRequest,
+    ExtractorError,
+    determine_ext,
+    get_element_by_class,
+    unified_timestamp,
+    extract_attributes,
+    clean_html,
+    unescapeHTML,
+)
+
+
+class MegaTVComBaseIE(InfoExtractor):
+    _PLAYER_DIV_ID = 'player_div_id'
+
+    def _extract_player_attrs(self, webpage):
+        PLAYER_DIV_RE = r'''(?x)
+        <div(?:
+            id=(?P<_q1>["'])(?P<%(pdi)s>%(pdi)s)(?P=_q1)|
+            [^>]*?
+        )+>
+        ''' % {'pdi': self._PLAYER_DIV_ID}
+        for mobj in re.finditer(PLAYER_DIV_RE, webpage):
+            if mobj.group(self._PLAYER_DIV_ID):
+                player_el = mobj.group(0)
+                break
+        else:
+            raise ExtractorError('no <div id="%s"> element found in webpage' %
+                                 self._PLAYER_DIV_ID)
+        return {
+            re.sub(r'^data-(?:kwik_)?', '', k): v
+            for k, v in extract_attributes(player_el).items()
+            if k not in ('id',)
+        }
+
+
+class MegaTVComIE(MegaTVComBaseIE):
+    IE_NAME = 'megatvcom'
+    IE_DESC = 'megatv.com videos'
+    _VALID_URL = r'https?://(?:www\.)?megatv\.com/(?:(?!\d{4})[^/]+/(?P<id>\d+)/[^/]+|\d{4}/\d{2}/\d{2}/.+)'
+
+    _TESTS = [{
+        'url': 'https://www.megatv.com/2021/10/23/egkainia-gia-ti-nea-skini-omega-tou-dimotikou-theatrou-peiraia/',
+        'md5': '2ebe96661cb81854889053cebb661068',
+        'info_dict': {
+            'id': '520979',
+            'ext': 'mp4',
+            'title': 'md5:70eef71a9cd2c1ecff7ee428354dded2',
+            'description': 'md5:0209fa8d318128569c0d256a5c404db1',
+            'timestamp': 1634975747,
+            'upload_date': '20211023',
+        },
+    }, {
+        'url': 'https://www.megatv.com/tvshows/527800/epeisodio-65-12/',
+        'md5': '8ab0c9d664cea11678670202b87bb2b1',
+        'info_dict': {
+            'id': '527800',
+            'ext': 'mp4',
+            'title': 'md5:fc322cb51f682eecfe2f54cd5ab3a157',
+            'description': 'md5:b2b7ed3690a78f2a0156eb790fdc00df',
+            'timestamp': 1636048859,
+            'upload_date': '20211104',
+        },
+    }]
+
+    def _match_article_id(self, webpage):
+        ART_RE = r'''(?x)
+        <article(?:
+            id=(?P<_q2>["'])Article_(?P<article>\d+)(?P=_q2)|
+            [^>]*?
+        )+>
+        '''
+        return compat_str(self._search_regex(ART_RE, webpage, 'article_id',
+                                             group='article'))
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        _is_article = video_id == 'None'
+        webpage = self._download_webpage(url,
+                                               'N/A' if _is_article else
+                                               video_id)
+        if _is_article:
+            video_id = self._match_article_id(webpage)
+        player_attrs = self._extract_player_attrs(webpage)
+        title = player_attrs.get('label') or self._og_search_title(webpage)
+        description = clean_html(get_element_by_class(
+            'article-wrapper' if _is_article else 'story_content',
+            webpage))
+        if not description:
+            description = self._og_search_description(webpage)
+        thumbnail = player_attrs.get('image') or \
+            self._og_search_thumbnail(webpage)
+        timestamp = unified_timestamp(self._html_search_meta(
+            'article:published_time', webpage))
+        try:
+            source = player_attrs['source']
+        except KeyError:
+            raise ExtractorError('no source found for %s' % video_id)
+        formats = self._extract_m3u8_formats(source, video_id, 'mp4') \
+            if determine_ext(source) == 'm3u8' else [source]
+        self._sort_formats(formats)
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+            'formats': formats,
+        }
+
+
+class MegaTVComEmbedIE(MegaTVComBaseIE):
+    IE_NAME = 'megatvcom:embed'
+    IE_DESC = 'megatv.com embedded videos'
+    _VALID_URL = r'https?://(?:www\.)?megatv\.com/embed/?\?p=\d+'
+
+    _TESTS = [{
+        'url': 'https://www.megatv.com/embed/?p=2020520979',
+        'md5': '2ebe96661cb81854889053cebb661068',
+        'info_dict': {
+            'id': '520979',
+            'ext': 'mp4',
+            'title': 'md5:70eef71a9cd2c1ecff7ee428354dded2',
+            'description': 'md5:0209fa8d318128569c0d256a5c404db1',
+            'timestamp': 1634975747,
+            'upload_date': '20211023',
+        },
+    }, {
+        'url': 'https://www.megatv.com/embed/?p=2020534081',
+        'md5': 'f9a15e315acbf01b128e8efa3f75aab3',
+        'info_dict': {
+            'id': '534081',
+            'ext': 'mp4',
+            'title': 'md5:062e9d5976ef854d8bdc1f5724d9b2d0',
+            'description': 'md5:36dbe4c3762d2ede9513eea8d07f6d52',
+            'timestamp': 1636376351,
+            'upload_date': '20211108',
+        },
+    }]
+
+    @classmethod
+    def _extract_urls(cls, webpage, origin_url=None):
+        # make the scheme in _VALID_URL optional
+        _URL_RE = r'(?:https?:)?//' + cls._VALID_URL.split('://', 1)[1]
+        EMBED_RE = r'''(?x)
+            <iframe[^>]+?src=(?P<_q1>%(quot_re)s)
+                (?P<url>%(url_re)s)(?P=_q1)
+        ''' % {'quot_re': r'["\']', 'url_re': _URL_RE}
+        for mobj in re.finditer(EMBED_RE, webpage):
+            url = unescapeHTML(mobj.group('url'))
+            if url.startswith('//'):
+                scheme = compat_urllib_parse_urlparse(origin_url).scheme \
+                    if origin_url else 'https'
+                url = '%s:%s' % (scheme, url)
+            yield url
+
+    def _match_canonical_url(self, webpage):
+        LINK_RE = r'''(?x)
+        <link(?:
+            rel=(?P<_q1>%(quot_re)s)(?P<canonical>canonical)(?P=_q1)|
+            href=(?P<_q2>%(quot_re)s)(?P<href>(?:(?!(?P=_q2)).)+)(?P=_q2)|
+            [^>]*?
+        )+>
+        ''' % {'quot_re': r'["\']'}
+        for mobj in re.finditer(LINK_RE, webpage):
+            canonical, href = mobj.group('canonical', 'href')
+            if canonical and href:
+                return unescapeHTML(href)
+
+    def _real_extract(self, url):
+        webpage = self._download_webpage(url, 'N/A')
+        player_attrs = self._extract_player_attrs(webpage)
+        canonical_url = player_attrs.get('share_url') or \
+            self._match_canonical_url(webpage)
+        if not canonical_url:
+            raise ExtractorError('canonical URL not found')
+        video_id = compat_parse_qs(compat_urllib_parse_urlparse(
+            canonical_url).query)['p'][0]
+
+        # Resolve the canonical URL, following redirects, and defer to
+        # megatvcom, as the metadata extracted from the embeddable page some
+        # times are slightly different, for the same video
+        canonical_url = self._request_webpage(
+            HEADRequest(canonical_url), video_id,
+            note='Resolve canonical URL',
+            errnote='Could not resolve canonical URL').geturl()
+        return self.url_result(
+            canonical_url,
+            MegaTVComIE.ie_key(),
+            video_id
+        )

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',