Merge 3872619ed5210e8a60f68cff7dbb64164bdc7b5c into 4e714f9df1ed2cccd51df60d45ff5504abe827b7

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

youtube_dl/extractor/ant1newsgr.py

@@ -0,0 +1,188 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+)
+from ..utils import (
+    HEADRequest,
+    ExtractorError,
+    determine_ext,
+    smuggle_url,
+    unsmuggle_url,
+    unescapeHTML,
+)
+
+
+class Ant1NewsGrBaseIE(InfoExtractor):
+    @staticmethod
+    def _smuggle_parent_info(url, **info_dict):
+        return smuggle_url(url, {'parent_info': info_dict})
+
+    @staticmethod
+    def _unsmuggle_parent_info(url):
+        unsmuggled_url, data = unsmuggle_url(url, default={'parent_info': {}})
+        return unsmuggled_url, data['parent_info']
+
+    def _download_api_data(self, netloc, cid, scheme='https'):
+        url_parts = (scheme, netloc, self._API_PATH, None, None, None)
+        url = compat_urlparse.urlunparse(url_parts)
+        query = {'cid': cid}
+        return self._download_json(
+            url, cid,
+            'Downloading JSON',
+            'Unable to download JSON',
+            query=query)
+
+    def _download_and_extract_api_data(self, video_id, *args, **kwargs):
+        info = self._download_api_data(*args, **kwargs)
+        try:
+            source = info['url']
+        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': info['title'],
+            'thumbnail': info['thumb'],
+            'formats': formats,
+        }
+
+
+class Ant1NewsGrWatchIE(Ant1NewsGrBaseIE):
+    IE_NAME = 'ant1newsgr:watch'
+    IE_DESC = 'ant1news.gr videos'
+    _VALID_URL = r'https?://(?:www\.)?ant1news\.gr/watch/(?P<id>\d+)/'
+    _API_PATH = '/templates/data/player'
+
+    _TEST = {
+        'url': 'https://www.ant1news.gr/watch/1506168/ant1-news-09112021-stis-18-45',
+        'md5': '60a984da5ffc98c9924e6d9dd46c6f04',
+        'info_dict': {
+            'id': '1506168',
+            'ext': 'mp4',
+            'title': 'md5:0ad00fa66ecf8aa233d26ab0dba7514a',
+            'description': 'md5:18665af715a6dcfeac1d6153a44f16b0',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        scheme, netloc, _, _, _, _ = compat_urllib_parse_urlparse(url)
+        info = self._download_and_extract_api_data(
+            video_id, netloc, video_id, scheme=scheme)
+        info['description'] = self._og_search_description(webpage)
+        return info
+
+
+class Ant1NewsGrArticleIE(Ant1NewsGrBaseIE):
+    IE_NAME = 'ant1newsgr:article'
+    IE_DESC = 'ant1news.gr articles'
+    _VALID_URL = r'https?://(?:www\.)?ant1news\.gr/[^/]+/article/(?P<id>\d+)/'
+
+    _TESTS = [{
+        'url': 'https://www.ant1news.gr/afieromata/article/549468/o-tzeims-mpont-sta-meteora-oi-apeiles-kai-o-xesikomos-ton-kalogeron',
+        'md5': 'eb635a194c15272c2611a751766b0200',
+        'info_dict': {
+            'id': '_xvg/m_cmbatw=',
+            'ext': 'mp4',
+            'title': 'md5:a93e8ecf2e4073bfdffcb38f59945411',
+        },
+        'expected_warnings': [r'^[Uu]nable to extract JSON-LD'],
+    }, {
+        'url': 'https://ant1news.gr/Society/article/620286/symmoria-anilikon-dikigoros-thymaton-ithelan-na-toys-apoteleiosoyn',
+        'info_dict': {
+            'id': '620286',
+            'title': 'md5:91fe569e952e4d146485740ae927662b',
+        },
+        'expected_warnings': [r'^[Uu]nable to extract JSON-LD'],
+        'playlist_mincount': 2,
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        info = self._search_json_ld(webpage, video_id,
+                                    expected_type='NewsArticle',
+                                    fatal=False)
+        # workaround as _json_ld does not recognize @graph nesting
+        if not info:
+            info['title'] = self._og_search_title(webpage)
+        embed_urls = list(Ant1NewsGrEmbedIE._extract_urls(webpage, url, **info))
+        if not embed_urls:
+            raise ExtractorError('no videos found for %s' % video_id)
+        if len(embed_urls) == 1:
+            return self.url_result(embed_urls[0], ie=Ant1NewsGrEmbedIE.ie_key(),
+                                   video_title=info['title'])
+        return self.playlist_from_matches(
+            embed_urls, video_id, info['title'], ie=Ant1NewsGrEmbedIE.ie_key())
+
+
+class Ant1NewsGrEmbedIE(Ant1NewsGrBaseIE):
+    IE_NAME = 'ant1newsgr:embed'
+    IE_DESC = 'ant1news.gr embedded videos'
+    _VALID_URL = r'''(?x)https?://(?:[a-zA-Z0-9\-]+\.)?
+        (?:antenna|ant1news)\.gr/templates/pages/player
+        \?(?:(?:cid=(?P<id>[^&#]+)|[^&=#]+=[^&#]+)&?)+'''
+    _API_PATH = '/news/templates/data/jsonPlayer'
+
+    _TEST = {
+        'url': 'https://www.antenna.gr/templates/pages/player?cid=3f_li_c_az_jw_y_u=&w=670&h=377',
+        'md5': '12872b12af18b5dbf76528786728de8c',
+        'info_dict': {
+            'id': '3f_li_c_az_jw_y_u=',
+            'ext': 'mp4',
+            'title': 'md5:a30c93332455f53e1e84ae0724f0adf7',
+        },
+    }
+
+    @classmethod
+    def _extract_urls(cls, webpage, origin_url=None, **parent_info):
+        # make the scheme in _VALID_URL optional
+        _URL_RE = r'(?:https?:)?//' + cls._VALID_URL.split('://', 1)[1]
+        # simplify the query string part of _VALID_URL; after extracting iframe
+        # src, the URL will be matched again
+        _URL_RE = _URL_RE.split(r'\?', 1)[0] + r'\?(?:(?!(?P=_q1)).)+'
+        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)
+            if not cls.suitable(url):
+                continue
+            yield cls._smuggle_parent_info(url, **parent_info)
+
+    def _real_extract(self, url):
+        url, parent_info = type(self)._unsmuggle_parent_info(url)
+        video_id = self._match_id(url)
+
+        # resolve any redirects, to derive the proper base URL for the API query
+        canonical_url = self._request_webpage(
+            HEADRequest(url), video_id,
+            note='Resolve canonical player URL',
+            errnote='Could not resolve canonical player URL').geturl()
+        scheme, netloc, _, _, query, _ = compat_urllib_parse_urlparse(
+            canonical_url)
+        query = compat_parse_qs(query)
+        cid = query['cid'][0]
+
+        info = self._download_and_extract_api_data(
+            video_id, netloc, cid, scheme=scheme)
+        if 'timestamp' not in info and 'timestamp' in parent_info:
+            info['timestamp'] = parent_info['timestamp']
+        return info

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 +1078,11 @@ from .rutube import (
     RutubePersonIE,
     RutubePlaylistIE,
 )
+from .ant1newsgr import (
+    Ant1NewsGrWatchIE,
+    Ant1NewsGrArticleIE,
+    Ant1NewsGrEmbedIE,
+)
 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 .ant1newsgr import Ant1NewsGrEmbedIE
 from .limelight import LimelightBaseIE
 from .anvato import AnvatoIE
 from .washingtonpost import WashingtonPostIE
@@ -3400,6 +3401,13 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 rutube_urls, video_id, video_title, ie=RutubeIE.ie_key())
 
+        # Look for ant1news.gr embeds
+        ant1newsgr_urls = list(Ant1NewsGrEmbedIE._extract_urls(webpage, url,
+                                                               title=video_title))
+        if ant1newsgr_urls:
+            return self.playlist_from_matches(
+                ant1newsgr_urls, video_id, video_title, ie=Ant1NewsGrEmbedIE.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/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',