Merge 3872619ed5 into 1036478d13

* WEB URLs are, MWEB not
* resolves #33017

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/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/youtube.py

@@ -9,6 +9,7 @@ import json
 import os.path
 import random
 import re
+import string
 import time
 import traceback
 
@@ -67,6 +68,7 @@ from ..utils import (
 
 class YoutubeBaseInfoExtractor(InfoExtractor):
     """Provide base functions for Youtube extractors"""
+
     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
     _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
 
@@ -138,7 +140,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 [2, 1, None, 1,
                  'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
                  None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ],
             username,
         ]
@@ -160,7 +162,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             None, 1, None, [1, None, None, None, [password, None, True]],
             [
                 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
-                1, [None, None, []], None, None, None, True
+                1, [None, None, []], None, None, None, True,
             ]]
 
         challenge_results = req(
@@ -213,7 +215,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                     user_hash, None, 2, None,
                     [
                         9, None, None, None, None, None, None, None,
-                        [None, tfa_code, True, 2]
+                        [None, tfa_code, True, 2],
                     ]]
 
                 tfa_results = req(
@@ -284,7 +286,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': '2.20201021.03.00',
-            }
+            },
         },
     }
 
@@ -385,7 +387,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
                 'client': {
                     'clientName': 'WEB',
                     'clientVersion': '2.20201021.03.00',
-                }
+                },
             },
             'query': query,
         }
@@ -462,7 +464,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             #       (HTML, videodetails, metadata, renderers)
             'name': ('content', 'author', (('ownerChannelName', None), 'title'), ['text']),
             'url': ('href', 'ownerProfileUrl', 'vanityChannelUrl',
-                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl'])
+                    ['navigationEndpoint', 'browseEndpoint', 'canonicalBaseUrl']),
         }
         if any((videodetails, metadata, renderers)):
             result = (
@@ -671,7 +673,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
                 'description': '',
                 'uploader': '8KVIDEO',
-                'title': 'UHDTV TEST 8K VIDEO.mp4'
+                'title': 'UHDTV TEST 8K VIDEO.mp4',
             },
             'params': {
                 'youtube_include_dash_manifest': True,
@@ -711,7 +713,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/@theamazingatheist',
                 'title': 'Burning Everyone\'s Koran',
                 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
-            }
+            },
         },
         # Age-gated videos
         {
@@ -839,7 +841,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             },
             'expected_warnings': [
                 'DASH manifest missing',
-            ]
+            ],
         },
         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
         {
@@ -1820,8 +1822,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         # cpn generation algorithm is reverse engineered from base.js.
         # In fact it works even with dummy cpn.
-        CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
+        CPN_ALPHABET = string.ascii_letters + string.digits + '-_'
-        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
+        cpn = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(16))
 
         # more consistent results setting it to right before the end
         qs = parse_qs(playback_url)
@@ -1881,8 +1883,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
         if mobj is None:
             raise ExtractorError('Invalid URL: %s' % url)
-        video_id = mobj.group(2)
+        return mobj.group(2)
-        return video_id
 
     def _extract_chapters_from_json(self, data, video_id, duration):
         chapters_list = try_get(
@@ -2035,7 +2036,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             headers = {
                 'X-YouTube-Client-Name': '85',
                 'X-YouTube-Client-Version': '2.0',
-                'Origin': 'https://www.youtube.com'
+                'Origin': 'https://www.youtube.com',
             }
 
             video_info = self._call_api('player', query, video_id, fatal=False, headers=headers)
@@ -2064,8 +2065,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)])
 
         search_meta = (
-            lambda x: self._html_search_meta(x, webpage, default=None)) \
+            (lambda x: self._html_search_meta(x, webpage, default=None))
-            if webpage else lambda x: None
+            if webpage else lambda _: None)
 
         video_details = player_response.get('videoDetails') or {}
         microformat = try_get(
@@ -2137,7 +2138,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         def build_fragments(f):
             return LazyList({
                 'url': update_url_query(f['url'], {
-                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize']))
+                    'range': '{0}-{1}'.format(range_start, min(range_start + CHUNK_SIZE - 1, f['filesize'])),
                 })
             } for range_start in range(0, f['filesize'], CHUNK_SIZE))
 
@@ -2236,7 +2237,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     'protocol': 'http_dash_segments',
                     'fragments': build_fragments(dct),
                 } if dct['filesize'] else {
-                    'downloader_options': {'http_chunk_size': CHUNK_SIZE}  # No longer useful?
+                    'downloader_options': {'http_chunk_size': CHUNK_SIZE},  # No longer useful?
                 })
 
             formats.append(dct)
@@ -2414,9 +2415,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'is_live': is_live,
         }
 
-        pctr = try_get(
+        pctr = traverse_obj(
             player_response,
-            lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
+            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
         if pctr:
             def process_language(container, base_url, lang_code, query):
                 lang_subs = []
@@ -2430,9 +2431,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     })
                 container[lang_code] = lang_subs
 
+            def process_subtitles():
                 subtitles = {}
-            for caption_track in (pctr.get('captionTracks') or []):
+                for caption_track in traverse_obj(pctr, (
-                base_url = caption_track.get('baseUrl')
+                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                    base_url = self._yt_urljoin(caption_track['baseUrl'])
                     if not base_url:
                         continue
                     if caption_track.get('kind') != 'asr':
@@ -2443,18 +2446,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                             subtitles, base_url, lang_code, {})
                         continue
                     automatic_captions = {}
-                for translation_language in (pctr.get('translationLanguages') or []):
+                    for translation_language in traverse_obj(pctr, (
-                    translation_language_code = translation_language.get('languageCode')
+                            'translationLanguages', lambda _, v: v.get('languageCode'))):
-                    if not translation_language_code:
+                        translation_language_code = translation_language['languageCode']
-                        continue
                         process_language(
                             automatic_captions, base_url, translation_language_code,
                             {'tlang': translation_language_code})
                     info['automatic_captions'] = automatic_captions
                 info['subtitles'] = subtitles
 
+            process_subtitles()
+
         parsed_url = compat_urllib_parse_urlparse(url)
-        for component in [parsed_url.fragment, parsed_url.query]:
+        for component in (parsed_url.fragment, parsed_url.query):
             query = compat_parse_qs(component)
             for k, v in query.items():
                 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
@@ -2684,7 +2688,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'title': 'Super Cooper Shorts - Shorts',
             'uploader': 'Super Cooper Shorts',
             'uploader_id': '@SuperCooperShorts',
-        }
+        },
     }, {
         # Channel that does not have a Shorts tab. Test should just download videos on Home tab instead
         'url': 'https://www.youtube.com/@emergencyawesome/shorts',
@@ -2738,7 +2742,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'description': 'md5:609399d937ea957b0f53cbffb747a14c',
             'uploader': 'ThirstForScience',
             'uploader_id': '@ThirstForScience',
-        }
+        },
     }, {
         'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
         'only_matching': True,
@@ -3037,7 +3041,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'uploader': '3Blue1Brown',
             'uploader_id': '@3blue1brown',
             'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
-        }
+        },
     }]
 
     @classmethod
@@ -3335,7 +3339,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             'client': {
                 'clientName': 'WEB',
                 'clientVersion': client_version,
-            }
+            },
         }
         visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str)
 
@@ -3354,7 +3358,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 headers['x-goog-visitor-id'] = visitor_data
             data['continuation'] = continuation['continuation']
             data['clickTracking'] = {
-                'clickTrackingParams': continuation['itct']
+                'clickTrackingParams': continuation['itct'],
             }
             count = 0
             retries = 3
@@ -3613,7 +3617,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'milan',
             'uploader_id': '@milan5503',
             'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
-        }
+        },
     }, {
         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
         'playlist_mincount': 455,
@@ -3623,7 +3627,7 @@ class YoutubePlaylistIE(InfoExtractor):
             'uploader': 'LBK',
             'uploader_id': '@music_king',
             'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA',
-        }
+        },
     }, {
         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
         'only_matching': True,
@@ -3734,7 +3738,7 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
     def _get_n_results(self, query, n):
@@ -3754,7 +3758,7 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
         'info_dict': {
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
-        }
+        },
     }]
 
 
@@ -3769,7 +3773,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
             'id': 'youtube-dl test video',
             'title': 'youtube-dl test video',
         },
-        'params': {'playlistend': 5}
+        'params': {'playlistend': 5},
     }, {
         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
         'only_matching': True,
@@ -3785,6 +3789,7 @@ class YoutubeSearchURLIE(YoutubeBaseInfoExtractor):
 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
     """
     Base class for feed extractors
+
     Subclasses must define the _FEED_NAME property.
     """
     _LOGIN_REQUIRED = True