Merge 96a0ad4778 into 1036478d13

* WEB URLs are, MWEB not
* resolves #33017

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/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/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 = ''.join(CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16))
+        CPN_ALPHABET = string.ascii_letters + string.digits + '-_'
+        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 video_id
+        return mobj.group(2)
 
     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)) \
-            if webpage else lambda x: None
+            (lambda x: self._html_search_meta(x, webpage, default=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 = []
@@ -2430,31 +2431,34 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     })
                 container[lang_code] = lang_subs
 
-            subtitles = {}
-            for caption_track in (pctr.get('captionTracks') or []):
-                base_url = caption_track.get('baseUrl')
-                if not base_url:
-                    continue
-                if caption_track.get('kind') != 'asr':
-                    lang_code = caption_track.get('languageCode')
-                    if not lang_code:
+            def process_subtitles():
+                subtitles = {}
+                for caption_track in traverse_obj(pctr, (
+                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                    base_url = self._yt_urljoin(caption_track['baseUrl'])
+                    if not base_url:
                         continue
-                    process_language(
-                        subtitles, base_url, lang_code, {})
-                    continue
-                automatic_captions = {}
-                for translation_language in (pctr.get('translationLanguages') or []):
-                    translation_language_code = translation_language.get('languageCode')
-                    if not translation_language_code:
+                    if caption_track.get('kind') != 'asr':
+                        lang_code = caption_track.get('languageCode')
+                        if not lang_code:
+                            continue
+                        process_language(
+                            subtitles, base_url, lang_code, {})
                         continue
-                    process_language(
-                        automatic_captions, base_url, translation_language_code,
-                        {'tlang': translation_language_code})
-                info['automatic_captions'] = automatic_captions
-            info['subtitles'] = subtitles
+                    automatic_captions = {}
+                    for translation_language in traverse_obj(pctr, (
+                            'translationLanguages', lambda _, v: v.get('languageCode'))):
+                        translation_language_code = translation_language['languageCode']
+                        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