Merge ea4948068d323a081324f69a805df14ddab13f6f 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/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/platzi.py

@@ -1,16 +1,23 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import (
-    compat_b64decode,
+    compat_kwargs,
     compat_str,
+    compat_urllib_parse_urlparse,
 )
 from ..utils import (
     clean_html,
+    dict_get,
     ExtractorError,
+    get_element_by_class,
     int_or_none,
+    parse_iso8601,
     str_or_none,
+    strip_or_none,
     try_get,
     url_or_none,
     urlencode_postdata,
@@ -22,6 +29,42 @@ class PlatziBaseIE(InfoExtractor):
     _LOGIN_URL = 'https://platzi.com/login/'
     _NETRC_MACHINE = 'platzi'
 
+    def _raise_extractor_error(self, video_id, reason, expected=True):
+        raise ExtractorError('[%s] %s: %s' % (self.IE_NAME, video_id, reason), expected=expected)
+
+    def _download_webpage(self, url_or_request, video_id, *args, **kwargs):
+        # CF likes Connection: keep-alive and so disfavours Py2
+        # retry on 403 may get in
+        kwargs['expected_status'] = 403
+        # header parameters required fpor Py3 to breach site's CF fence w/o 403
+        headers = kwargs.get('headers') or {}
+        new_hdrs = {}
+        if 'User-Agent' not in headers:
+            headers['User-Agent'] = 'Mozilla/5.0'  # (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/93.0.0.0 Safari/537.36'
+            kwargs['headers'] = new_hdrs = headers
+        if new_hdrs:
+            kwargs = compat_kwargs(kwargs)
+        for _ in range(2):
+            x = super(PlatziBaseIE, self)._download_webpage_handle(url_or_request, video_id, *args, **kwargs)
+            if x is False:
+                return x
+            if x[1].getcode() != 403:
+                break
+            kwargs.pop('expected_status', None)
+            note = kwargs.pop('note', '')
+            kwargs['note'] = (note or 'Downloading webpage') + ' - retrying'
+            kwargs = compat_kwargs(kwargs)
+        path = compat_urllib_parse_urlparse(x[1].geturl())
+        if path == '/':
+            self._raise_extractor_error(video_id, 'Redirected to home page: content expired?')
+        elif path == '/login':
+            self.raise_login_required()
+        else:
+            errs = clean_html(get_element_by_class('Errorpage-text', x[0]))
+            if errs:
+                self._raise_extractor_error(video_id, errs)
+        return x[0]
+
     def _real_initialize(self):
         self._login()
 
@@ -75,6 +118,26 @@ class PlatziIE(PlatziBaseIE):
                     '''
 
     _TESTS = [{
+        'url': 'https://platzi.com/clases/1927-intro-selenium/29383-bienvenida-al-curso',
+        'md5': '0af120f1ffd18a2246f19099d52b83e2',
+        'info_dict': {
+            'id': '29383',
+            'ext': 'mp4',
+            'title': 'Por qué aprender Selenium y qué verás',
+            'description': 'md5:bbe91d2760052ca4054a3149a6580436',
+            'timestamp': 1627400390,
+            'upload_date': '20210727',
+            'creator': 'Héctor Vega',
+            'series': 'Curso de Introducción a Selenium con Python',
+            'duration': 11700,
+            'categories': list,
+        },
+        'params': {
+            'format': 'bestvideo',
+            # 'skip_download': True,
+        },
+        'expected_warnings': ['HTTP Error 401']
+    }, {
         'url': 'https://platzi.com/clases/1311-next-js/12074-creando-nuestra-primera-pagina/',
         'md5': '8f56448241005b561c10f11a595b37e3',
         'info_dict': {
@@ -84,7 +147,7 @@ class PlatziIE(PlatziBaseIE):
             'description': 'md5:4c866e45034fc76412fbf6e60ae008bc',
             'duration': 420,
         },
-        'skip': 'Requires platzi account credentials',
+        'skip': 'Content expired',
     }, {
         'url': 'https://courses.platzi.com/classes/1367-communication-codestream/13430-background/',
         'info_dict': {
@@ -94,10 +157,7 @@ class PlatziIE(PlatziBaseIE):
             'description': 'md5:49c83c09404b15e6e71defaf87f6b305',
             'duration': 360,
         },
-        'skip': 'Requires platzi account credentials',
-        'params': {
-            'skip_download': True,
-        },
+        'skip': 'Content expired',
     }]
 
     def _real_extract(self, url):
@@ -105,50 +165,60 @@ class PlatziIE(PlatziBaseIE):
 
         webpage = self._download_webpage(url, lecture_id)
 
-        data = self._parse_json(
+        data_preloaded_state = self._parse_json(
             self._search_regex(
-                # client_data may contain "};" so that we have to try more
-                # strict regex first
-                (r'client_data\s*=\s*({.+?})\s*;\s*\n',
-                 r'client_data\s*=\s*({.+?})\s*;'),
-                webpage, 'client data'),
+                (r'window\s*.\s*__PRELOADED_STATE__\s*=\s*({.*?});?\s*</script'), webpage, 'client data'),
             lecture_id)
 
-        material = data['initialState']['material']
-        desc = material['description']
-        title = desc['title']
+        video_player = try_get(data_preloaded_state, lambda x: x['videoPlayer'], dict) or {}
+        title = strip_or_none(video_player.get('name')) or self._og_search_title(webpage)
+        servers = try_get(video_player, lambda x: x['video']['servers'], dict) or {}
+        if not servers and try_get(video_player, lambda x: x['blockedInfo']['blocked']):
+            why = video_player['blockedInfo'].get('type') or 'unspecified'
+            if why == 'unlogged':
+                self.raise_login_required()
+            self._raise_extractor_error(lecture_id, 'All video formats blocked because ' + why)
 
         formats = []
-        for server_id, server in material['videos'].items():
-            if not isinstance(server, dict):
+        headers = {'Referer': url}
+        extractions = {
+            'hls': lambda x: formats.extend(self._extract_m3u8_formats(
+                server_json[x], lecture_id, 'mp4',
+                entry_protocol='m3u8_native', m3u8_id='hls',
+                note='Downloading %s m3u8 information' % (server_json.get('id', x), ),
+                headers=headers, fatal=False)),
+            'dash': lambda x: formats.extend(self._extract_mpd_formats(
+                server_json[x], lecture_id, mpd_id='dash',
+                note='Downloading %s MPD manifest' % (server_json.get('id', x), ),
+                headers=headers, fatal=False)),
+        }
+        for server, server_json in servers.items():
+            if not isinstance(server_json, dict):
                 continue
-            for format_id in ('hls', 'dash'):
-                format_url = url_or_none(server.get(format_id))
-                if not format_url:
-                    continue
-                if format_id == 'hls':
-                    formats.extend(self._extract_m3u8_formats(
-                        format_url, lecture_id, 'mp4',
-                        entry_protocol='m3u8_native', m3u8_id=format_id,
-                        note='Downloading %s m3u8 information' % server_id,
-                        fatal=False))
-                elif format_id == 'dash':
-                    formats.extend(self._extract_mpd_formats(
-                        format_url, lecture_id, mpd_id=format_id,
-                        note='Downloading %s MPD manifest' % server_id,
-                        fatal=False))
+            for fmt in server_json.keys():
+                extraction = extractions.get(fmt)
+                if callable(extraction):
+                    extraction(fmt)
         self._sort_formats(formats)
+        for f in formats:
+            f.setdefault('http_headers', {})['Referer'] = headers['Referer']
 
-        content = str_or_none(desc.get('content'))
-        description = (clean_html(compat_b64decode(content).decode('utf-8'))
-                       if content else None)
-        duration = int_or_none(material.get('duration'), invscale=60)
+        def categories():
+            cat = strip_or_none(video_player.get('courseCategory'))
+            if cat:
+                return [cat]
 
         return {
             'id': lecture_id,
             'title': title,
-            'description': description,
-            'duration': duration,
+            'description': clean_html(video_player.get('courseDescription')) or self._og_search_description(webpage),
+            'duration': int_or_none(video_player.get('duration'), invscale=60),
+            'thumbnail': url_or_none(video_player.get('thumbnail')) or self._og_search_thumbnail(webpage),
+            'timestamp': parse_iso8601(dict_get(video_player, ('dateModified', 'datePublished'))),
+            'creator': strip_or_none(video_player.get('teacherName')) or clean_html(get_element_by_class('TeacherDetails-name', webpage)),
+            'comment_count': int_or_none(video_player.get('commentsNumber')),
+            'categories': categories(),
+            'series': strip_or_none(video_player.get('courseTitle')) or None,
             'formats': formats,
         }
 
@@ -157,17 +227,35 @@ class PlatziCourseIE(PlatziBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            platzi\.com/clases|           # es version
-                            courses\.platzi\.com/classes  # en version
+                            (?P<clas>
+                                platzi\.com/clases|           # es version
+                                courses\.platzi\.com/classes  # en version
+                            )|
+                            platzi\.com(?:/(?P<curs>cursos))?
                         )/(?P<id>[^/?\#&]+)
                     '''
     _TESTS = [{
+        'url': 'https://platzi.com/web-angular/',
+        'info_dict': {
+            'id': 'web-angular',
+            'title': 'Frontend con Angular',
+        },
+        'playlist_count': 9,
+    }, {
+        'url': 'https://platzi.com/cursos/angular/',
+        'info_dict': {
+            'id': '2478',
+            'title': 'Curso de Fundamentos de Angular',
+        },
+        'playlist_count': 21,
+    }, {
         'url': 'https://platzi.com/clases/next-js/',
         'info_dict': {
             'id': '1311',
             'title': 'Curso de Next.js',
         },
         'playlist_count': 22,
+        'skip': 'Oops (updating page)',
     }, {
         'url': 'https://courses.platzi.com/classes/communication-codestream/',
         'info_dict': {
@@ -175,23 +263,62 @@ class PlatziCourseIE(PlatziBaseIE):
             'title': 'Codestream Course',
         },
         'playlist_count': 14,
+        'skip': 'Content expired',
     }]
 
+    @classmethod
+    def _match_valid_url(cls, url):
+        return re.match(cls._VALID_URL, url)
+
     @classmethod
     def suitable(cls, url):
         return False if PlatziIE.suitable(url) else super(PlatziCourseIE, cls).suitable(url)
 
+    def __extract_things(self, webpage, thing_id, thing_pattern):
+        return self.playlist_from_matches(
+            re.finditer(thing_pattern, webpage),
+            playlist_id=thing_id,
+            playlist_title=self._og_search_title(webpage, default=None),
+            getter=lambda m: urljoin('https://platzi.com', m.group('path')))
+
+    def _extract_classes(self, webpage, course_id):
+        display_id = course_id
+        course_id = self._search_regex(
+            r'''(["'])courseId\1\s*:\s*(?P<id>\d+)''',
+            webpage, 'course id', group='id', fatal=False) or course_id
+        return self.__extract_things(
+            webpage, course_id,
+            r'''<a\b[^>]+\bhref\s*=\s*['"]?(?P<path>/clases/\d+-%s/[^/]+)'''
+            % (display_id, ))
+
+    def _extract_categories(self, webpage, cat_id):
+        return self.__extract_things(
+            webpage, cat_id,
+            r'''<a\b[^>]+\bhref\s*=\s*['"]?(?P<path>/cursos/[^/]+)''')
+
     def _real_extract(self, url):
-        course_name = self._match_id(url)
 
-        webpage = self._download_webpage(url, course_name)
+        m = self._match_valid_url(url)
+        classes, courses, this_id = m.group('clas', 'curs', 'id')
 
-        props = self._parse_json(
-            self._search_regex(r'data\s*=\s*({.+?})\s*;', webpage, 'data'),
-            course_name)['initialProps']
+        webpage = self._download_webpage(url, this_id)
 
+        if courses:
+            return self._extract_classes(webpage, this_id)
+
+        if not classes:
+            return self._extract_categories(webpage, this_id)
+
+        # this branch now seems always to give "Oops" pages
+        course_name = this_id
+
+        initialData = self._search_regex(
+            (r'window.initialData\s*=\s*({.+?})\s*;\s*\n', r'window.initialData\s*=\s*({.+?})\s*;'),
+            webpage, 'initialData')
+        props = self._parse_json(initialData, course_name, default={})
+        props = try_get(props, lambda x: x['initialProps'], dict) or {}
         entries = []
-        for chapter_num, chapter in enumerate(props['concepts'], 1):
+        for chapter_num, chapter in enumerate(props.get('concepts') or [], 1):
             if not isinstance(chapter, dict):
                 continue
             materials = chapter.get('materials')
@@ -221,4 +348,8 @@ class PlatziCourseIE(PlatziBaseIE):
         course_id = compat_str(try_get(props, lambda x: x['course']['id']))
         course_title = try_get(props, lambda x: x['course']['name'], compat_str)
 
-        return self.playlist_result(entries, course_id, course_title)
+        result = self.playlist_result(entries, course_id, course_title)
+        desc = clean_html(get_element_by_class('RouteDescription-content', webpage))
+        if desc:
+            result['description'] = desc
+        return result

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',