Merge 7e51ac7712d122beff12d2ef8483b1951f325747 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

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

youtube_dl/extractor/atresplayer.py

@@ -6,14 +6,19 @@ import re
 from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
+    base_url,
     ExtractorError,
     int_or_none,
     urlencode_postdata,
+    urljoin,
+    xpath_element,
+    xpath_text,
+    xpath_with_ns,
 )
 
 
 class AtresPlayerIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/[^/]+/[^/]+/[^/]+/[^/]+/(?P<display_id>.+?)_(?P<id>[0-9a-f]{24})'
+    _VALID_URL = r'https?://(?:www\.)?atresplayer\.com/[^/]+/[^/]+/[^/]+(?:/[^/]+)?/(?P<display_id>.+?)_(?P<id>[0-9a-f]{24})'
     _NETRC_MACHINE = 'atresplayer'
     _TESTS = [
         {
@@ -39,7 +44,6 @@ class AtresPlayerIE(InfoExtractor):
             'only_matching': True,
         },
     ]
-    _API_BASE = 'https://api.atresplayer.com/'
 
     def _real_initialize(self):
         self._login()
@@ -49,6 +53,8 @@ class AtresPlayerIE(InfoExtractor):
             error = self._parse_json(e.cause.read(), None)
             if error.get('error') == 'required_registered':
                 self.raise_login_required()
+            if error.get('error') == 'invalid_request':
+                raise ExtractorError('Authentication failed', expected=True)
             raise ExtractorError(error['error_description'], expected=True)
         raise
 
@@ -57,35 +63,65 @@ class AtresPlayerIE(InfoExtractor):
         if username is None:
             return
 
-        self._request_webpage(
-            self._API_BASE + 'login', None, 'Downloading login page')
-
         try:
-            target_url = self._download_json(
-                'https://account.atresmedia.com/api/login', None,
+            self._download_webpage(
+                'https://account.atresplayer.com/auth/v1/login', None,
                 'Logging in', headers={
                     'Content-Type': 'application/x-www-form-urlencoded'
                 }, data=urlencode_postdata({
                     'username': username,
                     'password': password,
-                }))['targetUrl']
+                }))
         except ExtractorError as e:
             self._handle_error(e, 400)
 
-        self._request_webpage(target_url, None, 'Following Target URL')
+    def _get_mpd_subtitles(self, mpd_xml, mpd_url):
+        subs = {}
+
+        def _add_ns(name):
+            return xpath_with_ns(name, {
+                'mpd': 'urn:mpeg:dash:schema:mpd:2011'
+            })
+
+        def _is_mime_type(node, mime_type):
+            return node.attrib.get('mimeType') == mime_type
+
+        text_nodes = mpd_xml.findall(
+            _add_ns('mpd:Period/mpd:AdaptationSet[@contentType="text"]'))
+        for adaptation_set in text_nodes:
+            lang = adaptation_set.attrib['lang']
+            representation = xpath_element(adaptation_set, _add_ns('mpd:Representation'))
+            subs_url = xpath_text(representation, _add_ns('mpd:BaseURL'))
+            if subs_url and (_is_mime_type(adaptation_set, 'text/vtt') or _is_mime_type(representation, 'text/vtt')):
+                subs.update({lang: [{
+                    'ext': 'vtt',
+                    'url': urljoin(mpd_url, subs_url),
+                }]})
+        return subs
 
     def _real_extract(self, url):
         display_id, video_id = re.match(self._VALID_URL, url).groups()
 
+        page = self._download_webpage(url, video_id, 'Downloading video page')
+        preloaded_state_regex = r'window\.__PRELOADED_STATE__\s*=\s*(\{(.*?)\});'
+        preloaded_state_text = self._html_search_regex(preloaded_state_regex, page, 'preloaded state')
+        preloaded_state = self._parse_json(preloaded_state_text, video_id)
+        link_info = next(iter(preloaded_state['links'].values()))
+
         try:
-            episode = self._download_json(
-                self._API_BASE + 'client/v1/player/episode/' + video_id, video_id)
+            metadata = self._download_json(link_info['href'], video_id)
+        except ExtractorError as e:
+            self._handle_error(e, 403)
+
+        try:
+            episode = self._download_json(metadata['urlVideo'], video_id)
         except ExtractorError as e:
             self._handle_error(e, 403)
 
         title = episode['titulo']
 
         formats = []
+        subtitles = {}
         for source in episode.get('sources', []):
             src = source.get('src')
             if not src:
@@ -96,8 +132,14 @@ class AtresPlayerIE(InfoExtractor):
                     src, video_id, 'mp4', 'm3u8_native',
                     m3u8_id='hls', fatal=False))
             elif src_type == 'application/dash+xml':
-                formats.extend(self._extract_mpd_formats(
-                    src, video_id, mpd_id='dash', fatal=False))
+                mpd = self._download_xml_handle(
+                    src, video_id, note='Downloading MPD manifest', fatal=False)
+                if mpd:
+                    mpd_doc, mpd_handle = mpd
+                    mpd_base_url = base_url(mpd_handle.geturl())
+                    subtitles = self._merge_subtitles(subtitles, self._get_mpd_subtitles(mpd_doc, mpd_base_url))
+                    formats.extend(self._parse_mpd_formats(
+                        mpd_doc, mpd_id='dash', mpd_base_url=mpd_base_url, mpd_url=src))
         self._sort_formats(formats)
 
         heartbeat = episode.get('heartbeat') or {}
@@ -115,4 +157,5 @@ class AtresPlayerIE(InfoExtractor):
             'channel': get_meta('channel'),
             'season': get_meta('season'),
             'episode_number': int_or_none(get_meta('episodeNumber')),
+            'subtitles': subtitles,
         }

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/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',

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError: