Back-port Paramount+ extractor (etc) from yt-dlp

With supporting changes to CBS and ThePlatform extractors

youtube_dl/extractor/cbs.py

@@ -8,6 +8,7 @@ from ..utils import (
     xpath_element,
     xpath_text,
     update_url_query,
+    url_or_none,
 )
 
 
@@ -25,10 +26,62 @@ class CBSBaseIE(ThePlatformFeedIE):
                     })
         return subtitles
 
+    def _extract_common_video_info(self, content_id, asset_types, mpx_acc, extra_info):
+        tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
+        tp_release_url = 'http://link.theplatform.com/s/' + tp_path
+        info = self._extract_theplatform_metadata(tp_path, content_id)
+
+        formats, subtitles = [], {}
+        last_e = None
+        for asset_type, query in asset_types.items():
+            try:
+                tp_formats, tp_subtitles = self._extract_theplatform_smil(
+                    update_url_query(tp_release_url, query), content_id,
+                    'Downloading %s SMIL data' % asset_type)
+            except ExtractorError as e:
+                last_e = e
+                if asset_type != 'fallback':
+                    continue
+                query['formats'] = ''  # blank query to check if expired
+            try:
+                tp_formats, tp_subtitles = self._extract_theplatform_smil(
+                    update_url_query(tp_release_url, query), content_id,
+                    'Downloading %s SMIL data, trying again with another format' % asset_type)
+            except ExtractorError as e:
+                last_e = e
+                continue
+            formats.extend(tp_formats)
+            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
+        if last_e and not formats:
+            raise last_e
+        self._sort_formats(formats)
+
+        extra_info.update({
+            'id': content_id,
+            'formats': formats,
+            'subtitles': subtitles,
+        })
+        info.update({k: v for k, v in extra_info.items() if v is not None})
+        return info
+
+    def _extract_video_info(self, *args, **kwargs):
+        # Extract assets + metadata and call _extract_common_video_info
+        raise NotImplementedError('This method must be implemented by subclasses')
+
+    def _real_extract(self, url):
+        return self._extract_video_info(self._match_id(url))
+
 
 class CBSIE(CBSBaseIE):
-    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:(?:cbs|paramountplus)\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
+    _VALID_URL = r'''(?x)
+        (?:
+            cbs:|
+            https?://(?:www\.)?(?:
+                cbs\.com/(?:shows/[^/]+/video|movies/[^/]+)/|
+                colbertlateshow\.com/(?:video|podcasts)/)
+        )(?P<id>[\w-]+)'''
 
+    # All tests are blocked outside US
     _TESTS = [{
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
         'info_dict': {
@@ -46,70 +99,70 @@ class CBSIE(CBSBaseIE):
             'skip_download': True,
         },
         '_skip': 'Blocked outside the US',
+    }, {
+        'url': 'https://www.cbs.com/shows/the-late-show-with-stephen-colbert/video/60icOhMb9NcjbcWnF_gub9XXHdeBcNk2/the-late-show-6-23-21-christine-baranski-joy-oladokun-',
+        'info_dict': {
+            'id': '60icOhMb9NcjbcWnF_gub9XXHdeBcNk2',
+            'title': 'The Late Show - 6/23/21 (Christine Baranski, Joy Oladokun)',
+            'timestamp': 1624507140,
+            'description': 'md5:e01af24e95c74d55e8775aef86117b95',
+            'uploader': 'CBSI-NEW',
+            'upload_date': '20210624',
+        },
+        'params': {
+            'ignore_no_formats_error': True,
+            'skip_download': True,
+        },
+        'expected_warnings': [
+            'This content expired on', 'No video formats found', 'Requested format is not available'],
     }, {
         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
         'only_matching': True,
     }, {
         'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
         'only_matching': True,
-    }, {
-        'url': 'https://www.paramountplus.com/shows/all-rise/video/QmR1WhNkh1a_IrdHZrbcRklm176X_rVc/all-rise-space/',
-        'only_matching': True,
     }]
 
     def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
         items_data = self._download_xml(
-            'http://can.cbs.com/thunder/player/videoPlayerService.php',
+            'https://can.cbs.com/thunder/player/videoPlayerService.php',
             content_id, query={'partner': site, 'contentId': content_id})
         video_data = xpath_element(items_data, './/item')
-        title = xpath_text(video_data, 'videoTitle', 'title', True)
+        title = xpath_text(video_data, 'videoTitle', 'title') or xpath_text(video_data, 'videotitle', 'title')
-        tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
-        tp_release_url = 'http://link.theplatform.com/s/' + tp_path
 
-        asset_types = []
+        asset_types = {}
-        subtitles = {}
+        has_drm = False
-        formats = []
-        last_e = None
         for item in items_data.findall('.//item'):
             asset_type = xpath_text(item, 'assetType')
-            if not asset_type or asset_type in asset_types or 'HLS_FPS' in asset_type or 'DASH_CENC' in asset_type:
-                continue
-            asset_types.append(asset_type)
             query = {
                 'mbr': 'true',
                 'assetTypes': asset_type,
             }
-            if asset_type.startswith('HLS') or asset_type in ('OnceURL', 'StreamPack'):
+            if not asset_type:
+                # fallback for content_ids that videoPlayerService doesn't return anything for
+                asset_type = 'fallback'
+                query['formats'] = 'M3U+none,MPEG4,M3U+appleHlsEncryption,MP3'
+                del query['assetTypes']
+            if asset_type in asset_types:
+                continue
+            elif any(excluded in asset_type for excluded in ('HLS_FPS', 'DASH_CENC', 'OnceURL')):
+                if 'DASH_CENC' in asset_type:
+                    has_drm = True
+                continue
+            if asset_type.startswith('HLS') or 'StreamPack' in asset_type:
                 query['formats'] = 'MPEG4,M3U'
             elif asset_type in ('RTMP', 'WIFI', '3G'):
                 query['formats'] = 'MPEG4,FLV'
-            try:
+            asset_types[asset_type] = query
-                tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                    update_url_query(tp_release_url, query), content_id,
-                    'Downloading %s SMIL data' % asset_type)
-            except ExtractorError as e:
-                last_e = e
-                continue
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        if last_e and not formats:
-            raise last_e
-        self._sort_formats(formats)
 
-        info = self._extract_theplatform_metadata(tp_path, content_id)
+        if not asset_types and has_drm:
-        info.update({
+            raise ExtractorError('Only DRM formats found', video_id=content_id, expected=True)
-            'id': content_id,
+
+        return self._extract_common_video_info(content_id, asset_types, mpx_acc, extra_info={
             'title': title,
             'series': xpath_text(video_data, 'seriesTitle'),
             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
-            'thumbnail': xpath_text(video_data, 'previewImageURL'),
+            'thumbnail': url_or_none(xpath_text(video_data, 'previewImageURL')),
-            'formats': formats,
-            'subtitles': subtitles,
         })
-        return info
-
-    def _real_extract(self, url):
-        content_id = self._match_id(url)
-        return self._extract_video_info(content_id)

youtube_dl/extractor/extractors.py

@@ -889,6 +889,10 @@ from .palcomp3 import (
     PalcoMP3VideoIE,
 )
 from .pandoratv import PandoraTVIE
+from .paramountplus import (
+     ParamountPlusIE,
+     ParamountPlusSeriesIE,
+)
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE

youtube_dl/extractor/paramountplus.py

@@ -0,0 +1,145 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .cbs import CBSBaseIE
+from ..utils import (
+    int_or_none,
+    url_or_none,
+)
+
+
+class ParamountPlusIE(CBSBaseIE):
+    _VALID_URL = r'''(?x)
+        (?:
+            paramountplus:|
+            https?://(?:www\.)?(?:
+                paramountplus\.com/(?:shows/[^/]+/video|movies/[^/]+)/
+        )(?P<id>[\w-]+))'''
+
+    # All tests are blocked outside US
+    _TESTS = [{
+        'url': 'https://www.paramountplus.com/shows/catdog/video/Oe44g5_NrlgiZE3aQVONleD6vXc8kP0k/catdog-climb-every-catdog-the-canine-mutiny/',
+        'info_dict': {
+            'id': 'Oe44g5_NrlgiZE3aQVONleD6vXc8kP0k',
+            'ext': 'mp4',
+            'title': 'CatDog - Climb Every CatDog/The Canine Mutiny',
+            'description': 'md5:7ac835000645a69933df226940e3c859',
+            'duration': 1426,
+            'timestamp': 920264400,
+            'upload_date': '19990301',
+            'uploader': 'CBSI-NEW',
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }, {
+        'url': 'https://www.paramountplus.com/shows/tooning-out-the-news/video/6hSWYWRrR9EUTz7IEe5fJKBhYvSUfexd/7-23-21-week-in-review-rep-jahana-hayes-howard-fineman-sen-michael-bennet-sheera-frenkel-cecilia-kang-/',
+        'info_dict': {
+            'id': '6hSWYWRrR9EUTz7IEe5fJKBhYvSUfexd',
+            'ext': 'mp4',
+            'title': '7/23/21 WEEK IN REVIEW (Rep. Jahana Hayes/Howard Fineman/Sen. Michael Bennet/Sheera Frenkel & Cecilia Kang)',
+            'description': 'md5:f4adcea3e8b106192022e121f1565bae',
+            'duration': 2506,
+            'timestamp': 1627063200,
+            'upload_date': '20210723',
+            'uploader': 'CBSI-NEW',
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }, {
+        'url': 'https://www.paramountplus.com/movies/daddys-home/vM2vm0kE6vsS2U41VhMRKTOVHyQAr6pC',
+        'info_dict': {
+            'id': 'vM2vm0kE6vsS2U41VhMRKTOVHyQAr6pC',
+            'ext': 'mp4',
+            'title': 'Daddy\'s Home',
+            'upload_date': '20151225',
+            'description': 'md5:9a6300c504d5e12000e8707f20c54745',
+            'uploader': 'CBSI-NEW',
+            'timestamp': 1451030400,
+        },
+        'params': {
+            'skip_download': 'm3u8',
+            'format': 'bestvideo',
+        },
+        'expected_warnings': ['Ignoring subtitle tracks'],  # TODO: Investigate this
+    }, {
+        'url': 'https://www.paramountplus.com/movies/sonic-the-hedgehog/5EKDXPOzdVf9voUqW6oRuocyAEeJGbEc',
+        'info_dict': {
+            'id': '5EKDXPOzdVf9voUqW6oRuocyAEeJGbEc',
+            'ext': 'mp4',
+            'uploader': 'CBSI-NEW',
+            'description': 'md5:bc7b6fea84ba631ef77a9bda9f2ff911',
+            'timestamp': 1577865600,
+            'title': 'Sonic the Hedgehog',
+            'upload_date': '20200101',
+        },
+        'params': {
+            'skip_download': 'm3u8',
+            'format': 'bestvideo',
+        },
+        'expected_warnings': ['Ignoring subtitle tracks'],
+    }, {
+        'url': 'https://www.paramountplus.com/shows/all-rise/video/QmR1WhNkh1a_IrdHZrbcRklm176X_rVc/all-rise-space/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.paramountplus.com/movies/million-dollar-american-princesses-meghan-and-harry/C0LpgNwXYeB8txxycdWdR9TjxpJOsdCq',
+        'only_matching': True,
+    }]
+
+    def _extract_video_info(self, content_id, mpx_acc=2198311517):
+        items_data = self._download_json(
+            'https://www.paramountplus.com/apps-api/v2.0/androidtv/video/cid/%s.json' % content_id,
+            content_id, query={'locale': 'en-us', 'at': 'ABCqWNNSwhIqINWIIAG+DFzcFUvF8/vcN6cNyXFFfNzWAIvXuoVgX+fK4naOC7V8MLI='}, headers=self.geo_verification_headers())
+
+        asset_types = {
+            item.get('assetType'): {
+                'format': 'SMIL',
+                'formats': 'MPEG4,M3U',
+            } for item in items_data['itemList']
+        }
+        item = items_data['itemList'][-1]
+        return self._extract_common_video_info(content_id, asset_types, mpx_acc, extra_info={
+            'title': item.get('title'),
+            'series': item.get('seriesTitle'),
+            'season_number': int_or_none(item.get('seasonNum')),
+            'episode_number': int_or_none(item.get('episodeNum')),
+            'duration': int_or_none(item.get('duration')),
+            'thumbnail': url_or_none(item.get('thumbnail')),
+        })
+
+
+class ParamountPlusSeriesIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?paramountplus\.com/shows/(?P<id>[a-zA-Z0-9-_]+)/?(?:[#?]|$)'
+    _TESTS = [{
+        'url': 'https://www.paramountplus.com/shows/drake-josh',
+        'playlist_mincount': 45,
+        'info_dict': {
+            'id': 'drake-josh',
+        }
+    }, {
+        'url': 'https://www.paramountplus.com/shows/hawaii_five_0/',
+        'playlist_mincount': 240,
+        'info_dict': {
+            'id': 'hawaii_five_0',
+        }
+    }, {
+        'url': 'https://www.paramountplus.com/shows/spongebob-squarepants/',
+        'playlist_mincount': 248,
+        'info_dict': {
+            'id': 'spongebob-squarepants',
+        }
+    }]
+    _API_URL = 'https://www.paramountplus.com/shows/{}/xhr/episodes/page/0/size/100000/xs/0/season/0/'
+
+    def _entries(self, show_name):
+        show_json = self._download_json(self._API_URL.format(show_name), video_id=show_name)
+        if show_json.get('success'):
+            for episode in show_json['result']['data']:
+                yield self.url_result(
+                    'https://www.paramountplus.com%s' % episode['url'],
+                    ie=ParamountPlusIE.ie_key(), video_id=episode['content_id'])
+
+    def _real_extract(self, url):
+        show_name = self._match_id(url)
+        return self.playlist_result(self._entries(show_name), playlist_id=show_name)

youtube_dl/extractor/theplatform.py

@@ -20,6 +20,7 @@ from ..utils import (
     float_or_none,
     int_or_none,
     sanitized_Request,
+    try_get,
     unsmuggle_url,
     update_url_query,
     xpath_with_ns,
@@ -34,6 +35,15 @@ _x = lambda p: xpath_with_ns(p, {'smil': default_ns})
 class ThePlatformBaseIE(OnceIE):
     _TP_TLD = 'com'
 
+    @classmethod
+    def _match_valid_url(cls, url):
+        # This does not use has/getattr intentionally - we want to know whether
+        # we have cached the regexp for *this* class, whereas getattr would also
+        # match the superclass
+        if '_VALID_URL_RE' not in cls.__dict__:
+            cls._VALID_URL_RE = re.compile(cls._VALID_URL)
+        return cls._VALID_URL_RE.match(url)
+
     def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
         meta = self._download_xml(
             smil_url, video_id, note=note, query={'format': 'SMIL'},
@@ -238,7 +248,7 @@ class ThePlatformIE(ThePlatformBaseIE, AdobePassIE):
             'countries': smuggled_data.get('geo_countries'),
         })
 
-        mobj = re.match(self._VALID_URL, url)
+        mobj = self._match_valid_url(url)
         provider_id = mobj.group('provider_id')
         video_id = mobj.group('id')
 
@@ -338,6 +348,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
             'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
             'uploader': 'NBCU-NEWS',
         },
+        'expected_warnings': ('Empty metadata',),
     }, {
         'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byGuid=nn_netcast_180306.Copy.01',
         'only_matching': True,
@@ -345,7 +356,11 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
 
     def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}, account_id=None):
         real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, filter_query)
-        entry = self._download_json(real_url, video_id)['entries'][0]
+        entry = self._download_json(real_url, video_id)
+        entry = try_get(entry, lambda x: x['entries'][0], dict)
+        if not entry:
+            self.report_warning('Empty metadata', video_id)
+            return None
         main_smil_url = 'http://link.theplatform.com/s/%s/media/guid/%d/%s' % (provider_id, account_id, entry['guid']) if account_id else entry.get('plmedia$publicUrl')
 
         formats = []
@@ -404,7 +419,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
         return ret
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        mobj = self._match_valid_url(url)
 
         video_id = mobj.group('id')
         provider_id = mobj.group('provider_id')