[sky] add support for Sports News articles and Brighcove videos(closes #13054)

youtube_dl/extractor/extractors.py

@@ -1052,6 +1052,7 @@ from .skynewsarabia import (
 from .sky import (
     SkyNewsIE,
     SkySportsIE,
+    SkySportsNewsIE,
 )
 from .slideshare import SlideshareIE
 from .slideslive import SlidesLiveIE

youtube_dl/extractor/sky.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     extract_attributes,
@@ -11,36 +13,59 @@ from ..utils import (
 
 
 class SkyBaseIE(InfoExtractor):
-    def _real_extract(self, url):
+    BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s'
-        video_id = self._match_id(url)
+    _SDC_EL_REGEX = r'(?s)(<div[^>]+data-(?:component-name|fn)="sdc-(?:articl|sit)e-video"[^>]*>)'
-        webpage = self._download_webpage(url, video_id)
-        video_data = extract_attributes(self._search_regex(
-            r'(<div.+?class="[^"]*sdc-article-video__media-ooyala[^"]*"[^>]+>)',
-            webpage, 'video data'))
 
-        video_url = 'ooyala:%s' % video_data['data-video-id']
+    def _process_ooyala_element(self, webpage, sdc_el, url):
-        if video_data.get('data-token-required') == 'true':
+        sdc = extract_attributes(sdc_el)
-            token_fetch_options = self._parse_json(video_data.get(
+        provider = sdc.get('data-provider')
-                'data-token-fetch-options', '{}'), video_id, fatal=False) or {}
+        if provider == 'ooyala':
-            token_fetch_url = token_fetch_options.get('url')
+            video_id = sdc['data-sdc-video-id']
-            if token_fetch_url:
+            video_url = 'ooyala:%s' % video_id
-                embed_token = self._download_webpage(urljoin(
+            ie_key = 'Ooyala'
-                    url, token_fetch_url), video_id, fatal=False)
+            ooyala_el = self._search_regex(
-                if embed_token:
+                r'(<div[^>]+class="[^"]*\bsdc-article-video__media-ooyala\b[^"]*"[^>]+data-video-id="%s"[^>]*>)' % video_id,
-                    video_url = smuggle_url(
+                webpage, 'video data', fatal=False)
-                        video_url, {'embed_token': embed_token.strip('"')})
+            if ooyala_el:
+                ooyala_attrs = extract_attributes(ooyala_el) or {}
+                if ooyala_attrs.get('data-token-required') == 'true':
+                    token_fetch_url = (self._parse_json(ooyala_attrs.get(
+                        'data-token-fetch-options', '{}'),
+                        video_id, fatal=False) or {}).get('url')
+                    if token_fetch_url:
+                        embed_token = self._download_json(urljoin(
+                            url, token_fetch_url), video_id, fatal=False)
+                        if embed_token:
+                            video_url = smuggle_url(
+                                video_url, {'embed_token': embed_token})
+        elif provider == 'brightcove':
+            video_id = sdc['data-video-id']
+            account_id = sdc.get('data-account-id') or '6058004172001'
+            player_id = sdc.get('data-player-id') or 'RC9PQUaJ6'
+            video_url = self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id)
+            ie_key = 'BrightcoveNew'
 
         return {
             '_type': 'url_transparent',
             'id': video_id,
             'url': video_url,
+            'ie_key': ie_key,
+        }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        info = self._process_ooyala_element(webpage, self._search_regex(
+            self._SDC_EL_REGEX, webpage, 'sdc element'), url)
+        info.update({
             'title': self._og_search_title(webpage),
             'description': strip_or_none(self._og_search_description(webpage)),
-            'ie_key': 'Ooyala',
+        })
-        }
+        return info
 
 
 class SkySportsIE(SkyBaseIE):
+    IE_NAME = 'sky:sports'
     _VALID_URL = r'https?://(?:www\.)?skysports\.com/watch/video/([^/]+/)*(?P<id>[0-9]+)'
     _TESTS = [{
         'url': 'http://www.skysports.com/watch/video/10328419/bale-its-our-time-to-shine',
@@ -62,15 +87,45 @@ class SkySportsIE(SkyBaseIE):
 
 
 class SkyNewsIE(SkyBaseIE):
+    IE_NAME = 'sky:news'
     _VALID_URL = r'https?://news\.sky\.com/video/[0-9a-z-]+-(?P<id>[0-9]+)'
     _TEST = {
         'url': 'https://news.sky.com/video/russian-plane-inspected-after-deadly-fire-11712962',
-        'md5': 'd6327e581473cea9976a3236ded370cd',
+        'md5': '411e8893fd216c75eaf7e4c65d364115',
         'info_dict': {
-            'id': '1ua21xaDE6lCtZDmbYfl8kwsKLooJbNM',
+            'id': 'ref:1ua21xaDE6lCtZDmbYfl8kwsKLooJbNM',
             'ext': 'mp4',
             'title': 'Russian plane inspected after deadly fire',
             'description': 'The Russian Investigative Committee has released video of the wreckage of a passenger plane which caught fire near Moscow.',
+            'uploader_id': '6058004172001',
+            'timestamp': 1567112345,
+            'upload_date': '20190829',
         },
-        'add_ie': ['Ooyala'],
+        'add_ie': ['BrightcoveNew'],
     }
+
+
+class SkySportsNewsIE(SkyBaseIE):
+    IE_NAME = 'sky:sports:news'
+    _VALID_URL = r'https?://(?:www\.)?skysports\.com/([^/]+/)*news/\d+/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://www.skysports.com/golf/news/12176/10871916/dustin-johnson-ready-to-conquer-players-championship-at-tpc-sawgrass',
+        'info_dict': {
+            'id': '10871916',
+            'title': 'Dustin Johnson ready to conquer Players Championship at TPC Sawgrass',
+            'description': 'Dustin Johnson is confident he can continue his dominant form in 2017 by adding the Players Championship to his list of victories.',
+        },
+        'playlist_count': 2,
+    }
+
+    def _real_extract(self, url):
+        article_id = self._match_id(url)
+        webpage = self._download_webpage(url, article_id)
+
+        entries = []
+        for sdc_el in re.findall(self._SDC_EL_REGEX, webpage):
+            entries.append(self._process_ooyala_element(webpage, sdc_el, url))
+
+        return self.playlist_result(
+            entries, article_id, self._og_search_title(webpage),
+            self._html_search_meta(['og:description', 'description'], webpage))