[cbsnews] fix extraction(closes #9659)(closes #15397)

youtube_dl/extractor/cbsnews.py

@@ -1,40 +1,62 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+import zlib
+
 from .common import InfoExtractor
 from .cbs import CBSIE
+from ..compat import (
+    compat_b64decode,
+    compat_urllib_parse_unquote,
+)
 from ..utils import (
     parse_duration,
 )
 
 
+class CBSNewsEmbedIE(CBSIE):
+    IE_NAME = 'cbsnews:embed'
+    _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/embed/video[^#]*#(?P<id>.+)'
+    _TESTS = [{
+        'url': 'https://www.cbsnews.com/embed/video/?v=1.c9b5b61492913d6660db0b2f03579ef25e86307a#1Vb7b9s2EP5XBAHbT6Gt98PAMKTJ0se6LVjWYWtdGBR1stlIpEBSTtwi%2F%2FvuJNkNhmHdGxgM2NL57vjd6zt%2B8PngdN%2Fyg79qeGvhzN%2FLGrS%2F%2BuBLB531V28%2B%2BO7Qg7%2Fy97r2z3xZ42NW8yLhDbA0S0KWlHnIijwKWJBHZZnHBa8Cgbpdf%2F89NM9Hi9fXifhpr8sr%2FlP848tn%2BTdXycX25zh4cdX%2FvHl6PmmPqnWQv9w8Ed%2B9GjYRim07bFEqdG%2BZVHuwTm65A7bVRrYtR5lAyMox7pigF6W4k%2By91mjspGsJ%2BwVae4%2BsvdnaO1p73HkXs%2FVisUDTGm7R8IcdnOROeq%2B19qT1amhA1VJtPenoTUgrtfKc9m7Rq8dP7nnjwOB7wg7ADdNt7VX64DWAWlKhPtmDEq22g4GF99x6Dk9E8OSsankHXqPNKDxC%2FdK7MLKTircTDgsI3mmj4OBdSq64dy7fd1x577RU1rt4cvMtOaulFYOd%2FLewRWvDO9lIgXFpZSnkZmjbv5SxKTPoQXClFbpsf%2Fhbbpzs0IB3vb8KkyzJQ%2BywOAgCrMpgRrz%2BKk4fvb7kFbR4XJCu0gAdtNO7woCwZTu%2BBUs9bam%2Fds71drVerpeisgrubLjAB4nnOSkWQnfr5W6o1ku5Xpr1MgrCbL0M0vUyDtfLLK15WiYp47xKWSLyjFVpwVmVJSLIoCjSOFkv3W7oKsVliwZJcB9nwXpZ5GEQQwY8jNKqKCBrgjTLeFxgdCIpazojDgnRtn43J6kG7nZ6cAbxh0EeFFk4%2B1u867cY5u4344n%2FxXjCqAjucdTHgLKojNKmSfO8KRsOFY%2FzKEYCKEJBzv90QA9nfm9gL%2BHulaFqUkz9ULUYxl62B3U%2FRVNLA8IhggaPycOoBuwOCESciDQVSSUgiOMsROB%2FhKfwCKOzEk%2B4k6rWd4uuT%2FwTDz7K7t3d3WLO8ISD95jSPQbayBacthbz86XVgxHwhex5zawzgDOmtp%2F3GPcXn0VXHdSS029%2Fj99UC%2FwJUvyKQ%2FzKyixIEVlYJOn4RxxuaH43Ty9fbJ5OObykHH435XAzJTHeOF4hhEUXD8URe%2FQ%2FBT%2BMpf8d5GN02Ox%2FfiGsl7TA7POu1xZ5%2BbTzcAVKMe48mqcC21hkacVEVScM26liVVBnrKkC4CLKyzAvHu0lhEaTKMFwI3a4SN9MsrfYzdBLq2vkwRD1gVviLT8kY9h2CHH6Y%2Bix6609weFtey4ESp60WtyeWMy%2BsmBuhsoKIyuoT%2Bq2R%2FrW5qi3g%2FvzS2j40DoixDP8%2BKP0yUdpXJ4l6Vla%2Bg9vce%2BC4yM5YlUcbA%2F0jLKdpmTwvsdN5z88nAIe08%2F0HgxeG1iv%2B6Hlhjh7uiW0SDzYNI92L401uha3JKYk268UVRzdOzNQvAaJqoXzAc80dAV440NZ1WVVAAMRYQ2KrGJFmDUsq8saWSnjvIj8t78y%2FRa3JRnbHVfyFpfwoDiGpPgjzekyUiKNlU3OMlwuLMmzgvEojllYVE2Z1HhImvsnk%2BuhusTEoB21PAtSFodeFK3iYhXEH9WOG2%2FkOE833sfeG%2Ff5cfHtEFNXgYes0%2FXj7aGivUgJ9XpusCtoNcNYVVnJVrrDo0OmJAutHCpuZul4W9lLcfy7BnuLPT02%2ByXsCTk%2B9zhzswIN04YueNSK%2BPtM0jS88QdLqSLJDTLsuGZJNolm2yO0PXh3UPnz9Ix5bfIAqxPjvETQsDCEiPG4QbqNyhBZISxybLnZYCrW5H3Axp690%2F0BJdXtDZ5ITuM4xj3f4oUHGzc5JeJmZKpp%2FjwKh4wMV%2FV1yx3emLoR0MwbG4K%2F%2BZgVep3PnzXGDHZ6a3i%2Fk%2BJrONDN13%2Bnq6tBTYk4o7cLGhBtqCC4KwacGHpEVuoH5JNro%2FE6JfE6d5RydbiR76k%2BW5wioDHBIjw1euhHjUGRB0y5A97KoaPx6MlL%2BwgboUVtUFRI%2FLemgTpdtF59ii7pab08kuPcfWzs0l%2FRI5takWnFpka0zOgWRtYcuf9aIxZMxlwr6IiGpsb6j2DQUXPl%2FimXI599Ev7fWjoPD78A',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        item = self._parse_json(zlib.decompress(compat_b64decode(
+            compat_urllib_parse_unquote(self._match_id(url))),
+            -zlib.MAX_WBITS), None)['video']['items'][0]
+        return self._extract_video_info(item['mpxRefId'], 'cbsnews')
+
+
 class CBSNewsIE(CBSIE):
     IE_NAME = 'cbsnews'
     IE_DESC = 'CBS News'
-    _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|videos)/(?P<id>[\da-z_-]+)'
+    _VALID_URL = r'https?://(?:www\.)?cbsnews\.com/(?:news|video)/(?P<id>[\da-z_-]+)'
 
     _TESTS = [
         {
             # 60 minutes
             'url': 'http://www.cbsnews.com/news/artificial-intelligence-positioned-to-be-a-game-changer/',
             'info_dict': {
-                'id': '_B6Ga3VJrI4iQNKsir_cdFo9Re_YJHE_',
+                'id': 'Y_nf_aEg6WwO9OLAq0MpKaPgfnBUxfW4',
-                'ext': 'mp4',
+                'ext': 'flv',
-                'title': 'Artificial Intelligence',
+                'title': 'Artificial Intelligence, real-life applications',
-                'description': 'md5:8818145f9974431e0fb58a1b8d69613c',
+                'description': 'md5:a7aaf27f1b4777244de8b0b442289304',
                 'thumbnail': r're:^https?://.*\.jpg$',
-                'duration': 1606,
+                'duration': 317,
                 'uploader': 'CBSI-NEW',
-                'timestamp': 1498431900,
+                'timestamp': 1476046464,
-                'upload_date': '20170625',
+                'upload_date': '20161009',
             },
             'params': {
-                # m3u8 download
+                # rtmp download
                 'skip_download': True,
             },
         },
         {
-            'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
+            'url': 'https://www.cbsnews.com/video/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
             'info_dict': {
                 'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y',
                 'ext': 'mp4',
@@ -60,37 +82,29 @@ class CBSNewsIE(CBSIE):
             # 48 hours
             'url': 'http://www.cbsnews.com/news/maria-ridulph-murder-will-the-nations-oldest-cold-case-to-go-to-trial-ever-get-solved/',
             'info_dict': {
-                'id': 'QpM5BJjBVEAUFi7ydR9LusS69DPLqPJ1',
-                'ext': 'mp4',
                 'title': 'Cold as Ice',
-                'description': 'Can a childhood memory of a friend\'s murder solve a 1957 cold case? "48 Hours" correspondent Erin Moriarty has the latest.',
+                'description': 'Can a childhood memory solve the 1957 murder of 7-year-old Maria Ridulph?',
-                'upload_date': '20170604',
-                'timestamp': 1496538000,
-                'uploader': 'CBSI-NEW',
-            },
-            'params': {
-                'skip_download': True,
             },
+            'playlist_mincount': 7,
         },
     ]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        display_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
 
-        video_info = self._parse_json(self._html_search_regex(
+        entries = []
-            r'(?:<ul class="media-list items" id="media-related-items"[^>]*><li data-video-info|<div id="cbsNewsVideoPlayer" data-video-player-options)=\'({.+?})\'',
+        for embed_url in re.findall(r'<iframe[^>]+data-src="(https?://(?:www\.)?cbsnews\.com/embed/video/[^#]*#[^"]+)"', webpage):
-            webpage, 'video JSON info', default='{}'), video_id, fatal=False)
+            entries.append(self.url_result(embed_url, CBSNewsEmbedIE.ie_key()))
-
+        if entries:
-        if video_info:
+            return self.playlist_result(
-            item = video_info['item'] if 'item' in video_info else video_info
+                entries, playlist_title=self._html_search_meta(['og:title', 'twitter:title'], webpage),
-        else:
+                playlist_description=self._html_search_meta(['og:description', 'twitter:description', 'description'], webpage))
-            state = self._parse_json(self._search_regex(
-                r'data-cbsvideoui-options=(["\'])(?P<json>{.+?})\1', webpage,
-                'playlist JSON info', group='json'), video_id)['state']
-            item = state['playlist'][state['pid']]
 
+        item = self._parse_json(self._html_search_regex(
+            r'CBSNEWS\.defaultPayload\s*=\s*({.+})',
+            webpage, 'video JSON info'), display_id)['items'][0]
         return self._extract_video_info(item['mpxRefId'], 'cbsnews')
 
 

youtube_dl/extractor/extractors.py

@@ -173,6 +173,7 @@ from .cbs import CBSIE
 from .cbslocal import CBSLocalIE
 from .cbsinteractive import CBSInteractiveIE
 from .cbsnews import (
+    CBSNewsEmbedIE,
     CBSNewsIE,
     CBSNewsLiveVideoIE,
 )