]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/nbc.py
Merge tag 'upstream/2016.06.25'
[youtubedl] / youtube_dl / extractor / nbc.py
index 3645d3033f74ae174e3eaa85ad55bbe677d9daba..f694e210b1dadceb030cb24f6498abe30de5b976 100644 (file)
@@ -3,57 +3,194 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-    compat_HTTPError,
-)
+from .theplatform import ThePlatformIE
 from ..utils import (
 from ..utils import (
-    ExtractorError,
     find_xpath_attr,
     find_xpath_attr,
+    lowercase_escape,
+    smuggle_url,
+    unescapeHTML,
 )
 
 
 class NBCIE(InfoExtractor):
 )
 
 
 class NBCIE(InfoExtractor):
-    _VALID_URL = r'http://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
+    _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'
 
     _TESTS = [
         {
             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
 
     _TESTS = [
         {
             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
-            # md5 checksum is not stable
             'info_dict': {
             'info_dict': {
-                'id': 'c9xnCo0YPOPH',
-                'ext': 'flv',
+                'id': '112966',
+                'ext': 'mp4',
                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
                 'title': 'Jimmy Fallon Surprises Fans at Ben & Jerry\'s',
                 'description': 'Jimmy gives out free scoops of his new "Tonight Dough" ice cream flavor by surprising customers at the Ben & Jerry\'s scoop shop.',
+                'timestamp': 1424246400,
+                'upload_date': '20150218',
+                'uploader': 'NBCU-COM',
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
             },
         },
         {
             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
             'info_dict': {
             },
         },
         {
             'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
             'info_dict': {
-                'id': 'XwU9KZkp98TH',
+                'id': '176',
                 'ext': 'flv',
                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
             },
                 'ext': 'flv',
                 'title': 'Ricky Gervais, Steven Van Zandt, ILoveMakonnen',
                 'description': 'A brand new episode of The Tonight Show welcomes Ricky Gervais, Steven Van Zandt and ILoveMakonnen.',
             },
+            'skip': '404 Not Found',
+        },
+        {
+            'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
+            'info_dict': {
+                'id': '2832821',
+                'ext': 'mp4',
+                'title': 'Star Wars Teaser',
+                'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
+                'timestamp': 1417852800,
+                'upload_date': '20141206',
+                'uploader': 'NBCU-COM',
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
             'skip': 'Only works from US',
         },
             'skip': 'Only works from US',
         },
+        {
+            # This video has expired but with an escaped embedURL
+            'url': 'http://www.nbc.com/parenthood/episode-guide/season-5/just-like-at-home/515',
+            'only_matching': True,
+        },
+        {
+            # HLS streams requires the 'hdnea3' cookie
+            'url': 'http://www.nbc.com/Kings/video/goliath/n1806',
+            'info_dict': {
+                'id': 'n1806',
+                'ext': 'mp4',
+                'title': 'Goliath',
+                'description': 'When an unknown soldier saves the life of the King\'s son in battle, he\'s thrust into the limelight and politics of the kingdom.',
+                'timestamp': 1237100400,
+                'upload_date': '20090315',
+                'uploader': 'NBCU-COM',
+            },
+            'params': {
+                'skip_download': True,
+            },
+            'skip': 'Only works from US',
+        }
     ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
     ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        theplatform_url = self._search_regex(
-            '(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
-            webpage, 'theplatform url').replace('_no_endcard', '')
+        theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
+            [
+                r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
+                r'<iframe[^>]+src="((?:https?:)?//player\.theplatform\.com/[^"]+)"',
+                r'"embedURL"\s*:\s*"([^"]+)"'
+            ],
+            webpage, 'theplatform url').replace('_no_endcard', '').replace('\\/', '/')))
         if theplatform_url.startswith('//'):
             theplatform_url = 'http:' + theplatform_url
         if theplatform_url.startswith('//'):
             theplatform_url = 'http:' + theplatform_url
-        return self.url_result(theplatform_url)
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'ThePlatform',
+            'url': smuggle_url(theplatform_url, {'source_url': url}),
+            'id': video_id,
+        }
+
+
+class NBCSportsVPlayerIE(InfoExtractor):
+    _VALID_URL = r'https?://vplayer\.nbcsports\.com/(?:[^/]+/)+(?P<id>[0-9a-zA-Z_]+)'
+
+    _TESTS = [{
+        'url': 'https://vplayer.nbcsports.com/p/BxmELC/nbcsports_share/select/9CsDKds0kvHI',
+        'info_dict': {
+            'id': '9CsDKds0kvHI',
+            'ext': 'flv',
+            'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
+            'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
+            'timestamp': 1426270238,
+            'upload_date': '20150313',
+            'uploader': 'NBCU-SPORTS',
+        }
+    }, {
+        'url': 'http://vplayer.nbcsports.com/p/BxmELC/nbc_embedshare/select/_hqLjQ95yx8Z',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_url(webpage):
+        iframe_m = re.search(
+            r'<iframe[^>]+src="(?P<url>https?://vplayer\.nbcsports\.com/[^"]+)"', webpage)
+        if iframe_m:
+            return iframe_m.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        theplatform_url = self._og_search_video_url(webpage)
+        return self.url_result(theplatform_url, 'ThePlatform')
+
+
+class NBCSportsIE(InfoExtractor):
+    # Does not include https because its certificate is invalid
+    _VALID_URL = r'https?://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
+
+    _TEST = {
+        'url': 'http://www.nbcsports.com//college-basketball/ncaab/tom-izzo-michigan-st-has-so-much-respect-duke',
+        'info_dict': {
+            'id': 'PHJSaFWbrTY9',
+            'ext': 'flv',
+            'title': 'Tom Izzo, Michigan St. has \'so much respect\' for Duke',
+            'description': 'md5:ecb459c9d59e0766ac9c7d5d0eda8113',
+            'uploader': 'NBCU-SPORTS',
+            'upload_date': '20150330',
+            'timestamp': 1427726529,
+        }
+    }
 
 
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        return self.url_result(
+            NBCSportsVPlayerIE._extract_url(webpage), 'NBCSportsVPlayer')
+
+
+class CSNNEIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.csnne\.com/video/(?P<id>[0-9a-z-]+)'
+
+    _TEST = {
+        'url': 'http://www.csnne.com/video/snc-evening-update-wright-named-red-sox-no-5-starter',
+        'info_dict': {
+            'id': 'yvBLLUgQ8WU0',
+            'ext': 'mp4',
+            'title': 'SNC evening update: Wright named Red Sox\' No. 5 starter.',
+            'description': 'md5:1753cfee40d9352b19b4c9b3e589b9e3',
+            'timestamp': 1459369979,
+            'upload_date': '20160330',
+            'uploader': 'NBCU-SPORTS',
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'ThePlatform',
+            'url': self._html_search_meta('twitter:player:stream', webpage),
+            'display_id': display_id,
+        }
 
 
-class NBCNewsIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
+
+class NBCNewsIE(ThePlatformIE):
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(?:nbcnews|today|msnbc)\.com/
         (?:video/.+?/(?P<id>\d+)|
         (?:video/.+?/(?P<id>\d+)|
-        (?:feature|nightly-news)/[^/]+/(?P<title>.+))
+        ([^/]+/)*(?:.*-)?(?P<mpx_id>[^/?]+))
         '''
 
     _TESTS = [
         '''
 
     _TESTS = [
@@ -68,36 +205,92 @@ class NBCNewsIE(InfoExtractor):
             },
         },
         {
             },
         },
         {
-            'url': 'http://www.nbcnews.com/feature/edward-snowden-interview/how-twitter-reacted-snowden-interview-n117236',
-            'md5': 'b2421750c9f260783721d898f4c42063',
+            'url': 'http://www.nbcnews.com/watch/nbcnews-com/how-twitter-reacted-to-the-snowden-interview-269389891880',
+            'md5': 'af1adfa51312291a017720403826bb64',
             'info_dict': {
             'info_dict': {
-                'id': 'I1wpAI_zmhsQ',
+                'id': '269389891880',
                 'ext': 'mp4',
                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
                 'ext': 'mp4',
                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
+                'uploader': 'NBCU-NEWS',
+                'timestamp': 1401363060,
+                'upload_date': '20140529',
             },
             },
-            'add_ie': ['ThePlatform'],
         },
         {
             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
             'info_dict': {
         },
         {
             'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
             'md5': 'fdbf39ab73a72df5896b6234ff98518a',
             'info_dict': {
-                'id': 'Wjf9EDR3A_60',
+                'id': '529953347624',
                 'ext': 'mp4',
                 'title': 'FULL EPISODE: Family Business',
                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
             },
                 'ext': 'mp4',
                 'title': 'FULL EPISODE: Family Business',
                 'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
             },
+            'skip': 'This page is unavailable.',
         },
         {
             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
         },
         {
             'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
-            'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
+            'md5': '73135a2e0ef819107bbb55a5a9b2a802',
             'info_dict': {
             'info_dict': {
-                'id': 'sekXqyTVnmN3',
+                'id': '394064451844',
                 'ext': 'mp4',
                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
                 'ext': 'mp4',
                 'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
                 'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
+                'timestamp': 1423104900,
+                'uploader': 'NBCU-NEWS',
+                'upload_date': '20150205',
             },
         },
             },
         },
+        {
+            'url': 'http://www.nbcnews.com/business/autos/volkswagen-11-million-vehicles-could-have-suspect-software-emissions-scandal-n431456',
+            'md5': 'a49e173825e5fcd15c13fc297fced39d',
+            'info_dict': {
+                'id': '529953347624',
+                'ext': 'mp4',
+                'title': 'Volkswagen U.S. Chief:\xa0 We Have Totally Screwed Up',
+                'description': 'md5:c8be487b2d80ff0594c005add88d8351',
+                'upload_date': '20150922',
+                'timestamp': 1442917800,
+                'uploader': 'NBCU-NEWS',
+            },
+        },
+        {
+            'url': 'http://www.today.com/video/see-the-aurora-borealis-from-space-in-stunning-new-nasa-video-669831235788',
+            'md5': '118d7ca3f0bea6534f119c68ef539f71',
+            'info_dict': {
+                'id': '669831235788',
+                'ext': 'mp4',
+                'title': 'See the aurora borealis from space in stunning new NASA video',
+                'description': 'md5:74752b7358afb99939c5f8bb2d1d04b1',
+                'upload_date': '20160420',
+                'timestamp': 1461152093,
+                'uploader': 'NBCU-NEWS',
+            },
+        },
+        {
+            'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
+            'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
+            'info_dict': {
+                'id': '314487875924',
+                'ext': 'mp4',
+                'title': 'The chaotic GOP immigration vote',
+                'description': 'The Republican House votes on a border bill that has no chance of getting through the Senate or signed by the President and is drawing criticism from all sides.',
+                'thumbnail': 're:^https?://.*\.jpg$',
+                'timestamp': 1406937606,
+                'upload_date': '20140802',
+                'uploader': 'NBCU-NEWS',
+                'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
+            },
+        },
+        {
+            'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
+            'only_matching': True,
+        },
+        {
+            # From http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html
+            'url': 'http://www.nbcnews.com/widget/video-embed/701714499682',
+            'only_matching': True,
+        },
     ]
 
     def _real_extract(self, url):
     ]
 
     def _real_extract(self, url):
@@ -112,52 +305,33 @@ class NBCNewsIE(InfoExtractor):
                 'title': info.find('headline').text,
                 'ext': 'flv',
                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
                 'title': info.find('headline').text,
                 'ext': 'flv',
                 'url': find_xpath_attr(info, 'media', 'type', 'flashVideo').text,
-                'description': compat_str(info.find('caption').text),
+                'description': info.find('caption').text,
                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
             }
         else:
             # "feature" and "nightly-news" pages use theplatform.com
                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
             }
         else:
             # "feature" and "nightly-news" pages use theplatform.com
-            title = mobj.group('title')
-            webpage = self._download_webpage(url, title)
-            bootstrap_json = self._search_regex(
-                r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
-                webpage, 'bootstrap json', flags=re.MULTILINE)
-            bootstrap = self._parse_json(bootstrap_json, video_id)
-            info = bootstrap['results'][0]['video']
-            mpxid = info['mpxId']
-
-            base_urls = [
-                info['fallbackPlaylistUrl'],
-                info['associatedPlaylistUrl'],
-            ]
-
-            for base_url in base_urls:
-                if not base_url:
-                    continue
-                playlist_url = base_url + '?form=MPXNBCNewsAPI'
-
-                try:
-                    all_videos = self._download_json(playlist_url, title)
-                except ExtractorError as ee:
-                    if isinstance(ee.cause, compat_HTTPError):
-                        continue
-                    raise
-
-                if not all_videos or 'videos' not in all_videos:
-                    continue
-
-                try:
-                    info = next(v for v in all_videos['videos'] if v['mpxId'] == mpxid)
-                    break
-                except StopIteration:
-                    continue
-
-            if info is None:
-                raise ExtractorError('Could not find video in playlists')
+            video_id = mobj.group('mpx_id')
+            if not video_id.isdigit():
+                webpage = self._download_webpage(url, video_id)
+                info = None
+                bootstrap_json = self._search_regex(
+                    [r'(?m)(?:var\s+(?:bootstrapJson|playlistData)|NEWS\.videoObj)\s*=\s*({.+});?\s*$',
+                     r'videoObj\s*:\s*({.+})', r'data-video="([^"]+)"'],
+                    webpage, 'bootstrap json', default=None)
+                bootstrap = self._parse_json(
+                    bootstrap_json, video_id, transform_source=unescapeHTML)
+                if 'results' in bootstrap:
+                    info = bootstrap['results'][0]['video']
+                elif 'video' in bootstrap:
+                    info = bootstrap['video']
+                else:
+                    info = bootstrap
+                video_id = info['mpxId']
 
             return {
 
             return {
-                '_type': 'url',
-                # We get the best quality video
-                'url': info['videoAssets'][-1]['publicUrl'],
-                'ie_key': 'ThePlatform',
+                '_type': 'url_transparent',
+                'id': video_id,
+                # http://feed.theplatform.com/f/2E2eJC/nbcnews also works
+                'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byId=%s' % video_id,
+                'ie_key': 'ThePlatformFeed',
             }
             }