]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/nbc.py
Merge tag 'upstream/2016.02.22'
[youtubedl] / youtube_dl / extractor / nbc.py
index 690c46b6a57be11edf36899b959318af5e482119..2202cfa334765653b44e180306d8e40e0be2049d 100644 (file)
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-)
+from ..compat import compat_HTTPError
 from ..utils import (
     ExtractorError,
     find_xpath_attr,
+    lowercase_escape,
+    smuggle_url,
+    unescapeHTML,
 )
 
 
 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/chicago-fire/video/i-am-a-firefighter/2734188',
-            # md5 checksum is not stable
+            'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
             'info_dict': {
-                'id': 'bTmnLCvIbaaH',
-                'ext': 'flv',
-                'title': 'I Am a Firefighter',
-                'description': 'An emergency puts Dawson\'sf irefighter skills to the ultimate test in this four-part digital series.',
+                '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.',
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
             },
         },
         {
             '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.',
             },
+            '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',
+            },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
             '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,
+        }
     ]
 
     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
-        return self.url_result(theplatform_url)
+        return {
+            '_type': 'url_transparent',
+            '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',
+        }
+    }, {
+        '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'http://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',
+        }
+    }
+
+    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 NBCNewsIE(InfoExtractor):
-    _VALID_URL = r'''(?x)https?://www\.nbcnews\.com/
-        ((video/.+?/(?P<id>\d+))|
-        (feature/[^/]+/(?P<title>.+)))
+    _VALID_URL = r'''(?x)https?://(?:www\.)?nbcnews\.com/
+        (?:video/.+?/(?P<id>\d+)|
+        (?:watch|feature|nightly-news)/[^/]+/(?P<title>.+))
         '''
 
     _TESTS = [
@@ -78,6 +159,30 @@ class NBCNewsIE(InfoExtractor):
             },
             'add_ie': ['ThePlatform'],
         },
+        {
+            'url': 'http://www.nbcnews.com/feature/dateline-full-episodes/full-episode-family-business-n285156',
+            'md5': 'fdbf39ab73a72df5896b6234ff98518a',
+            'info_dict': {
+                'id': 'Wjf9EDR3A_60',
+                'ext': 'mp4',
+                'title': 'FULL EPISODE: Family Business',
+                'description': 'md5:757988edbaae9d7be1d585eb5d55cc04',
+            },
+        },
+        {
+            'url': 'http://www.nbcnews.com/nightly-news/video/nightly-news-with-brian-williams-full-broadcast-february-4-394064451844',
+            'md5': 'b5dda8cddd8650baa0dcb616dd2cf60d',
+            'info_dict': {
+                'id': 'sekXqyTVnmN3',
+                'ext': 'mp4',
+                'title': 'Nightly News with Brian Williams Full Broadcast (February 4)',
+                'description': 'md5:1c10c1eccbe84a26e5debb4381e2d3c5',
+            },
+        },
+        {
+            'url': 'http://www.nbcnews.com/watch/dateline/full-episode--deadly-betrayal-386250819952',
+            'only_matching': True,
+        },
     ]
 
     def _real_extract(self, url):
@@ -92,17 +197,17 @@ class NBCNewsIE(InfoExtractor):
                 '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" pages use theplatform.com
+            # "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 bootstrapJson = ({.+})\s*$', webpage, 'bootstrap json',
-                flags=re.MULTILINE)
-            bootstrap = json.loads(bootstrap_json)
+                r'var\s+(?:bootstrapJson|playlistData)\s*=\s*({.+});?\s*$',
+                webpage, 'bootstrap json', flags=re.MULTILINE)
+            bootstrap = self._parse_json(bootstrap_json, video_id)
             info = bootstrap['results'][0]['video']
             mpxid = info['mpxId']
 
@@ -115,10 +220,19 @@ class NBCNewsIE(InfoExtractor):
                 if not base_url:
                     continue
                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
-                all_videos = self._download_json(playlist_url, title)['videos']
 
                 try:
-                    info = next(v for v in all_videos if v['mpxId'] == mpxid)
+                    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
@@ -132,3 +246,28 @@ class NBCNewsIE(InfoExtractor):
                 'url': info['videoAssets'][-1]['publicUrl'],
                 'ie_key': 'ThePlatform',
             }
+
+
+class MSNBCIE(InfoExtractor):
+    # https URLs redirect to corresponding http ones
+    _VALID_URL = r'http://www\.msnbc\.com/[^/]+/watch/(?P<id>[^/]+)'
+    _TEST = {
+        'url': 'http://www.msnbc.com/all-in-with-chris-hayes/watch/the-chaotic-gop-immigration-vote-314487875924',
+        'md5': '6d236bf4f3dddc226633ce6e2c3f814d',
+        'info_dict': {
+            'id': 'n_hayes_Aimm_140801_272214',
+            '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',
+            'categories': ['MSNBC/Topics/Franchise/Best of last night', 'MSNBC/Topics/General/Congress'],
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        embed_url = self._html_search_meta('embedURL', webpage)
+        return self.url_result(embed_url)