Merge tag 'upstream/2015.06.04.1'

[youtubedl] / youtube_dl / extractor / nbc.py

diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py
index aa34665d1669f32ab31a02618c58ef9c4b130fe2..dc2091be0d0c8706b2f3b6d78d88fa22fcb8b6d1 100644 (file)
--- a/youtube_dl/extractor/nbc.py
+++ b/youtube_dl/extractor/nbc.py
@@ -1,40 +1,130 @@
 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

-import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import find_xpath_attr, compat_str
+from ..compat import (
+    compat_str,
+    compat_HTTPError,
+)
+from ..utils import (
+    ExtractorError,
+    find_xpath_attr,
+    lowercase_escape,
+    unescapeHTML,
+)

 
 
 class NBCIE(InfoExtractor):
 
 
 class NBCIE(InfoExtractor):

-    _VALID_URL = r'http://www\.nbc\.com/[^/]+/video/[^/]+/(?P<id>n?\d+)'
+    _VALID_URL = r'https?://www\.nbc\.com/(?:[^/]+/)+(?P<id>n?\d+)'

 
 

-    _TEST = {
-        'url': 'http://www.nbc.com/chicago-fire/video/i-am-a-firefighter/2734188',
-        'md5': '54d0fbc33e0b853a65d7b4de5c06d64e',
-        'info_dict': {
-            'id': 'u1RInQZRN7QJ',
-            '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.',
+    _TESTS = [
+        {
+            'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
+            # md5 checksum is not stable
+            'info_dict': {
+                'id': 'c9xnCo0YPOPH',
+                'ext': 'flv',
+                '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.',
+            },

         },
         },

-    }
+        {
+            'url': 'http://www.nbc.com/the-tonight-show/episodes/176',
+            'info_dict': {
+                'id': 'XwU9KZkp98TH',
+                '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': 'Only works from US',
+        },
+        {
+            'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
+            'info_dict': {
+                'id': '8iUuyzWDdYUZ',
+                'ext': 'flv',
+                'title': 'Star Wars Teaser',
+                'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
+            },
+            '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',
+            'skip': 'Expired'
+        }
+    ]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)

         webpage = self._download_webpage(url, video_id)
         webpage = self._download_webpage(url, video_id)

-        theplatform_url = self._search_regex('class="video-player video-player-full" data-mpx-url="(.*?)"', webpage, 'theplatform url')
+        theplatform_url = unescapeHTML(lowercase_escape(self._html_search_regex(
+            [
+                r'(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
+                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)
 
 
         if theplatform_url.startswith('//'):
             theplatform_url = 'http:' + theplatform_url
         return self.url_result(theplatform_url)
 
 

+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 becuase 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):
 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+)|
+        (?:feature|nightly-news)/[^/]+/(?P<title>.+))

         '''
 
     _TESTS = [
         '''
 
     _TESTS = [


@@ -53,12 +143,32 @@ class NBCNewsIE(InfoExtractor):
             'md5': 'b2421750c9f260783721d898f4c42063',
             'info_dict': {
                 'id': 'I1wpAI_zmhsQ',
             'md5': 'b2421750c9f260783721d898f4c42063',
             'info_dict': {
                 'id': 'I1wpAI_zmhsQ',

-                'ext': 'flv',
+                'ext': 'mp4',

                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
             },
             'add_ie': ['ThePlatform'],
         },
                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
             },
             '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',
+            },
+        },

     ]
 
     def _real_extract(self, url):
     ]
 
     def _real_extract(self, url):


@@ -77,19 +187,44 @@ class NBCNewsIE(InfoExtractor):
                 'thumbnail': find_xpath_attr(info, 'media', 'type', 'thumbnail').text,
             }
         else:
                 '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(
             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']
             info = bootstrap['results'][0]['video']

-            playlist_url = info['fallbackPlaylistUrl'] + '?form=MPXNBCNewsAPI'

             mpxid = info['mpxId']
             mpxid = info['mpxId']

-            all_videos = self._download_json(playlist_url, title)['videos']
-            # The response contains additional videos
-            info = next(v for v in all_videos if v['mpxId'] == 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')

 
             return {
                 '_type': 'url',
 
             return {
                 '_type': 'url',