Merge tag 'upstream/2015.01.16'

[youtubedl] / youtube_dl / extractor / nbc.py

diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py
index d2e4acbada5b99c5c3eac4fe6b966ce77dab1ef9..690c46b6a57be11edf36899b959318af5e482119 100644 (file)
--- a/youtube_dl/extractor/nbc.py
+++ b/youtube_dl/extractor/nbc.py
@@ -4,32 +4,47 @@ import re
 import json
 
 from .common import InfoExtractor
 import json
 
 from .common import InfoExtractor

-from ..utils import (
+from ..compat import (

     compat_str,
     compat_str,

+)
+from ..utils import (

     ExtractorError,
     find_xpath_attr,
 )
 
 
 class NBCIE(InfoExtractor):
     ExtractorError,
     find_xpath_attr,
 )
 
 
 class NBCIE(InfoExtractor):

-    _VALID_URL = r'http://www\.nbc\.com/[^/]+/video/[^/]+/(?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.',
+    _VALID_URL = r'http://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
+            '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.',
+            },

         },
         },

-    }
+        {
+            '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',
+        },
+    ]

 
     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 = self._search_regex(
+            '(?:class="video-player video-player-full" data-mpx-url|class="player" src)="(.*?)"',
+            webpage, 'theplatform url').replace('_no_endcard', '')

         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)


@@ -57,7 +72,7 @@ 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',
             },
                 'title': 'How Twitter Reacted To The Snowden Interview',
                 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64',
             },


@@ -97,6 +112,8 @@ class NBCNewsIE(InfoExtractor):
             ]
 
             for base_url in base_urls:
             ]
 
             for base_url in base_urls:

+                if not base_url:
+                    continue

                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
                 all_videos = self._download_json(playlist_url, title)['videos']
 
                 playlist_url = base_url + '?form=MPXNBCNewsAPI'
                 all_videos = self._download_json(playlist_url, title)['videos']