Merge tag 'upstream/2015.02.28'

[youtubedl] / youtube_dl / extractor / nbc.py

diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py
index 690c46b6a57be11edf36899b959318af5e482119..3645d3033f74ae174e3eaa85ad55bbe677d9daba 100644 (file)
--- a/youtube_dl/extractor/nbc.py
+++ b/youtube_dl/extractor/nbc.py
@@ -1,11 +1,11 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
 from ..compat import (
     compat_str,
+    compat_HTTPError,
 )
 from ..utils import (
     ExtractorError,
@@ -18,13 +18,13 @@ class NBCIE(InfoExtractor):
 
     _TESTS = [
         {
-            'url': 'http://www.nbc.com/chicago-fire/video/i-am-a-firefighter/2734188',
+            'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
             # md5 checksum is not stable
             'info_dict': {
-                'id': 'bTmnLCvIbaaH',
+                'id': 'c9xnCo0YPOPH',
                 '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.',
+                '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.',
             },
         },
         {
@@ -51,9 +51,9 @@ class NBCIE(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 = [
@@ -78,6 +78,26 @@ 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',
+            },
+        },
     ]
 
     def _real_extract(self, url):
@@ -96,13 +116,13 @@ class NBCNewsIE(InfoExtractor):
                 '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 +135,19 @@ 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