[nbc] Use NBC's id and fix _TESTS

[youtube-dl] / youtube_dl / extractor / nbc.py

diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py
index ccdbfb6c9c42c3eb22101eb2f7080b4bda1df00d..18d01f423c620e041883e55839bc866e321cb51d 100644 (file)
--- a/youtube_dl/extractor/nbc.py
+++ b/youtube_dl/extractor/nbc.py
@@ -3,14 +3,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-    compat_HTTPError,
-)
+from ..compat import compat_HTTPError
 from ..utils import (
     ExtractorError,
     find_xpath_attr,
     lowercase_escape,
+    smuggle_url,
     unescapeHTML,
 )
 
@@ -21,32 +19,39 @@ class NBCIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.nbc.com/the-tonight-show/segments/112966',
-            # md5 checksum is not stable
             '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.',
             },
+            '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': 'Only works from US',
+            'skip': '404 Not Found',
         },
         {
             'url': 'http://www.nbc.com/saturday-night-live/video/star-wars-teaser/2832821',
             'info_dict': {
-                'id': '8iUuyzWDdYUZ',
-                'ext': 'flv',
+                'id': '2832821',
+                'ext': 'mp4',
                 'title': 'Star Wars Teaser',
                 'description': 'md5:0b40f9cbde5b671a7ff62fceccc4f442',
             },
+            'params': {
+                # m3u8 download
+                'skip_download': True,
+            },
             'skip': 'Only works from US',
         },
         {
@@ -62,12 +67,17 @@ class NBCIE(InfoExtractor):
         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):
@@ -101,7 +111,7 @@ class NBCSportsVPlayerIE(InfoExtractor):
 
 
 class NBCSportsIE(InfoExtractor):
-    # Does not include https becuase its certificate is invalid
+    # Does not include https because its certificate is invalid
     _VALID_URL = r'http://www\.nbcsports\.com//?(?:[^/]+/)+(?P<id>[0-9a-z-]+)'
 
     _TEST = {
@@ -187,7 +197,7 @@ 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:
@@ -236,3 +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)