Merge pull request #2281 from matthewfranglen/master

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

diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py
index 127fbeb4edde7fdcc7d76ac95ab67b09cb7823a7..4521451ac933baabce786c76813e3e8407298b65 100644 (file)
--- a/youtube_dl/extractor/mtv.py
+++ b/youtube_dl/extractor/mtv.py
@@ -6,11 +6,13 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     ExtractorError,
+    find_xpath_attr,
     fix_xml_ampersands,
     url_basename,
     RegexNotFoundError,
 )
 
+
 def _media_xml_tag(tag):
     return '{http://search.yahoo.com/mrss/}%s' % tag
 
@@ -65,6 +67,7 @@ class MTVServicesInfoExtractor(InfoExtractor):
         mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
         if 'acceptMethods' not in mediagen_url:
             mediagen_url += '&acceptMethods=fms'
+
         mediagen_doc = self._download_xml(mediagen_url, video_id,
             'Downloading video urls')
 
@@ -74,8 +77,22 @@ class MTVServicesInfoExtractor(InfoExtractor):
         else:
             description = None
 
+        title_el = None
+        if title_el is None:
+            title_el = find_xpath_attr(
+                itemdoc, './/{http://search.yahoo.com/mrss/}category',
+                'scheme', 'urn:mtvn:video_title')
+        if title_el is None:
+            title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
+        if title_el is None:
+            title_el = itemdoc.find('.//title')
+        title = title_el.text
+        if title is None:
+            raise ExtractorError('Could not find video title')
+        title = title.strip()
+
         return {
-            'title': itemdoc.find('title').text,
+            'title': title,
             'formats': self._extract_video_formats(mediagen_doc),
             'id': video_id,
             'thumbnail': self._get_thumbnail_url(uri, itemdoc),
@@ -95,11 +112,16 @@ class MTVServicesInfoExtractor(InfoExtractor):
         title = url_basename(url)
         webpage = self._download_webpage(url, title)
         try:
-            # the url is in the format http://media.mtvnservices.com/fb/{mgid}.swf
-            fb_url = self._og_search_video_url(webpage)
-            mgid = url_basename(fb_url).rpartition('.')[0]
+            # the url can be http://media.mtvnservices.com/fb/{mgid}.swf
+            # or http://media.mtvnservices.com/{mgid}
+            og_url = self._og_search_video_url(webpage)
+            mgid = url_basename(og_url)
+            if mgid.endswith('.swf'):
+                mgid = mgid[:-4]
         except RegexNotFoundError:
-            mgid = self._search_regex(r'data-mgid="(.*?)"', webpage, u'mgid')
+            mgid = self._search_regex(
+                [r'data-mgid="(.*?)"', r'swfobject.embedSWF\(".*?(mgid:.*?)"'],
+                webpage, u'mgid')
         return self._get_videos_info(mgid)
 
 
@@ -164,7 +186,7 @@ class MTVIggyIE(MTVServicesInfoExtractor):
         'info_dict': {
             'id': '984696',
             'ext': 'mp4',
-            'title': 'Short',
+            'title': 'Arcade Fire: Behind the Scenes at the Biggest Music Experiment Yet',
         }
     }
     _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'