Merge branch 'master' of github.com:rg3/youtube-dl
authorPhilipp Hagemeister <phihag@phihag.de>
Tue, 21 Jan 2014 23:21:27 +0000 (00:21 +0100)
committerPhilipp Hagemeister <phihag@phihag.de>
Tue, 21 Jan 2014 23:21:27 +0000 (00:21 +0100)
Conflicts:
youtube_dl/extractor/mtv.py

1  2 
youtube_dl/extractor/mtv.py

index 32cfa36322bb23335b3baf32f138d04937f679c3,127fbeb4edde7fdcc7d76ac95ab67b09cb7823a7..af889a8afeb9f50169ceb3f59d9dee83c27fee65
@@@ -1,17 -1,16 +1,18 @@@
  from __future__ import unicode_literals
  
  import re
- import xml.etree.ElementTree
  
  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
  
@@@ -38,10 -37,9 +39,9 @@@ class MTVServicesInfoExtractor(InfoExtr
          else:
              return thumb_node.attrib['url']
  
-     def _extract_video_formats(self, metadataXml):
-         if '/error_country_block.swf' in metadataXml:
+     def _extract_video_formats(self, mdoc):
+         if re.match(r'.*/error_country_block\.swf$', mdoc.find('.//src').text) is not None:
              raise ExtractorError('This video is not available from your country.', expected=True)
-         mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8'))
  
          formats = []
          for rendition in mdoc.findall('.//rendition'):
@@@ -67,8 -65,8 +67,9 @@@
          mediagen_url = re.sub(r'&[^=]*?={.*?}(?=(&|$))', '', mediagen_url)
          if 'acceptMethods' not in mediagen_url:
              mediagen_url += '&acceptMethods=fms'
-         mediagen_page = self._download_webpage(mediagen_url, video_id,
-                                                'Downloading video urls')
++
+         mediagen_doc = self._download_xml(mediagen_url, video_id,
+             'Downloading video urls')
  
          description_node = itemdoc.find('description')
          if description_node is not None:
          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')
 +
          return {
 -            'title': itemdoc.find('title').text,
 +            'title': title,
-             'formats': self._extract_video_formats(mediagen_page),
+             'formats': self._extract_video_formats(mediagen_doc),
              'id': video_id,
              'thumbnail': self._get_thumbnail_url(uri, itemdoc),
              'description': description,
              'Downloading info', transform_source=fix_xml_ampersands)
          return [self._get_video_info(item) for item in idoc.findall('.//item')]
  
+     def _real_extract(self, url):
+         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]
+         except RegexNotFoundError:
+             mgid = self._search_regex(r'data-mgid="(.*?)"', webpage, u'mgid')
+         return self._get_videos_info(mgid)
  
  class MTVIE(MTVServicesInfoExtractor):
      _VALID_URL = r'''(?x)^https?://
      
              uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, 'uri')
          return self._get_videos_info(uri)
+ class MTVIggyIE(MTVServicesInfoExtractor):
+     IE_NAME = 'mtviggy.com'
+     _VALID_URL = r'https?://www\.mtviggy\.com/videos/.+'
+     _TEST = {
+         'url': 'http://www.mtviggy.com/videos/arcade-fire-behind-the-scenes-at-the-biggest-music-experiment-yet/',
+         'info_dict': {
+             'id': '984696',
+             'ext': 'mp4',
+             'title': 'Short',
+         }
+     }
+     _FEED_URL = 'http://all.mtvworldverticals.com/feed-xml/'