Imported Upstream version 2013.10.01

[youtubedl] / youtube_dl / extractor / arte.py

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index e7a91a1eb5e835c9b6e8bd9f16302a9bc7a8bf90..69b3b0ad7820600ef5107ad3d79230c0e4edcaac 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -5,6 +5,7 @@ import xml.etree.ElementTree
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,

+    find_xpath_attr,

     unified_strdate,
 )
 
     unified_strdate,
 )
 


@@ -16,13 +17,14 @@ class ArteTvIE(InfoExtractor):
     """
     _EMISSION_URL = r'(?:http://)?www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
     _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?P<lang>fr|de)/.*-(?P<id>.*?).html'
     """
     _EMISSION_URL = r'(?:http://)?www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
     _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?P<lang>fr|de)/.*-(?P<id>.*?).html'

+    _LIVEWEB_URL = r'(?:http://)?liveweb.arte.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'

     _LIVE_URL = r'index-[0-9]+\.html$'
 
     IE_NAME = u'arte.tv'
 
     @classmethod
     def suitable(cls, url):
     _LIVE_URL = r'index-[0-9]+\.html$'
 
     IE_NAME = u'arte.tv'
 
     @classmethod
     def suitable(cls, url):

-        return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL))
+        return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL, cls._LIVEWEB_URL))

 
     # TODO implement Live Stream
     # from ..utils import compat_urllib_parse
 
     # TODO implement Live Stream
     # from ..utils import compat_urllib_parse


@@ -67,6 +69,12 @@ class ArteTvIE(InfoExtractor):
             lang = mobj.group('lang')
             return self._extract_video(url, id, lang)
 
             lang = mobj.group('lang')
             return self._extract_video(url, id, lang)
 

+        mobj = re.match(self._LIVEWEB_URL, url)
+        if mobj is not None:
+            name = mobj.group('name')
+            lang = mobj.group('lang')
+            return self._extract_liveweb(url, name, lang)
+

         if re.search(self._LIVE_URL, video_id) is not None:
             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
         if re.search(self._LIVE_URL, video_id) is not None:
             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)


@@ -84,7 +92,7 @@ class ArteTvIE(InfoExtractor):
 
         info_dict = {'id': player_info['VID'],
                      'title': player_info['VTI'],
 
         info_dict = {'id': player_info['VID'],
                      'title': player_info['VTI'],

-                     'description': player_info['VDE'],
+                     'description': player_info.get('VDE'),

                      'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]),
                      'thumbnail': player_info['programImage'],
                      'ext': 'flv',
                      'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]),
                      'thumbnail': player_info['programImage'],
                      'ext': 'flv',


@@ -97,12 +105,14 @@ class ArteTvIE(InfoExtractor):
                 l = 'F'
             elif lang == 'de':
                 l = 'A'
                 l = 'F'
             elif lang == 'de':
                 l = 'A'

-            regexes = [r'VO?%s' % l, r'V%s-ST.' % l]
+            regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]

             return any(re.match(r, f['versionCode']) for r in regexes)
         # Some formats may not be in the same language as the url
         formats = filter(_match_lang, formats)
         # We order the formats by quality
         formats = sorted(formats, key=lambda f: int(f['height']))
             return any(re.match(r, f['versionCode']) for r in regexes)
         # Some formats may not be in the same language as the url
         formats = filter(_match_lang, formats)
         # We order the formats by quality
         formats = sorted(formats, key=lambda f: int(f['height']))

+        # Prefer videos without subtitles in the same language
+        formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f['versionCode']) is None)

         # Pick the best quality
         format_info = formats[-1]
         if format_info['mediaType'] == u'rtmp':
         # Pick the best quality
         format_info = formats[-1]
         if format_info['mediaType'] == u'rtmp':


@@ -119,7 +129,7 @@ class ArteTvIE(InfoExtractor):
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
         ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
         ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
         ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
         ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)

-        config_node = ref_xml_doc.find('.//video[@lang="%s"]' % lang)
+        config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)

         config_xml_url = config_node.attrib['ref']
         config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
 
         config_xml_url = config_node.attrib['ref']
         config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
 


@@ -143,3 +153,22 @@ class ArteTvIE(InfoExtractor):
                 'url': video_url,
                 'ext': 'flv',
                 }
                 'url': video_url,
                 'ext': 'flv',
                 }

+
+    def _extract_liveweb(self, url, name, lang):
+        """Extract form http://liveweb.arte.tv/"""
+        webpage = self._download_webpage(url, name)
+        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id')
+        config_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
+                                            video_id, u'Downloading information')
+        config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+        event_doc = config_doc.find('event')
+        url_node = event_doc.find('video').find('urlHd')
+        if url_node is None:
+            url_node = video_doc.find('urlSd')
+
+        return {'id': video_id,
+                'title': event_doc.find('name%s' % lang.capitalize()).text,
+                'url': url_node.text.replace('MP4', 'mp4'),
+                'ext': 'flv',
+                'thumbnail': self._og_search_thumbnail(webpage),
+                }