New test release.

[youtubedl] / youtube_dl / extractor / arte.py

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index 183274eb75f0947d11434867f30ba3ef9f6fcc0c..e7a91a1eb5e835c9b6e8bd9f16302a9bc7a8bf90 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -1,11 +1,9 @@
 import re
 import json
 import re
 import json

+import xml.etree.ElementTree

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (

-    # This is used by the not implemented extractLiveStream method
-    compat_urllib_parse,
-

     ExtractorError,
     unified_strdate,
 )
     ExtractorError,
     unified_strdate,
 )


@@ -16,8 +14,8 @@ class ArteTvIE(InfoExtractor):
     www.arte.tv/guide, the extraction process is different for each one.
     The videos expire in 7 days, so we can't add tests.
     """
     www.arte.tv/guide, the extraction process is different for each one.
     The videos expire in 7 days, so we can't add tests.
     """

-    _EMISSION_URL = r'(?:http://)?www\.arte.tv/guide/(?:fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
-    _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?: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'

     _LIVE_URL = r'index-[0-9]+\.html$'
 
     IE_NAME = u'arte.tv'
     _LIVE_URL = r'index-[0-9]+\.html$'
 
     IE_NAME = u'arte.tv'


@@ -27,6 +25,7 @@ class ArteTvIE(InfoExtractor):
         return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL))
 
     # TODO implement Live Stream
         return any(re.match(regex, url) for regex in (cls._EMISSION_URL, cls._VIDEOS_URL))
 
     # TODO implement Live Stream

+    # from ..utils import compat_urllib_parse

     # def extractLiveStream(self, url):
     #     video_lang = url.split('/')[-4]
     #     info = self.grep_webpage(
     # def extractLiveStream(self, url):
     #     video_lang = url.split('/')[-4]
     #     info = self.grep_webpage(


@@ -56,23 +55,24 @@ class ArteTvIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._EMISSION_URL, url)
         if mobj is not None:
     def _real_extract(self, url):
         mobj = re.match(self._EMISSION_URL, url)
         if mobj is not None:

-            name = mobj.group('name')
+            lang = mobj.group('lang')

             # This is not a real id, it can be for example AJT for the news
             # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
             video_id = mobj.group('id')
             # This is not a real id, it can be for example AJT for the news
             # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
             video_id = mobj.group('id')

-            return self._extract_emission(url, video_id)
+            return self._extract_emission(url, video_id, lang)

 
         mobj = re.match(self._VIDEOS_URL, url)
         if mobj is not None:
             id = mobj.group('id')
 
         mobj = re.match(self._VIDEOS_URL, url)
         if mobj is not None:
             id = mobj.group('id')

-            return self._extract_video(url, id)
+            lang = mobj.group('lang')
+            return self._extract_video(url, id, 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)
             # return
 
 
         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)
             # return
 

-    def _extract_emission(self, url, video_id):
+    def _extract_emission(self, url, video_id, lang):

         """Extract from www.arte.tv/guide"""
         webpage = self._download_webpage(url, video_id)
         json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
         """Extract from www.arte.tv/guide"""
         webpage = self._download_webpage(url, video_id)
         json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')


@@ -91,6 +91,16 @@ class ArteTvIE(InfoExtractor):
                      }
 
         formats = player_info['VSR'].values()
                      }
 
         formats = player_info['VSR'].values()

+        def _match_lang(f):
+            # Return true if that format is in the language of the url
+            if lang == 'fr':
+                l = 'F'
+            elif lang == 'de':
+                l = 'A'
+            regexes = [r'VO?%s' % l, r'V%s-ST.' % 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']))
         # Pick the best quality
         # We order the formats by quality
         formats = sorted(formats, key=lambda f: int(f['height']))
         # Pick the best quality


@@ -103,13 +113,15 @@ class ArteTvIE(InfoExtractor):
 
         return info_dict
 
 
         return info_dict
 

-    def _extract_video(self, url, video_id):
+    def _extract_video(self, url, video_id, lang):

         """Extract from videos.arte.tv"""
         """Extract from videos.arte.tv"""

-        config_xml_url = url.replace('/videos/', '/do_delegate/videos/')
-        config_xml_url = config_xml_url.replace('.html', ',view,asPlayerXml.xml')
-        config_xml = self._download_webpage(config_xml_url, video_id)
-        config_xml_url = self._html_search_regex(r'<video lang=".*?" ref="(.*?)"', config_xml, 'config xml url')
-        config_xml = self._download_webpage(config_xml_url, video_id)
+        ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
+        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_xml_url = config_node.attrib['ref']
+        config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')

 
         video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
         def _key(m):
 
         video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
         def _key(m):