Update changelog.

[youtubedl] / youtube_dl / extractor / arte.py

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index 3a57ce52785a7d4950ec694b71fb9bdff69d1e55..2a00da3ee0b7f865cb2b14e2c4920361c35022f8 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -4,16 +4,19 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+)

 from ..utils import (
     find_xpath_attr,
     unified_strdate,
 from ..utils import (
     find_xpath_attr,
     unified_strdate,

-    get_element_by_id,

     get_element_by_attribute,
     int_or_none,
     qualities,
 )
 
     get_element_by_attribute,
     int_or_none,
     qualities,
 )
 

-# There are different sources of video in arte.tv, the extraction process 
+# There are different sources of video in arte.tv, the extraction process

 # is different for each one. The videos usually expire in 7 days, so we can't
 # add tests.
 
 # is different for each one. The videos usually expire in 7 days, so we can't
 # add tests.
 


@@ -37,7 +40,7 @@ class ArteTvIE(InfoExtractor):
             config_xml_url, video_id, note='Downloading configuration')
 
         formats = [{
             config_xml_url, video_id, note='Downloading configuration')
 
         formats = [{

-            'forma_id': q.attrib['quality'],
+            'format_id': q.attrib['quality'],

             # The playpath starts at 'mp4:', if we don't manually
             # split the url, rtmpdump will incorrectly parse them
             'url': q.text.split('mp4:', 1)[0],
             # The playpath starts at 'mp4:', if we don't manually
             # split the url, rtmpdump will incorrectly parse them
             'url': q.text.split('mp4:', 1)[0],


@@ -78,7 +81,13 @@ class ArteTVPlus7IE(InfoExtractor):
     def _extract_from_webpage(self, webpage, video_id, lang):
         json_url = self._html_search_regex(
             [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
     def _extract_from_webpage(self, webpage, video_id, lang):
         json_url = self._html_search_regex(
             [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],

-            webpage, 'json vp url')
+            webpage, 'json vp url', default=None)
+        if not json_url:
+            iframe_url = self._html_search_regex(
+                r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',
+                webpage, 'iframe url', group='url')
+            json_url = compat_parse_qs(
+                compat_urllib_parse_urlparse(iframe_url).query)['json_url'][0]

         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):
         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):


@@ -133,7 +142,7 @@ class ArteTVPlus7IE(InfoExtractor):
                 'width': int_or_none(f.get('width')),
                 'height': int_or_none(f.get('height')),
                 'tbr': int_or_none(f.get('bitrate')),
                 'width': int_or_none(f.get('width')),
                 'height': int_or_none(f.get('height')),
                 'tbr': int_or_none(f.get('bitrate')),

-                'quality': qfunc(f['quality']),
+                'quality': qfunc(f.get('quality')),

                 'source_preference': source_pref,
             }
 
                 'source_preference': source_pref,
             }
 


@@ -146,6 +155,7 @@ class ArteTVPlus7IE(InfoExtractor):
 
             formats.append(format)
 
 
             formats.append(format)
 

+        self._check_formats(formats, video_id)

         self._sort_formats(formats)
 
         info_dict['formats'] = formats
         self._sort_formats(formats)
 
         info_dict['formats'] = formats


@@ -194,7 +204,9 @@ class ArteTVFutureIE(ArteTVPlus7IE):
     def _real_extract(self, url):
         anchor_id, lang = self._extract_url_info(url)
         webpage = self._download_webpage(url, anchor_id)
     def _real_extract(self, url):
         anchor_id, lang = self._extract_url_info(url)
         webpage = self._download_webpage(url, anchor_id)

-        row = get_element_by_id(anchor_id, webpage)
+        row = self._search_regex(
+            r'(?s)id="%s"[^>]*>.+?(<div[^>]*arte_vp_url[^>]*>)' % anchor_id,
+            webpage, 'row')

         return self._extract_from_webpage(row, anchor_id, lang)
 
 
         return self._extract_from_webpage(row, anchor_id, lang)