[arte:future] Fix extraction

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

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index 8273bd6c9ae3cdff82052c8f63efc68be97561b3..964d38fdf00b09505d632aa9dd55f80f7f904a74 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -4,10 +4,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_urlparse,
+)
 from ..utils import (
     find_xpath_attr,
     unified_strdate,
-    get_element_by_id,
     get_element_by_attribute,
     int_or_none,
     qualities,
@@ -65,9 +68,13 @@ class ArteTVPlus7IE(InfoExtractor):
     def _extract_url_info(cls, url):
         mobj = re.match(cls._VALID_URL, url)
         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')
+        query = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
+        if 'vid' in query:
+            video_id = query['vid'][0]
+        else:
+            # 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 video_id, lang
 
     def _real_extract(self, url):
@@ -76,9 +83,21 @@ class ArteTVPlus7IE(InfoExtractor):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):
+        patterns_templates = (r'arte_vp_url=["\'](.*?%s.*?)["\']', r'data-url=["\']([^"]+%s[^"]+)["\']')
+        ids = (video_id, '')
+        # some pages contain multiple videos (like
+        # http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D),
+        # so we first try to look for json URLs that contain the video id from
+        # the 'vid' parameter.
+        patterns = [t % re.escape(_id) for _id in ids for t in patterns_templates]
         json_url = self._html_search_regex(
-            [r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'],
-            webpage, 'json vp url')
+            patterns, 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):
@@ -180,23 +199,26 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
 
 class ArteTVFutureIE(ArteTVPlus7IE):
     IE_NAME = 'arte.tv:future'
-    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
-
-    _TEST = {
-        'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
-        'info_dict': {
-            'id': '5201',
-            'ext': 'mp4',
-            'title': 'Les champignons au secours de la planète',
-            'upload_date': '20131101',
+    _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(?P<id>.+)'
+
+    _TESTS = [
+        {
+            'url': 'http://future.arte.tv/fr/info-sciences/les-ecrevisses-aussi-sont-anxieuses',
+            'info_dict': {
+                'id': '050940-028-A',
+                'ext': 'mp4',
+                'title': 'Les écrevisses aussi peuvent être anxieuses',
+            },
         },
-    }
-
-    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)
-        return self._extract_from_webpage(row, anchor_id, lang)
+        {
+            'url': 'http://future.arte.tv/fr/la-science-est-elle-responsable',
+            'info_dict': {
+                'id': '061982-002-A',
+                'ext': 'mp4',
+                'title': 'Brian P. Schmidt - Prix Nobel de physique 2011',
+            },
+        }
+    ]
 
 
 class ArteTVDDCIE(ArteTVPlus7IE):