w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[arte] Prefer json URLs that contain the video id from the 'vid' parameter in the URL (fixes #7920)

Browse source
This commit is contained in:
Jaime Marquínez Ferrándiz 2015-12-23 17:55:58 +01:00
parent 261b4c23c7
commit a8f1d167f6
1 changed files with 15 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
youtube_dl/extractor/arte.py
View file

@ -68,9 +68,13 @@ class ArteTVPlus7IE(InfoExtractor):
def _extract_url_info(cls, url): def _extract_url_info(cls, url):
mobj = re.match(cls._VALID_URL, url) mobj = re.match(cls._VALID_URL, url)
lang = mobj.group('lang') lang = mobj.group('lang')
# This is not a real id, it can be for example AJT for the news query = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
# http://www.arte.tv/guide/fr/emissions/AJT/arte-journal if 'vid' in query:
video_id = mobj.group('id') 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 return video_id, lang
def _real_extract(self, url): def _real_extract(self, url):
@ -79,9 +83,15 @@ class ArteTVPlus7IE(InfoExtractor):
return self._extract_from_webpage(webpage, video_id, lang) return self._extract_from_webpage(webpage, video_id, lang)
def _extract_from_webpage(self, 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( json_url = self._html_search_regex(
[r'arte_vp_url=["\'](.*?)["\']', r'data-url=["\']([^"]+)["\']'], patterns, webpage, 'json vp url', default=None)
webpage, 'json vp url', default=None)
if not json_url: if not json_url:
iframe_url = self._html_search_regex( iframe_url = self._html_search_regex(
r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1', r'<iframe[^>]+src=(["\'])(?P<url>.+\bjson_url=.+?)\1',