midgard/youtube-dl
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[imgur] support regular URL

Browse source
This commit is contained in:
John Boehr 2015-02-18 18:12:48 -08:00
parent 1ac1af9b47
commit 1a13940c8d
1 changed files with 15 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

17
youtube_dl/extractor/imgur.py
View file

@ -7,11 +7,11 @@ from ..utils import (
int_or_none, int_or_none,
js_to_json, js_to_json,
mimetype2ext, mimetype2ext,
ExtractorError,
) )
class ImgurIE(InfoExtractor): class ImgurIE(InfoExtractor):
_VALID_URL = r'https?://i\.imgur\.com/(?P<id>[a-zA-Z0-9]+)\.(?:mp4|gifv)' _VALID_URL = r'https?://(?:i\.)?imgur\.com/(?P<id>[a-zA-Z0-9]+)(?:\.)?(?:mp4|gifv)?'
_TESTS = [{ _TESTS = [{
'url': 'https://i.imgur.com/A61SaA1.gifv', 'url': 'https://i.imgur.com/A61SaA1.gifv',
@ -21,12 +21,25 @@ class ImgurIE(InfoExtractor):
'title': 'MRW gifv is up and running without any bugs', 'title': 'MRW gifv is up and running without any bugs',
'description': 'The Internet\'s visual storytelling community. Explore, share, and discuss the best visual stories the Internet has to offer.', 'description': 'The Internet\'s visual storytelling community. Explore, share, and discuss the best visual stories the Internet has to offer.',
}, },
}, {
'url': 'https://imgur.com/A61SaA1',
'info_dict': {
'id': 'A61SaA1',
'ext': 'mp4',
'title': 'MRW gifv is up and running without any bugs',
'description': 'The Internet\'s visual storytelling community. Explore, share, and discuss the best visual stories the Internet has to offer.',
},
}] }]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
sources = re.findall(r'<source src="([^"]+)" type="([^"]+)"', webpage)
if not sources:
raise ExtractorError(
'No sources found for video %s' % video_id, expected=True)
width = int_or_none(self._search_regex( width = int_or_none(self._search_regex(
r'<param name="width" value="([0-9]+)"', r'<param name="width" value="([0-9]+)"',
webpage, 'width', fatal=False)) webpage, 'width', fatal=False))