kumi/yt-dlp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

[viki] Convert to new subtitles system

Browse source
This commit is contained in:
Jaime Marquínez Ferrándiz 2015-02-18 20:37:16 +01:00
parent afbdd3acc3
commit 4f7cea6c53
2 changed files with 22 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

13
test/test_subtitles.py
View file

@ -23,6 +23,7 @@ from youtube_dl.extractor import (
ComedyCentralIE,
NRKTVIE,
RaiIE,
VikiIE,
)
@ -372,5 +373,17 @@ class TestRaiSubtitles(BaseTestSubtitles):
self.assertEqual(md5(subtitles['it']), 'b1d90a98755126b61e667567a1f6680a')
class TestVikiSubtitles(BaseTestSubtitles):
url = 'http://www.viki.com/videos/1060846v-punch-episode-18'
IE = VikiIE
def test_allsubtitles(self):
self.DL.params['writesubtitles'] = True
self.DL.params['allsubtitles'] = True
subtitles = self.getSubtitles()
self.assertEqual(set(subtitles.keys()), set(['en']))
self.assertEqual(md5(subtitles['en']), 'b0b781eeb45efd3f6398a925b259150b')
if __name__ == '__main__':
unittest.main()

17
youtube_dl/extractor/viki.py
View file

@ -2,16 +2,17 @@ from __future__ import unicode_literals
import re
from ..compat import compat_urlparse
from ..utils import (
ExtractorError,
unescapeHTML,
unified_strdate,
US_RATINGS,
)
from .subtitles import SubtitlesInfoExtractor
from .common import InfoExtractor
class VikiIE(SubtitlesInfoExtractor):
class VikiIE(InfoExtractor):
IE_NAME = 'viki'
_VALID_URL = r'^https?://(?:www\.)?viki\.com/videos/(?P<id>[0-9]+v)'
@ -69,9 +70,6 @@ class VikiIE(SubtitlesInfoExtractor):
# subtitles
video_subtitles = self.extract_subtitles(video_id, info_webpage)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(video_id, info_webpage)
return
return {
'id': video_id,
@ -85,12 +83,15 @@ class VikiIE(SubtitlesInfoExtractor):
'upload_date': upload_date,
}
def _get_available_subtitles(self, video_id, info_webpage):
def _get_subtitles(self, video_id, info_webpage):
res = {}
for sturl_html in re.findall(r'<track src="([^"]+)"/>', info_webpage):
for sturl_html in re.findall(r'<track src="([^"]+)"', info_webpage):
sturl = unescapeHTML(sturl_html)
m = re.search(r'/(?P<lang>[a-z]+)\.vtt', sturl)
if not m:
continue
res[m.group('lang')] = sturl
res[m.group('lang')] = [{
'url': compat_urlparse.urljoin('http://www.viki.com', sturl),
'ext': 'vtt',
}]
return res