[ninegag] Fix extraction

This commit is contained in:
Philipp Hagemeister 2014-03-13 16:40:53 +01:00
parent 9d2ecdbc71
commit 84769e708c
2 changed files with 24 additions and 12 deletions

View file

@ -144,6 +144,10 @@ def generator(test_case):
self.assertTrue( self.assertTrue(
isinstance(got, compat_str) and match_rex.match(got), isinstance(got, compat_str) and match_rex.match(got),
u'field %s (value: %r) should match %r' % (info_field, got, match_str)) u'field %s (value: %r) should match %r' % (info_field, got, match_str))
elif isinstance(expected, type):
got = info_dict.get(info_field)
self.assertTrue(isinstance(got, expected),
u'Expected type %r, but got value %r of type %r' % (expected, got, type(got)))
else: else:
if isinstance(expected, compat_str) and expected.startswith('md5:'): if isinstance(expected, compat_str) and expected.startswith('md5:'):
got = 'md5:' + md5(info_dict.get(info_field)) got = 'md5:' + md5(info_dict.get(info_field))

View file

@ -15,7 +15,9 @@ class NineGagIE(InfoExtractor):
"file": "1912.mp4", "file": "1912.mp4",
"info_dict": { "info_dict": {
"description": "This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)", "description": "This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)",
"title": "\"People Are Awesome 2013\" Is Absolutely Awesome" "title": "\"People Are Awesome 2013\" Is Absolutely Awesome",
"view_count": int,
"thumbnail": "re:^https?://",
}, },
'add_ie': ['Youtube'] 'add_ie': ['Youtube']
} }
@ -25,21 +27,27 @@ class NineGagIE(InfoExtractor):
video_id = mobj.group('id') video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
data_json = self._html_search_regex(r'''(?x)
<div\s*id="tv-video"\s*data-video-source="youtube"\s*
data-video-meta="([^"]+)"''', webpage, 'video metadata')
data = json.loads(data_json) youtube_id = self._html_search_regex(
r'(?s)id="jsid-video-post-container".*?data-external-id="([^"]+)"',
webpage, 'video ID')
description = self._html_search_regex(
r'(?s)<div class="video-caption">.*?<p>(.*?)</p>', webpage,
'description', fatal=False)
view_count_str = self._html_search_regex(
r'<p><b>([0-9][0-9,]*)</b> views</p>', webpage, 'view count',
fatal=False)
view_count = (
None if view_count_str is None
else int(view_count_str.replace(',', '')))
return { return {
'_type': 'url_transparent', '_type': 'url_transparent',
'url': data['youtubeVideoId'], 'url': youtube_id,
'ie_key': 'Youtube', 'ie_key': 'Youtube',
'id': video_id, 'id': video_id,
'title': data['title'], 'title': self._og_search_title(webpage),
'description': data['description'], 'description': description,
'view_count': int(data['view_count']), 'view_count': view_count,
'like_count': int(data['statistic']['like']), 'thumbnail': self._og_search_thumbnail(webpage),
'dislike_count': int(data['statistic']['dislike']),
'thumbnail': data['thumbnail_url'],
} }