-import json
+from __future__ import unicode_literals
+
import re
from .common import InfoExtractor
_VALID_URL = r'^https?://(?:www\.)?9gag\.tv/v/(?P<id>[0-9]+)'
_TEST = {
- u"url": u"http://9gag.tv/v/1912",
- u"file": u"1912.mp4",
- u"info_dict": {
- u"description": u"This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)",
- u"title": u"\"People Are Awesome 2013\" Is Absolutely Awesome"
+ "url": "http://9gag.tv/v/1912",
+ "info_dict": {
+ "id": "1912",
+ "ext": "mp4",
+ "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",
+ "view_count": int,
+ "thumbnail": "re:^https?://",
},
- u'add_ie': [u'Youtube']
+ 'add_ie': ['Youtube']
}
def _real_extract(self, url):
video_id = mobj.group('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, u'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 {
'_type': 'url_transparent',
- 'url': data['youtubeVideoId'],
+ 'url': youtube_id,
'ie_key': 'Youtube',
'id': video_id,
- 'title': data['title'],
- 'description': data['description'],
- 'view_count': int(data['view_count']),
- 'like_count': int(data['statistic']['like']),
- 'dislike_count': int(data['statistic']['dislike']),
- 'thumbnail': data['thumbnail_url'],
+ 'title': self._og_search_title(webpage),
+ 'description': description,
+ 'view_count': view_count,
+ 'thumbnail': self._og_search_thumbnail(webpage),
}