Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bambuser.py

   1 import re
   2 import json
   3 import itertools
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_request,
   8 )
   9
  10
  11 class BambuserIE(InfoExtractor):
  12     IE_NAME = u'bambuser'
  13     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
  14     _API_KEY = '005f64509e19a868399060af746a00aa'
  15
  16     _TEST = {
  17         u'url': u'http://bambuser.com/v/4050584',
  18         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
  19         #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641',
  20         u'info_dict': {
  21             u'id': u'4050584',
  22             u'ext': u'flv',
  23             u'title': u'Education engineering days - lightning talks',
  24             u'duration': 3741,
  25             u'uploader': u'pixelversity',
  26             u'uploader_id': u'344706',
  27         },
  28         u'params': {
  29             # It doesn't respect the 'Range' header, it would download the whole video
  30             # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
  31             u'skip_download': True,
  32         },
  33     }
  34
  35     def _real_extract(self, url):
  36         mobj = re.match(self._VALID_URL, url)
  37         video_id = mobj.group('id')
  38         info_url = ('http://player-c.api.bambuser.com/getVideo.json?'
  39             '&api_key=%s&vid=%s' % (self._API_KEY, video_id))
  40         info_json = self._download_webpage(info_url, video_id)
  41         info = json.loads(info_json)['result']
  42
  43         return {
  44             'id': video_id,
  45             'title': info['title'],
  46             'url': info['url'],
  47             'thumbnail': info.get('preview'),
  48             'duration': int(info['length']),
  49             'view_count': int(info['views_total']),
  50             'uploader': info['username'],
  51             'uploader_id': info['uid'],
  52         }
  53
  54
  55 class BambuserChannelIE(InfoExtractor):
  56     IE_NAME = u'bambuser:channel'
  57     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
  58     # The maximum number we can get with each request
  59     _STEP = 50
  60
  61     def _real_extract(self, url):
  62         mobj = re.match(self._VALID_URL, url)
  63         user = mobj.group('user')
  64         urls = []
  65         last_id = ''
  66         for i in itertools.count(1):
  67             req_url = ('http://bambuser.com/xhr-api/index.php?username={user}'
  68                 '&sort=created&access_mode=0%2C1%2C2&limit={count}'
  69                 '&method=broadcast&format=json&vid_older_than={last}'
  70                 ).format(user=user, count=self._STEP, last=last_id)
  71             req = compat_urllib_request.Request(req_url)
  72             # Without setting this header, we wouldn't get any result
  73             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
  74             info_json = self._download_webpage(req, user,
  75                 u'Downloading page %d' % i)
  76             results = json.loads(info_json)['result']
  77             if len(results) == 0:
  78                 break
  79             last_id = results[-1]['vid']
  80             urls.extend(self.url_result(v['page'], 'Bambuser') for v in results)
  81
  82         return {
  83             '_type': 'playlist',
  84             'title': user,
  85             'entries': urls,
  86         }