youtube_dl/extractor/bambuser.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5 import itertools
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_urllib_request,
  10 )
  11
  12
  13 class BambuserIE(InfoExtractor):
  14     IE_NAME = 'bambuser'
  15     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
  16     _API_KEY = '005f64509e19a868399060af746a00aa'
  17
  18     _TEST = {
  19         'url': 'http://bambuser.com/v/4050584',
  20         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
  21         #u'md5': 'fba8f7693e48fd4e8641b3fd5539a641',
  22         'info_dict': {
  23             'id': '4050584',
  24             'ext': 'flv',
  25             'title': 'Education engineering days - lightning talks',
  26             'duration': 3741,
  27             'uploader': 'pixelversity',
  28             'uploader_id': '344706',
  29         },
  30         'params': {
  31             # It doesn't respect the 'Range' header, it would download the whole video
  32             # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
  33             'skip_download': True,
  34         },
  35     }
  36
  37     def _real_extract(self, url):
  38         mobj = re.match(self._VALID_URL, url)
  39         video_id = mobj.group('id')
  40         info_url = ('http://player-c.api.bambuser.com/getVideo.json?'
  41             '&api_key=%s&vid=%s' % (self._API_KEY, video_id))
  42         info_json = self._download_webpage(info_url, video_id)
  43         info = json.loads(info_json)['result']
  44
  45         return {
  46             'id': video_id,
  47             'title': info['title'],
  48             'url': info['url'],
  49             'thumbnail': info.get('preview'),
  50             'duration': int(info['length']),
  51             'view_count': int(info['views_total']),
  52             'uploader': info['username'],
  53             'uploader_id': info['uid'],
  54         }
  55
  56
  57 class BambuserChannelIE(InfoExtractor):
  58     IE_NAME = 'bambuser:channel'
  59     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
  60     # The maximum number we can get with each request
  61     _STEP = 50
  62
  63     def _real_extract(self, url):
  64         mobj = re.match(self._VALID_URL, url)
  65         user = mobj.group('user')
  66         urls = []
  67         last_id = ''
  68         for i in itertools.count(1):
  69             req_url = ('http://bambuser.com/xhr-api/index.php?username={user}'
  70                 '&sort=created&access_mode=0%2C1%2C2&limit={count}'
  71                 '&method=broadcast&format=json&vid_older_than={last}'
  72                 ).format(user=user, count=self._STEP, last=last_id)
  73             req = compat_urllib_request.Request(req_url)
  74             # Without setting this header, we wouldn't get any result
  75             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
  76             info_json = self._download_webpage(req, user,
  77                 'Downloading page %d' % i)
  78             results = json.loads(info_json)['result']
  79             if len(results) == 0:
  80                 break
  81             last_id = results[-1]['vid']
  82             urls.extend(self.url_result(v['page'], 'Bambuser') for v in results)
  83
  84         return {
  85             '_type': 'playlist',
  86             'title': user,
  87             'entries': urls,
  88         }