_ Git - youtube-dl/blob - youtube_dl/extractor/bambuser.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import itertools
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_request
   8 from ..utils import ExtractorError
   9
  10
  11 class BambuserIE(InfoExtractor):
  12     IE_NAME = 'bambuser'
  13     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
  14     _API_KEY = '005f64509e19a868399060af746a00aa'
  15
  16     _TEST = {
  17         'url': 'http://bambuser.com/v/4050584',
  18         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
  19         # 'md5': 'fba8f7693e48fd4e8641b3fd5539a641',
  20         'info_dict': {
  21             'id': '4050584',
  22             'ext': 'flv',
  23             'title': 'Education engineering days - lightning talks',
  24             'duration': 3741,
  25             'uploader': 'pixelversity',
  26             'uploader_id': '344706',
  27         },
  28         '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             'skip_download': True,
  32         },
  33     }
  34
  35     def _real_extract(self, url):
  36         video_id = self._match_id(url)
  37
  38         info = self._download_json(
  39             'http://player-c.api.bambuser.com/getVideo.json?api_key=%s&vid=%s'
  40             % (self._API_KEY, video_id), video_id)
  41
  42         error = info.get('error')
  43         if error:
  44             raise ExtractorError(
  45                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
  46
  47         result = info['result']
  48
  49         return {
  50             'id': video_id,
  51             'title': result['title'],
  52             'url': result['url'],
  53             'thumbnail': result.get('preview'),
  54             'duration': int(result['length']),
  55             'view_count': int(result['views_total']),
  56             'uploader': result['username'],
  57             'uploader_id': result['owner']['uid'],
  58         }
  59
  60
  61 class BambuserChannelIE(InfoExtractor):
  62     IE_NAME = 'bambuser:channel'
  63     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
  64     # The maximum number we can get with each request
  65     _STEP = 50
  66     _TEST = {
  67         'url': 'http://bambuser.com/channel/pixelversity',
  68         'info_dict': {
  69             'title': 'pixelversity',
  70         },
  71         'playlist_mincount': 60,
  72     }
  73
  74     def _real_extract(self, url):
  75         mobj = re.match(self._VALID_URL, url)
  76         user = mobj.group('user')
  77         urls = []
  78         last_id = ''
  79         for i in itertools.count(1):
  80             req_url = (
  81                 'http://bambuser.com/xhr-api/index.php?username={user}'
  82                 '&sort=created&access_mode=0%2C1%2C2&limit={count}'
  83                 '&method=broadcast&format=json&vid_older_than={last}'
  84             ).format(user=user, count=self._STEP, last=last_id)
  85             req = compat_urllib_request.Request(req_url)
  86             # Without setting this header, we wouldn't get any result
  87             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
  88             data = self._download_json(
  89                 req, user, 'Downloading page %d' % i)
  90             results = data['result']
  91             if not results:
  92                 break
  93             last_id = results[-1]['vid']
  94             urls.extend(self.url_result(v['page'], 'Bambuser') for v in results)
  95
  96         return {
  97             '_type': 'playlist',
  98             'title': user,
  99             'entries': urls,
 100         }