_ 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 (
   8     compat_urllib_request,
   9     compat_str,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     float_or_none,
  15 )
  16
  17
  18 class BambuserIE(InfoExtractor):
  19     IE_NAME = 'bambuser'
  20     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
  21     _API_KEY = '005f64509e19a868399060af746a00aa'
  22
  23     _TEST = {
  24         'url': 'http://bambuser.com/v/4050584',
  25         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
  26         # 'md5': 'fba8f7693e48fd4e8641b3fd5539a641',
  27         'info_dict': {
  28             'id': '4050584',
  29             'ext': 'flv',
  30             'title': 'Education engineering days - lightning talks',
  31             'duration': 3741,
  32             'uploader': 'pixelversity',
  33             'uploader_id': '344706',
  34             'timestamp': 1382976692,
  35             'upload_date': '20131028',
  36             'view_count': int,
  37         },
  38         'params': {
  39             # It doesn't respect the 'Range' header, it would download the whole video
  40             # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
  41             'skip_download': True,
  42         },
  43     }
  44
  45     def _real_extract(self, url):
  46         video_id = self._match_id(url)
  47
  48         info = self._download_json(
  49             'http://player-c.api.bambuser.com/getVideo.json?api_key=%s&vid=%s'
  50             % (self._API_KEY, video_id), video_id)
  51
  52         error = info.get('error')
  53         if error:
  54             raise ExtractorError(
  55                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
  56
  57         result = info['result']
  58
  59         return {
  60             'id': video_id,
  61             'title': result['title'],
  62             'url': result['url'],
  63             'thumbnail': result.get('preview'),
  64             'duration': int_or_none(result.get('length')),
  65             'uploader': result.get('username'),
  66             'uploader_id': compat_str(result.get('owner', {}).get('uid')),
  67             'timestamp': int_or_none(result.get('created')),
  68             'fps': float_or_none(result.get('framerate')),
  69             'view_count': int_or_none(result.get('views_total')),
  70             'comment_count': int_or_none(result.get('comment_count')),
  71         }
  72
  73
  74 class BambuserChannelIE(InfoExtractor):
  75     IE_NAME = 'bambuser:channel'
  76     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
  77     # The maximum number we can get with each request
  78     _STEP = 50
  79     _TEST = {
  80         'url': 'http://bambuser.com/channel/pixelversity',
  81         'info_dict': {
  82             'title': 'pixelversity',
  83         },
  84         'playlist_mincount': 60,
  85     }
  86
  87     def _real_extract(self, url):
  88         mobj = re.match(self._VALID_URL, url)
  89         user = mobj.group('user')
  90         urls = []
  91         last_id = ''
  92         for i in itertools.count(1):
  93             req_url = (
  94                 'http://bambuser.com/xhr-api/index.php?username={user}'
  95                 '&sort=created&access_mode=0%2C1%2C2&limit={count}'
  96                 '&method=broadcast&format=json&vid_older_than={last}'
  97             ).format(user=user, count=self._STEP, last=last_id)
  98             req = compat_urllib_request.Request(req_url)
  99             # Without setting this header, we wouldn't get any result
 100             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
 101             data = self._download_json(
 102                 req, user, 'Downloading page %d' % i)
 103             results = data['result']
 104             if not results:
 105                 break
 106             last_id = results[-1]['vid']
 107             urls.extend(self.url_result(v['page'], 'Bambuser') for v in results)
 108
 109         return {
 110             '_type': 'playlist',
 111             'title': user,
 112             'entries': urls,
 113         }