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_parse,
   9     compat_str,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     float_or_none,
  15     sanitized_Request,
  16 )
  17
  18
  19 class BambuserIE(InfoExtractor):
  20     IE_NAME = 'bambuser'
  21     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
  22     _API_KEY = '005f64509e19a868399060af746a00aa'
  23     _LOGIN_URL = 'https://bambuser.com/user'
  24     _NETRC_MACHINE = 'bambuser'
  25
  26     _TEST = {
  27         'url': 'http://bambuser.com/v/4050584',
  28         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
  29         # 'md5': 'fba8f7693e48fd4e8641b3fd5539a641',
  30         'info_dict': {
  31             'id': '4050584',
  32             'ext': 'flv',
  33             'title': 'Education engineering days - lightning talks',
  34             'duration': 3741,
  35             'uploader': 'pixelversity',
  36             'uploader_id': '344706',
  37             'timestamp': 1382976692,
  38             'upload_date': '20131028',
  39             'view_count': int,
  40         },
  41         'params': {
  42             # It doesn't respect the 'Range' header, it would download the whole video
  43             # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
  44             'skip_download': True,
  45         },
  46     }
  47
  48     def _login(self):
  49         (username, password) = self._get_login_info()
  50         if username is None:
  51             return
  52
  53         login_form = {
  54             'form_id': 'user_login',
  55             'op': 'Log in',
  56             'name': username,
  57             'pass': password,
  58         }
  59
  60         request = sanitized_Request(
  61             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
  62         request.add_header('Referer', self._LOGIN_URL)
  63         response = self._download_webpage(
  64             request, None, 'Logging in as %s' % username)
  65
  66         login_error = self._html_search_regex(
  67             r'(?s)<div class="messages error">(.+?)</div>',
  68             response, 'login error', default=None)
  69         if login_error:
  70             raise ExtractorError(
  71                 'Unable to login: %s' % login_error, expected=True)
  72
  73     def _real_initialize(self):
  74         self._login()
  75
  76     def _real_extract(self, url):
  77         video_id = self._match_id(url)
  78
  79         info = self._download_json(
  80             'http://player-c.api.bambuser.com/getVideo.json?api_key=%s&vid=%s'
  81             % (self._API_KEY, video_id), video_id)
  82
  83         error = info.get('error')
  84         if error:
  85             raise ExtractorError(
  86                 '%s returned error: %s' % (self.IE_NAME, error), expected=True)
  87
  88         result = info['result']
  89
  90         return {
  91             'id': video_id,
  92             'title': result['title'],
  93             'url': result['url'],
  94             'thumbnail': result.get('preview'),
  95             'duration': int_or_none(result.get('length')),
  96             'uploader': result.get('username'),
  97             'uploader_id': compat_str(result.get('owner', {}).get('uid')),
  98             'timestamp': int_or_none(result.get('created')),
  99             'fps': float_or_none(result.get('framerate')),
 100             'view_count': int_or_none(result.get('views_total')),
 101             'comment_count': int_or_none(result.get('comment_count')),
 102         }
 103
 104
 105 class BambuserChannelIE(InfoExtractor):
 106     IE_NAME = 'bambuser:channel'
 107     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
 108     # The maximum number we can get with each request
 109     _STEP = 50
 110     _TEST = {
 111         'url': 'http://bambuser.com/channel/pixelversity',
 112         'info_dict': {
 113             'title': 'pixelversity',
 114         },
 115         'playlist_mincount': 60,
 116     }
 117
 118     def _real_extract(self, url):
 119         mobj = re.match(self._VALID_URL, url)
 120         user = mobj.group('user')
 121         urls = []
 122         last_id = ''
 123         for i in itertools.count(1):
 124             req_url = (
 125                 'http://bambuser.com/xhr-api/index.php?username={user}'
 126                 '&sort=created&access_mode=0%2C1%2C2&limit={count}'
 127                 '&method=broadcast&format=json&vid_older_than={last}'
 128             ).format(user=user, count=self._STEP, last=last_id)
 129             req = sanitized_Request(req_url)
 130             # Without setting this header, we wouldn't get any result
 131             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
 132             data = self._download_json(
 133                 req, user, 'Downloading page %d' % i)
 134             results = data['result']
 135             if not results:
 136                 break
 137             last_id = results[-1]['vid']
 138             urls.extend(self.url_result(v['page'], 'Bambuser') for v in results)
 139
 140         return {
 141             '_type': 'playlist',
 142             'title': user,
 143             'entries': urls,
 144         }