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