X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fbambuser.py;h=ccd31c4c7093d54e86df50a42600d08e12e55005;hb=b21e25702f05018b64064b4aec9007e6e383b476;hp=cf8da22e3b71afa0e48afe6affdd10b996932337;hpb=72a5b4f70216fe1a5b1c22be34653ae0ff81058a;p=youtube-dl diff --git a/youtube_dl/extractor/bambuser.py b/youtube_dl/extractor/bambuser.py index cf8da22e3..ccd31c4c7 100644 --- a/youtube_dl/extractor/bambuser.py +++ b/youtube_dl/extractor/bambuser.py @@ -1,23 +1,36 @@ +from __future__ import unicode_literals + import re import json +import itertools from .common import InfoExtractor +from ..utils import ( + compat_urllib_request, +) class BambuserIE(InfoExtractor): + IE_NAME = 'bambuser' _VALID_URL = r'https?://bambuser\.com/v/(?P\d+)' _API_KEY = '005f64509e19a868399060af746a00aa' _TEST = { - u'url': u'http://bambuser.com/v/4050584', - u'md5': u'fba8f7693e48fd4e8641b3fd5539a641', - u'info_dict': { - u'id': u'4050584', - u'ext': u'flv', - u'title': u'Education engineering days - lightning talks', - u'duration': 3741, - u'uploader': u'pixelversity', - u'uploader_id': u'344706', + 'url': 'http://bambuser.com/v/4050584', + # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388 + #u'md5': 'fba8f7693e48fd4e8641b3fd5539a641', + 'info_dict': { + 'id': '4050584', + 'ext': 'flv', + 'title': 'Education engineering days - lightning talks', + 'duration': 3741, + 'uploader': 'pixelversity', + 'uploader_id': '344706', + }, + 'params': { + # It doesn't respect the 'Range' header, it would download the whole video + # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59 + 'skip_download': True, }, } @@ -33,10 +46,43 @@ class BambuserIE(InfoExtractor): 'id': video_id, 'title': info['title'], 'url': info['url'], - 'thumbnail': info['preview'], + 'thumbnail': info.get('preview'), 'duration': int(info['length']), 'view_count': int(info['views_total']), 'uploader': info['username'], 'uploader_id': info['uid'], } + +class BambuserChannelIE(InfoExtractor): + IE_NAME = 'bambuser:channel' + _VALID_URL = r'https?://bambuser\.com/channel/(?P.*?)(?:/|#|\?|$)' + # The maximum number we can get with each request + _STEP = 50 + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + user = mobj.group('user') + urls = [] + last_id = '' + for i in itertools.count(1): + req_url = ('http://bambuser.com/xhr-api/index.php?username={user}' + '&sort=created&access_mode=0%2C1%2C2&limit={count}' + '&method=broadcast&format=json&vid_older_than={last}' + ).format(user=user, count=self._STEP, last=last_id) + req = compat_urllib_request.Request(req_url) + # Without setting this header, we wouldn't get any result + req.add_header('Referer', 'http://bambuser.com/channel/%s' % user) + info_json = self._download_webpage(req, user, + 'Downloading page %d' % i) + results = json.loads(info_json)['result'] + if len(results) == 0: + break + last_id = results[-1]['vid'] + urls.extend(self.url_result(v['page'], 'Bambuser') for v in results) + + return { + '_type': 'playlist', + 'title': user, + 'entries': urls, + }