X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Finfoq.py;h=65f0b76cb0e2a413e82005405b012b4a93fae107;hb=7560096db5ce0002b3cf3f3bcbbb53ef05e6fb13;hp=9a13c52a8fa4d0b4a416d31411f37a63255bccf1;hpb=9d069c477895862d93e223b6d8cc475d42746033;p=youtube-dl diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py index 9a13c52a8..65f0b76cb 100644 --- a/youtube_dl/extractor/infoq.py +++ b/youtube_dl/extractor/infoq.py @@ -1,62 +1,58 @@ +from __future__ import unicode_literals + import base64 import re from .common import InfoExtractor from ..utils import ( compat_urllib_parse, - - ExtractorError, ) class InfoQIE(InfoExtractor): - _VALID_URL = r'^(?:https?://)?(?:www\.)?infoq\.com/[^/]+/[^/]+$' + _VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P[^/]+)$' + _TEST = { - u"name": u"InfoQ", - u"url": u"http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things", - u"file": u"12-jan-pythonthings.mp4", - u"info_dict": { - u"description": u"Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.", - u"title": u"A Few of My Favorite [Python] Things" + u'name': u'InfoQ', + u'url': u'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things', + u'md5': u'fcaa3d995e04080dcb9465d86b5eef62', + u'info_dict': { + u'id': u'12-jan-pythonthings', + u'ext': u'mp4', + u'description': u'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.', + u'title': u'A Few of My Favorite [Python] Things', }, - u"params": { - u"skip_download": true - } } def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') - webpage = self._download_webpage(url, video_id=url) - self.report_extraction(url) + webpage = self._download_webpage(url, video_id) - # Extract video URL - mobj = re.search(r"jsclassref ?= ?'([^']*)'", webpage) - if mobj is None: - raise ExtractorError(u'Unable to extract video url') - real_id = compat_urllib_parse.unquote(base64.b64decode(mobj.group(1).encode('ascii')).decode('utf-8')) - video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id + self.report_extraction(video_id) - # Extract title - video_title = self._search_regex(r'contentTitle = "(.*?)";', - webpage, u'title') + video_title = self._html_search_regex(r'(.*?)', webpage, 'title') + video_description = self._html_search_meta('description', webpage, 'description') - # Extract description - video_description = self._html_search_regex(r'', - webpage, u'description', fatal=False) + # The server URL is hardcoded + video_url = 'rtmpe://video.infoq.com/cfx/st/' - video_filename = video_url.split('/')[-1] + # Extract video URL + encoded_id = self._search_regex(r"jsclassref ?= ?'([^']*)'", webpage, 'encoded id') + real_id = compat_urllib_parse.unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8')) + playpath = 'mp4:' + real_id + + video_filename = playpath.split('/')[-1] video_id, extension = video_filename.split('.') - info = { + return [{ 'id': video_id, - 'url': video_url, - 'uploader': None, - 'upload_date': None, 'title': video_title, - 'ext': extension, # Extension is always(?) mp4, but seems to be flv - 'thumbnail': None, 'description': video_description, - } - - return [info] \ No newline at end of file + 'formats': [{ + 'url': video_url, + 'ext': extension, + 'play_path': playpath, + }], + }]