X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/848723ea972c09f28787db91d8c06e98a274ab89..76d85602f8a22ca3817c3a86f4f0e8969c0b02a9:/youtube_dl/extractor/infoq.py diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py index ed32373..71cfd12 100644 --- a/youtube_dl/extractor/infoq.py +++ b/youtube_dl/extractor/infoq.py @@ -1,55 +1,68 @@ from __future__ import unicode_literals import base64 -import re from .common import InfoExtractor -from ..utils import ( - compat_urllib_parse, +from ..compat import ( + compat_urllib_parse_unquote, + compat_urlparse, ) class InfoQIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P[^/]+)$' - _TEST = { - "name": "InfoQ", - "url": "http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things", - "file": "12-jan-pythonthings.mp4", - "info_dict": { - "description": "Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.", - "title": "A Few of My Favorite [Python] Things", - }, - "params": { - "skip_download": True, + _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P[^/]+)' + + _TESTS = [{ + 'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things', + 'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2', + 'info_dict': { + 'id': '12-jan-pythonthings', + 'ext': 'mp4', + 'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.', + 'title': 'A Few of My Favorite [Python] Things', }, - } + }, { + 'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - # 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')) - video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id + video_title = self._html_search_regex(r'(.*?)', webpage, 'title') + video_description = self._html_search_meta('description', webpage, 'description') - # Extract title - video_title = self._search_regex(r'contentTitle = "(.*?)";', - webpage, 'title') + # The server URL is hardcoded + video_url = 'rtmpe://video.infoq.com/cfx/st/' - # Extract description - video_description = self._html_search_regex(r'', - webpage, 'description', fatal=False) + # Extract video URL + encoded_id = self._search_regex( + r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id') + real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8')) + playpath = 'mp4:' + real_id - video_filename = video_url.split('/')[-1] + video_filename = playpath.split('/')[-1] video_id, extension = video_filename.split('.') + http_base = self._search_regex( + r'EXPRESSINSTALL_SWF\s*=\s*[^"]*"((?:https?:)?//[^/"]+/)', webpage, + 'HTTP base URL') + + formats = [{ + 'format_id': 'rtmp', + 'url': video_url, + 'ext': extension, + 'play_path': playpath, + }, { + 'format_id': 'http', + 'url': compat_urlparse.urljoin(url, http_base) + real_id, + }] + self._sort_formats(formats) + return { 'id': video_id, - 'url': video_url, 'title': video_title, - 'ext': extension, # Extension is always(?) mp4, but seems to be flv 'description': video_description, + 'formats': formats, }