From e1fb24569065e93b0406a0f19a3ab75611a8c968 Mon Sep 17 00:00:00 2001 From: =?utf8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Tue, 16 Jul 2013 20:50:16 +0200 Subject: [PATCH] Add CondeNastIE MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit It supports some of the websites of the Condé Nast group: WIRED, GQ, Vogue, Glamour, W Magazine and Vanity Fair. --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/condenast.py | 106 ++++++++++++++++++++++++++++++ 2 files changed, 107 insertions(+) create mode 100644 youtube_dl/extractor/condenast.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 6738c4168..79868aa06 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -9,6 +9,7 @@ from .brightcove import BrightcoveIE from .canalplus import CanalplusIE from .collegehumor import CollegeHumorIE from .comedycentral import ComedyCentralIE +from .condenast import CondeNastIE from .criterion import CriterionIE from .cspan import CSpanIE from .dailymotion import DailymotionIE diff --git a/youtube_dl/extractor/condenast.py b/youtube_dl/extractor/condenast.py new file mode 100644 index 000000000..f336a3c62 --- /dev/null +++ b/youtube_dl/extractor/condenast.py @@ -0,0 +1,106 @@ +# coding: utf-8 + +import re +import json + +from .common import InfoExtractor +from ..utils import ( + compat_urllib_parse, + orderedSet, + compat_urllib_parse_urlparse, + compat_urlparse, +) + + +class CondeNastIE(InfoExtractor): + """ + Condé Nast is a media group, some of its sites use a custom HTML5 player + that works the same in all of them. + """ + + # The keys are the supported sites and the values are the name to be shown + # to the user and in the extractor description. + _SITES = {'wired': u'WIRED', + 'gq': u'GQ', + 'vogue': u'Vogue', + 'glamour': u'Glamour', + 'wmagazine': u'W Magazine', + 'vanityfair': u'Vanity Fair', + } + + _VALID_URL = r'http://(video|www).(?P%s).com/(?Pwatch|series|video)/(?P.+)' % '|'.join(_SITES.keys()) + IE_DESC = u'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values())) + + _TEST = { + u'url': u'http://video.wired.com/watch/3d-printed-speakers-lit-with-led', + u'file': u'5171b343c2b4c00dd0c1ccb3.mp4', + u'md5': u'1921f713ed48aabd715691f774c451f7', + u'info_dict': { + u'title': u'3D Printed Speakers Lit With LED', + u'description': u'Check out these beautiful 3D printed LED speakers. You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.', + } + } + + def _extract_series(self, url, webpage): + title = self._html_search_regex(r'
.*?

(.+?)

', + webpage, u'series title', flags=re.DOTALL) + url_object = compat_urllib_parse_urlparse(url) + base_url = '%s://%s' % (url_object.scheme, url_object.netloc) + m_paths = re.finditer(r'

.*?(.+?)

', + r'
(.+?)
', + ], + webpage, u'description', + fatal=False, flags=re.DOTALL) + params = self._search_regex(r'var params = {(.+?)}[;,]', webpage, + u'player params', flags=re.DOTALL) + video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, u'video id') + player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, u'player id') + target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, u'target') + data = compat_urllib_parse.urlencode({'videoId': video_id, + 'playerId': player_id, + 'target': target, + }) + base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]', + webpage, u'base info url', + default='http://player.cnevids.com/player/loader.js?') + info_url = base_info_url + data + info_page = self._download_webpage(info_url, video_id, + u'Downloading video info') + video_info = self._search_regex(r'var video = ({.+?});', info_page, u'video info') + video_info = json.loads(video_info) + + def _formats_sort_key(f): + type_ord = 1 if f['type'] == 'video/mp4' else 0 + quality_ord = 1 if f['quality'] == 'high' else 0 + return (quality_ord, type_ord) + best_format = sorted(video_info['sources'][0], key=_formats_sort_key)[-1] + + return {'id': video_id, + 'url': best_format['src'], + 'ext': best_format['type'].split('/')[-1], + 'title': video_info['title'], + 'thumbnail': video_info['poster_frame'], + 'description': description, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + site = mobj.group('site') + url_type = mobj.group('type') + id = mobj.group('id') + + self.to_screen(u'Extracting from %s with the Condé Nast extractor' % self._SITES[site]) + webpage = self._download_webpage(url, id) + + if url_type == 'series': + return self._extract_series(url, webpage) + else: + return self._extract_video(webpage) -- 2.22.2