X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fheise.py;h=82e11a7d88735f2105d0dff70c4304ecbf274ff0;hb=950b5f296986ed0a2dd9feeb69dbb950592b6047;hp=f97b1e0854e0655d4cab866ace1dccd58a13f7ec;hpb=f0b8e3607db6bc2e7cdfcf3175e85d9bccb22229;p=youtube-dl diff --git a/youtube_dl/extractor/heise.py b/youtube_dl/extractor/heise.py index f97b1e085..82e11a7d8 100644 --- a/youtube_dl/extractor/heise.py +++ b/youtube_dl/extractor/heise.py @@ -2,80 +2,110 @@ from __future__ import unicode_literals from .common import InfoExtractor +from .youtube import YoutubeIE from ..utils import ( - get_meta_content, + determine_ext, + int_or_none, parse_iso8601, + xpath_text, ) class HeiseIE(InfoExtractor): - _VALID_URL = r'''(?x) - https?://(?:www\.)?heise\.de/video/artikel/ - .+?(?P[0-9]+)\.html(?:$|[?#]) - ''' - _TEST = { - 'url': ( - 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html' - ), + _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P[0-9]+)\.html' + _TESTS = [{ + 'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html', 'md5': 'ffed432483e922e88545ad9f2f15d30e', 'info_dict': { 'id': '2404147', 'ext': 'mp4', - 'title': ( - "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone" - ), - 'format_id': 'mp4_720', + 'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone", + 'format_id': 'mp4_720p', 'timestamp': 1411812600, 'upload_date': '20140927', - 'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.', + 'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20', + 'thumbnail': r're:^https?://.*/gallery/$', } - } + }, { + # YouTube embed + 'url': 'http://www.heise.de/newsticker/meldung/Netflix-In-20-Jahren-vom-Videoverleih-zum-TV-Revolutionaer-3814130.html', + 'md5': 'e403d2b43fea8e405e88e3f8623909f1', + 'info_dict': { + 'id': '6kmWbXleKW4', + 'ext': 'mp4', + 'title': 'NEU IM SEPTEMBER | Netflix', + 'description': 'md5:2131f3c7525e540d5fd841de938bd452', + 'upload_date': '20170830', + 'uploader': 'Netflix Deutschland, Österreich und Schweiz', + 'uploader_id': 'netflixdach', + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html', + 'only_matching': True, + }, { + 'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom', + 'only_matching': True, + }, { + 'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - json_url = self._search_regex( - r'json_url:\s*"([^"]+)"', webpage, 'json URL') - config = self._download_json(json_url, video_id) - - info = { - 'id': video_id, - 'thumbnail': config.get('poster'), - 'timestamp': parse_iso8601(get_meta_content('date', webpage)), - 'description': self._og_search_description(webpage), - } - title = get_meta_content('fulltitle', webpage) - if title: - info['title'] = title - elif config.get('title'): - info['title'] = config['title'] - else: - info['title'] = self._og_search_title(webpage) + title = self._html_search_meta('fulltitle', webpage, default=None) + if not title or title == "c't": + title = self._search_regex( + r']+class="videoplayerjw"[^>]+data-title="([^"]+)"', + webpage, 'title') - formats = [] - for t, rs in config['formats'].items(): - if not rs or not hasattr(rs, 'items'): - self._downloader.report_warning( - 'formats: {0}: no resolutions'.format(t)) - continue - - for height_str, obj in rs.items(): - format_id = '{0}_{1}'.format(t, height_str) + yt_urls = YoutubeIE._extract_urls(webpage) + if yt_urls: + return self.playlist_from_matches(yt_urls, video_id, title, ie=YoutubeIE.ie_key()) - if not obj or not obj.get('url'): - self._downloader.report_warning( - 'formats: {0}: no url'.format(format_id)) - continue + container_id = self._search_regex( + r'
]+data-container="([0-9]+)"', + webpage, 'container ID') + sequenz_id = self._search_regex( + r'
]+data-sequenz="([0-9]+)"', + webpage, 'sequenz ID') - formats.append({ - 'url': obj['url'], - 'format_id': format_id, - 'height': self._int(height_str, 'height'), - }) + doc = self._download_xml( + 'http://www.heise.de/videout/feed', video_id, query={ + 'container': container_id, + 'sequenz': sequenz_id, + }) + formats = [] + for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'): + label = source_node.attrib['label'] + height = int_or_none(self._search_regex( + r'^(.*?_)?([0-9]+)p$', label, 'height', default=None)) + video_url = source_node.attrib['file'] + ext = determine_ext(video_url, '') + formats.append({ + 'url': video_url, + 'format_note': label, + 'format_id': '%s_%s' % (ext, label), + 'height': height, + }) self._sort_formats(formats) - info['formats'] = formats - return info + description = self._og_search_description( + webpage, default=None) or self._html_search_meta( + 'description', webpage) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or + self._og_search_thumbnail(webpage)), + 'timestamp': parse_iso8601( + self._html_search_meta('date', webpage)), + 'formats': formats, + }