diff options
Diffstat (limited to 'youtube_dl/extractor/arte.py')
| -rw-r--r-- | youtube_dl/extractor/arte.py | 145 |
1 files changed, 45 insertions, 100 deletions
diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py index 979481b21..646377e4b 100644 --- a/youtube_dl/extractor/arte.py +++ b/youtube_dl/extractor/arte.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..utils import ( @@ -19,114 +18,41 @@ from ..utils import ( # is different for each one. The videos usually expire in 7 days, so we can't # add tests. -class ArteTvIE(InfoExtractor): - _VIDEOS_URL = r'(?:http://)?videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html' - _LIVEWEB_URL = r'(?:http://)?liveweb\.arte\.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)' - _LIVE_URL = r'index-[0-9]+\.html$' +class ArteTvIE(InfoExtractor): + _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html' IE_NAME = 'arte.tv' - @classmethod - def suitable(cls, url): - return any(re.match(regex, url) for regex in (cls._VIDEOS_URL, cls._LIVEWEB_URL)) - - # TODO implement Live Stream - # from ..utils import compat_urllib_parse - # def extractLiveStream(self, url): - # video_lang = url.split('/')[-4] - # info = self.grep_webpage( - # url, - # r'src="(.*?/videothek_js.*?\.js)', - # 0, - # [ - # (1, 'url', 'Invalid URL: %s' % url) - # ] - # ) - # http_host = url.split('/')[2] - # next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url'))) - # info = self.grep_webpage( - # next_url, - # r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' + - # '(http://.*?\.swf).*?' + - # '(rtmp://.*?)\'', - # re.DOTALL, - # [ - # (1, 'path', 'could not extract video path: %s' % url), - # (2, 'player', 'could not extract video player: %s' % url), - # (3, 'url', 'could not extract video url: %s' % url) - # ] - # ) - # video_url = '%s/%s' % (info.get('url'), info.get('path')) - def _real_extract(self, url): - mobj = re.match(self._VIDEOS_URL, url) - if mobj is not None: - id = mobj.group('id') - lang = mobj.group('lang') - return self._extract_video(url, id, lang) - - mobj = re.match(self._LIVEWEB_URL, url) - if mobj is not None: - name = mobj.group('name') - lang = mobj.group('lang') - return self._extract_liveweb(url, name, lang) - - if re.search(self._LIVE_URL, url) is not None: - raise ExtractorError('Arte live streams are not yet supported, sorry') - # self.extractLiveStream(url) - # return - - raise ExtractorError('No video found') - - def _extract_video(self, url, video_id, lang): - """Extract from videos.arte.tv""" + mobj = re.match(self._VALID_URL, url) + lang = mobj.group('lang') + video_id = mobj.group('id') + ref_xml_url = url.replace('/videos/', '/do_delegate/videos/') ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml') ref_xml_doc = self._download_xml( ref_xml_url, video_id, note='Downloading metadata') config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang) config_xml_url = config_node.attrib['ref'] - config_xml = self._download_webpage( + config = self._download_xml( config_xml_url, video_id, note='Downloading configuration') - video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml)) - def _key(m): - quality = m.group('quality') - if quality == 'hd': - return 2 - else: - return 1 - # We pick the best quality - video_urls = sorted(video_urls, key=_key) - video_url = list(video_urls)[-1].group('url') - - title = self._html_search_regex(r'<name>(.*?)</name>', config_xml, 'title') - thumbnail = self._html_search_regex(r'<firstThumbnailUrl>(.*?)</firstThumbnailUrl>', - config_xml, 'thumbnail') - return {'id': video_id, - 'title': title, - 'thumbnail': thumbnail, - 'url': video_url, - 'ext': 'flv', - } - - def _extract_liveweb(self, url, name, lang): - """Extract form http://liveweb.arte.tv/""" - webpage = self._download_webpage(url, name) - video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, 'event id') - config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id, - video_id, 'Downloading information') - event_doc = config_doc.find('event') - url_node = event_doc.find('video').find('urlHd') - if url_node is None: - url_node = event_doc.find('urlSd') - - return {'id': video_id, - 'title': event_doc.find('name%s' % lang.capitalize()).text, - 'url': url_node.text.replace('MP4', 'mp4'), - 'ext': 'flv', - 'thumbnail': self._og_search_thumbnail(webpage), - } + formats = [{ + 'forma_id': q.attrib['quality'], + 'url': q.text, + 'ext': 'flv', + 'quality': 2 if q.attrib['quality'] == 'hd' else 1, + } for q in config.findall('./urls/url')] + self._sort_formats(formats) + + title = config.find('.//name').text + thumbnail = config.find('.//firstThumbnailUrl').text + return { + 'id': video_id, + 'title': title, + 'thumbnail': thumbnail, + 'formats': formats, + } class ArteTVPlus7IE(InfoExtractor): @@ -152,9 +78,7 @@ class ArteTVPlus7IE(InfoExtractor): return self._extract_from_json_url(json_url, video_id, lang) def _extract_from_json_url(self, json_url, video_id, lang): - json_info = self._download_webpage(json_url, video_id, 'Downloading info json') - self.report_extraction(video_id) - info = json.loads(json_info) + info = self._download_json(json_url, video_id) player_info = info['videoJsonPlayer'] info_dict = { @@ -176,6 +100,8 @@ class ArteTVPlus7IE(InfoExtractor): l = 'F' elif lang == 'de': l = 'A' + else: + l = lang regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l] return any(re.match(r, f['versionCode']) for r in regexes) # Some formats may not be in the same language as the url @@ -305,3 +231,22 @@ class ArteTVConcertIE(ArteTVPlus7IE): 'description': 'md5:486eb08f991552ade77439fe6d82c305', }, } + + +class ArteTVEmbedIE(ArteTVPlus7IE): + IE_NAME = 'arte.tv:embed' + _VALID_URL = r'''(?x) + http://www\.arte\.tv + /playerv2/embed\.php\?json_url= + (?P<json_url> + http://arte\.tv/papi/tvguide/videos/stream/player/ + (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]* + ) + ''' + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + lang = mobj.group('lang') + json_url = mobj.group('json_url') + return self._extract_from_json_url(json_url, video_id, lang) |