X-Git-Url: http://git.bitcoin.ninja/index.cgi?p=youtube-dl;a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fscreenwavemedia.py;h=7d77e8825d7420185ed1f9f707efb3462f19cab1;hp=cf9b8e0de984b73c8a5230d4b522740b5ee61c84;hb=dcdb292fddc82ae11f4c0b647815a45c88a6b6d5;hpb=69f491f14e132a54e8ca7de291183dced64d461d diff --git a/youtube_dl/extractor/screenwavemedia.py b/youtube_dl/extractor/screenwavemedia.py index cf9b8e0de..7d77e8825 100644 --- a/youtube_dl/extractor/screenwavemedia.py +++ b/youtube_dl/extractor/screenwavemedia.py @@ -1,194 +1,146 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals import re from .common import InfoExtractor from ..utils import ( - ExtractorError, - month_by_name, int_or_none, + unified_strdate, + js_to_json, ) + class ScreenwaveMediaIE(InfoExtractor): - _VALID_URL = r'(?:http://)?(?' \ - r':(?Pplayer\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=(?P.+))' \ - r'|(?P(?:www\.)?cinemassacre\.com/(?P[0-9]{4})/(?P[0-9]{2})/(?P[0-9]{2})/(?P[^?#/]+))' \ - r'|(?P(?:www\.)?teamfourstar\.com/video/(?P[a-z0-9\-]+)/?)' \ - r')' - - _TESTS = [ - { - 'url': 'http://cinemassacre.com/2012/11/10/avgn-the-movie-trailer/', - 'md5': 'fde81fbafaee331785f58cd6c0d46190', - 'info_dict': { - 'id': 'Cinemasssacre-19911', - 'ext': 'mp4', - 'upload_date': '20121110', - 'title': '“Angry Video Game Nerd: The Movie” – Trailer', - 'description': 'md5:fb87405fcb42a331742a0dce2708560b', - }, - }, - { - 'url': 'http://cinemassacre.com/2013/10/02/the-mummys-hand-1940', - 'md5': 'd72f10cd39eac4215048f62ab477a511', - 'info_dict': { - 'id': 'Cinemasssacre-521be8ef82b16', - 'ext': 'mp4', - 'upload_date': '20131002', - 'title': 'The Mummy’s Hand (1940)', - }, - } - ] + _VALID_URL = r'(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=(?P[A-Za-z0-9-]+)' + EMBED_PATTERN = r'src=(["\'])(?P(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=.+?)\1' + _TESTS = [{ + 'url': 'http://player.screenwavemedia.com/play/play.php?playerdiv=videoarea&companiondiv=squareAd&id=Cinemassacre-19911', + 'only_matching': True, + }] - def _cinemassacre_get_info(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('cm_display_id') + def _real_extract(self, url): + video_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - video_date = mobj.group('cm_date_Y') + mobj.group('cm_date_m') + mobj.group('cm_date_d') - mobj = re.search(r'src="(?Phttp://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', webpage) - if not mobj: - raise ExtractorError('Can\'t extract embed url and video id') - playerdata_url = mobj.group('embed_url') + playerdata = self._download_webpage( + 'http://player.screenwavemedia.com/player.php?id=%s' % video_id, + video_id, 'Downloading player webpage') - video_title = self._html_search_regex( - r'(?P<title>.+?)\|', webpage, 'title') - video_description = self._html_search_regex( - r'<div class="entry-content">(?P<description>.+?)</div>', - webpage, 'description', flags=re.DOTALL, fatal=False) - video_thumbnail = self._og_search_thumbnail(webpage) + vidtitle = self._search_regex( + r'\'vidtitle\'\s*:\s*"([^"]+)"', playerdata, 'vidtitle').replace('\\/', '/') + + playerconfig = self._download_webpage( + 'http://player.screenwavemedia.com/player.js', + video_id, 'Downloading playerconfig webpage') + + videoserver = self._search_regex(r'SWMServer\s*=\s*"([\d\.]+)"', playerdata, 'videoserver') + + sources = self._parse_json( + js_to_json( + re.sub( + r'(?s)/\*.*?\*/', '', + self._search_regex( + r'sources\s*:\s*(\[[^\]]+?\])', playerconfig, + 'sources', + ).replace( + "' + thisObj.options.videoserver + '", + videoserver + ).replace( + "' + playerVidId + '", + video_id + ) + ) + ), + video_id, fatal=False + ) + + # Fallback to hardcoded sources if JS changes again + if not sources: + self.report_warning('Falling back to a hardcoded list of streams') + sources = [{ + 'file': 'http://%s/vod/%s_%s.mp4' % (videoserver, video_id, format_id), + 'type': 'mp4', + 'label': format_label, + } for format_id, format_label in ( + ('low', '144p Low'), ('med', '160p Med'), ('high', '360p High'), ('hd1', '720p HD1'))] + sources.append({ + 'file': 'http://%s/vod/smil:%s.smil/playlist.m3u8' % (videoserver, video_id), + 'type': 'hls', + }) + + formats = [] + for source in sources: + file_ = source.get('file') + if not file_: + continue + if source.get('type') == 'hls': + formats.extend(self._extract_m3u8_formats(file_, video_id, ext='mp4')) + else: + format_id = self._search_regex( + r'_(.+?)\.[^.]+$', file_, 'format id', default=None) + if not self._is_valid_url(file_, video_id, format_id or 'video'): + continue + format_label = source.get('label') + height = int_or_none(self._search_regex( + r'^(\d+)[pP]', format_label, 'height', default=None)) + formats.append({ + 'url': file_, + 'format_id': format_id, + 'format': format_label, + 'ext': source.get('type'), + 'height': height, + }) + self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id')) return { - 'title': video_title, - 'description': video_description, - 'upload_date': video_date, - 'thumbnail': video_thumbnail, - '_embed_url': playerdata_url, + 'id': video_id, + 'title': vidtitle, + 'formats': formats, } - def _teamfourstar_get_info(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('tfs_display_id') + +class TeamFourIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?teamfourstar\.com/video/(?P<id>[a-z0-9\-]+)/?' + _TEST = { + 'url': 'http://teamfourstar.com/video/a-moment-with-tfs-episode-4/', + 'info_dict': { + 'id': 'TeamFourStar-5292a02f20bfa', + 'ext': 'mp4', + 'upload_date': '20130401', + 'description': 'Check out this and more on our website: http://teamfourstar.com\nTFS Store: http://sharkrobot.com/team-four-star\nFollow on Twitter: http://twitter.com/teamfourstar\nLike on FB: http://facebook.com/teamfourstar', + 'title': 'A Moment With TFS Episode 4', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + } + + def _real_extract(self, url): + display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - mobj = re.search(r'src="(?P<embed_url>http://player\.screenwavemedia\.com/play/[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', webpage) - if not mobj: - raise ExtractorError('Can\'t extract embed url and video id') - playerdata_url = mobj.group('embed_url') + playerdata_url = self._search_regex( + r'src="(http://player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?[^"]*\bid=.+?)"', + webpage, 'player data URL') video_title = self._html_search_regex( - r'<div class="heroheadingtitle">(?P<title>.+?)</div>', webpage, 'title') - video_date = self._html_search_regex( - r'<div class="heroheadingdate">(?P<date>.+?)</div>', webpage, 'date') - mobj = re.match('(?P<month>[A-Z][a-z]+) (?P<day>\d+), (?P<year>\d+)', video_date) - video_date = '%04u%02u%02u' % (int(mobj.group('year')), month_by_name(mobj.group('month')), int(mobj.group('day'))) + r'<div class="heroheadingtitle">(?P<title>.+?)</div>', + webpage, 'title') + video_date = unified_strdate(self._html_search_regex( + r'<div class="heroheadingdate">(?P<date>.+?)</div>', + webpage, 'date', fatal=False)) video_description = self._html_search_regex( - r'<div class="postcontent">(?P<description>.+?)</div>', webpage, 'description', flags=re.DOTALL) + r'(?s)<div class="postcontent">(?P<description>.+?)</div>', + webpage, 'description', fatal=False) video_thumbnail = self._og_search_thumbnail(webpage) return { + '_type': 'url_transparent', + 'display_id': display_id, 'title': video_title, 'description': video_description, 'upload_date': video_date, 'thumbnail': video_thumbnail, - '_embed_url': playerdata_url, + 'url': playerdata_url, } - - def _screenwavemedia_get_info(self, url): - mobj = re.match(self._VALID_URL, url) - if not mobj: - raise ExtractorError('Can\'t extract embed url and video id') - video_id = mobj.group('video_id') - - playerdata = self._download_webpage(url, video_id, 'Downloading player webpage') - - vidtitle = self._search_regex( - r'\'vidtitle\'\s*:\s*"([^\']+)"', playerdata, 'vidtitle').replace('\\/', '/') - vidurl = self._search_regex( - r'\'vidurl\'\s*:\s*"([^\']+)"', playerdata, 'vidurl').replace('\\/', '/') - pageurl = self._search_regex( - r'\'pageurl\'\s*:\s*"([^\']+)"', playerdata, 'pageurl', fatal=False).replace('\\/', '/') - - videolist_url = None - - mobj = re.search(r"'videoserver'\s*:\s*'(?P<videoserver>[^']+)'", playerdata) - if mobj: - videoserver = mobj.group('videoserver') - mobj = re.search(r'\'vidid\'\s*:\s*"(?P<vidid>[^\']+)"', playerdata) - vidid = mobj.group('vidid') if mobj else video_id - videolist_url = 'http://%s/vod/smil:%s.smil/jwplayer.smil' % (videoserver, vidid) - else: - mobj = re.search(r"file\s*:\s*'(?P<smil>http.+?/jwplayer\.smil)'", playerdata) - if mobj: - videolist_url = mobj.group('smil') - - if videolist_url: - videolist = self._download_xml(videolist_url, video_id, 'Downloading videolist XML') - formats = [] - baseurl = vidurl[:vidurl.rfind('/') + 1] - for video in videolist.findall('.//video'): - src = video.get('src') - if not src: - continue - file_ = src.partition(':')[-1] - width = int_or_none(video.get('width')) - height = int_or_none(video.get('height')) - bitrate = int_or_none(video.get('system-bitrate')) - format = { - 'url': baseurl + file_, - 'format_id': src.rpartition('.')[0].rpartition('_')[-1], - } - if width or height: - format.update({ - 'tbr': bitrate // 1000 if bitrate else None, - 'width': width, - 'height': height, - }) - else: - format.update({ - 'abr': bitrate // 1000 if bitrate else None, - 'vcodec': 'none', - }) - formats.append(format) - self._sort_formats(formats) - else: - formats = [{ - 'url': vidurl, - }] - - return { - 'id': video_id, - 'title': vidtitle, - 'formats': formats, - '_episode_page': pageurl, - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - - swm_info = None - site_info = None - - if mobj.group('generic'): - swm_info = self._screenwavemedia_get_info(url) - url = swm_info['_episode_page'] - mobj = re.match(self._VALID_URL, url) - - if mobj: - if mobj.group('cinemassacre'): - site_info = self._cinemassacre_get_info(url) - elif mobj.group('teamfourstar'): - site_info = self._teamfourstar_get_info(url) - - if not swm_info: - if site_info: - swm_info = self._screenwavemedia_get_info(site_info['_embed_url']) - - if not swm_info: - raise ExtractorError("Failed to extract metadata for this URL") - - if site_info: - swm_info.update(site_info) - - return swm_info