X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/87a0165ca7e39af4dacb7ec637063b2cd35ae40b..725f8b2d4cc43b8401946dfed69d22cfe40810fc:/youtube_dl/extractor/crackle.py?ds=sidebyside diff --git a/youtube_dl/extractor/crackle.py b/youtube_dl/extractor/crackle.py index cc68f1c..fc014f8 100644 --- a/youtube_dl/extractor/crackle.py +++ b/youtube_dl/extractor/crackle.py @@ -1,29 +1,45 @@ # coding: utf-8 from __future__ import unicode_literals, division +import re + from .common import InfoExtractor -from ..utils import int_or_none +from ..compat import ( + compat_str, + compat_HTTPError, +) +from ..utils import ( + determine_ext, + float_or_none, + int_or_none, + parse_age_limit, + parse_duration, + ExtractorError +) class CrackleIE(InfoExtractor): - _VALID_URL = r'(?:crackle:|https?://(?:www\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P\d+)' + _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P\d+)' _TEST = { - 'url': 'http://www.crackle.com/comedians-in-cars-getting-coffee/2498934', + # geo restricted to CA + 'url': 'https://www.crackle.com/andromeda/2502343', 'info_dict': { - 'id': '2498934', + 'id': '2502343', 'ext': 'mp4', - 'title': 'Everybody Respects A Bloody Nose', - 'description': 'Jerry is kaffeeklatsching in L.A. with funnyman J.B. Smoove (Saturday Night Live, Real Husbands of Hollywood). They’re headed for brew at 10 Speed Coffee in a 1964 Studebaker Avanti.', - 'thumbnail': 're:^https?://.*\.jpg', - 'duration': 906, - 'series': 'Comedians In Cars Getting Coffee', - 'season_number': 8, - 'episode_number': 4, - 'subtitles': { - 'en-US': [{ - 'ext': 'ttml', - }] - }, + 'title': 'Under The Night', + 'description': 'md5:d2b8ca816579ae8a7bf28bfff8cefc8a', + 'duration': 2583, + 'view_count': int, + 'average_rating': 0, + 'age_limit': 14, + 'genre': 'Action, Sci-Fi', + 'creator': 'Allan Kroeker', + 'artist': 'Keith Hamilton Cobb, Kevin Sorbo, Lisa Ryder, Lexa Doig, Robert Hewitt Wolfe', + 'release_year': 2000, + 'series': 'Andromeda', + 'episode': 'Under The Night', + 'season_number': 1, + 'episode_number': 1, }, 'params': { # m3u8 download @@ -31,76 +47,118 @@ class CrackleIE(InfoExtractor): } } - # extracted from http://legacyweb-us.crackle.com/flash/ReferrerRedirect.ashx - _THUMBNAIL_TEMPLATE = 'http://images-us-am.crackle.com/%stnl_1920x1080.jpg?ts=20140107233116?c=635333335057637614' - _MEDIA_FILE_SLOTS = { - 'c544.flv': { - 'width': 544, - 'height': 306, - }, - '360p.mp4': { - 'width': 640, - 'height': 360, - }, - '480p.mp4': { - 'width': 852, - 'height': 478, - }, - '480p_1mbps.mp4': { - 'width': 852, - 'height': 478, - }, - } - def _real_extract(self, url): video_id = self._match_id(url) - config_doc = self._download_xml( - 'http://legacyweb-us.crackle.com/flash/QueryReferrer.ashx?site=16', - video_id, 'Downloading config') - - item = self._download_xml( - 'http://legacyweb-us.crackle.com/app/revamp/vidwallcache.aspx?flags=-1&fm=%s' % video_id, - video_id).find('i') - title = item.attrib['t'] - - subtitles = {} - formats = self._extract_m3u8_formats( - 'http://content.uplynk.com/ext/%s/%s.m3u8' % (config_doc.attrib['strUplynkOwnerId'], video_id), - video_id, 'mp4', m3u8_id='hls', fatal=None) - thumbnail = None - path = item.attrib.get('p') - if path: - thumbnail = self._THUMBNAIL_TEMPLATE % path - http_base_url = 'http://ahttp.crackle.com/' + path - for mfs_path, mfs_info in self._MEDIA_FILE_SLOTS.items(): - formats.append({ - 'url': http_base_url + mfs_path, - 'format_id': 'http-' + mfs_path.split('.')[0], - 'width': mfs_info['width'], - 'height': mfs_info['height'], - }) - for cc in item.findall('cc'): - locale = cc.attrib.get('l') - v = cc.attrib.get('v') - if locale and v: - if locale not in subtitles: - subtitles[locale] = [] - subtitles[locale] = [{ - 'url': '%s/%s%s_%s.xml' % (config_doc.attrib['strSubtitleServer'], path, locale, v), - 'ext': 'ttml', - }] - self._sort_formats(formats, ('width', 'height', 'tbr', 'format_id')) - - return { - 'id': video_id, - 'title': title, - 'description': item.attrib.get('d'), - 'duration': int(item.attrib.get('r'), 16) / 1000 if item.attrib.get('r') else None, - 'series': item.attrib.get('sn'), - 'season_number': int_or_none(item.attrib.get('se')), - 'episode_number': int_or_none(item.attrib.get('ep')), - 'thumbnail': thumbnail, - 'subtitles': subtitles, - 'formats': formats, - } + country_code = self._downloader.params.get('geo_bypass_country', None) + countries = [country_code] if country_code else ( + 'US', 'AU', 'CA', 'AS', 'FM', 'GU', 'MP', 'PR', 'PW', 'MH', 'VI') + + last_e = None + + for country in countries: + try: + media = self._download_json( + 'https://web-api-us.crackle.com/Service.svc/details/media/%s/%s' + % (video_id, country), video_id, + 'Downloading media JSON as %s' % country, + 'Unable to download media JSON', query={ + 'disableProtocols': 'true', + 'format': 'json' + }) + except ExtractorError as e: + # 401 means geo restriction, trying next country + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: + last_e = e + continue + raise + + media_urls = media.get('MediaURLs') + if not media_urls or not isinstance(media_urls, list): + continue + + title = media['Title'] + + formats = [] + for e in media['MediaURLs']: + if e.get('UseDRM') is True: + continue + format_url = e.get('Path') + if not format_url or not isinstance(format_url, compat_str): + continue + ext = determine_ext(format_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + format_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + elif ext == 'mpd': + formats.extend(self._extract_mpd_formats( + format_url, video_id, mpd_id='dash', fatal=False)) + self._sort_formats(formats) + + description = media.get('Description') + duration = int_or_none(media.get( + 'DurationInSeconds')) or parse_duration(media.get('Duration')) + view_count = int_or_none(media.get('CountViews')) + average_rating = float_or_none(media.get('UserRating')) + age_limit = parse_age_limit(media.get('Rating')) + genre = media.get('Genre') + release_year = int_or_none(media.get('ReleaseYear')) + creator = media.get('Directors') + artist = media.get('Cast') + + if media.get('MediaTypeDisplayValue') == 'Full Episode': + series = media.get('ShowName') + episode = title + season_number = int_or_none(media.get('Season')) + episode_number = int_or_none(media.get('Episode')) + else: + series = episode = season_number = episode_number = None + + subtitles = {} + cc_files = media.get('ClosedCaptionFiles') + if isinstance(cc_files, list): + for cc_file in cc_files: + if not isinstance(cc_file, dict): + continue + cc_url = cc_file.get('Path') + if not cc_url or not isinstance(cc_url, compat_str): + continue + lang = cc_file.get('Locale') or 'en' + subtitles.setdefault(lang, []).append({'url': cc_url}) + + thumbnails = [] + images = media.get('Images') + if isinstance(images, list): + for image_key, image_url in images.items(): + mobj = re.search(r'Img_(\d+)[xX](\d+)', image_key) + if not mobj: + continue + thumbnails.append({ + 'url': image_url, + 'width': int(mobj.group(1)), + 'height': int(mobj.group(2)), + }) + + return { + 'id': video_id, + 'title': title, + 'description': description, + 'duration': duration, + 'view_count': view_count, + 'average_rating': average_rating, + 'age_limit': age_limit, + 'genre': genre, + 'creator': creator, + 'artist': artist, + 'release_year': release_year, + 'series': series, + 'episode': episode, + 'season_number': season_number, + 'episode_number': episode_number, + 'thumbnails': thumbnails, + 'subtitles': subtitles, + 'formats': formats, + } + + raise last_e