X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/47d80ec0b18245caeb97018d4c1af18d0b5b972b..39393b81acfaf4045fb7f20454a0226f0dc9142e:/youtube_dl/extractor/dctp.py?ds=sidebyside diff --git a/youtube_dl/extractor/dctp.py b/youtube_dl/extractor/dctp.py index 14ba887..e700f8d 100644 --- a/youtube_dl/extractor/dctp.py +++ b/youtube_dl/extractor/dctp.py @@ -2,53 +2,104 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import unified_strdate +from ..compat import compat_str +from ..utils import ( + float_or_none, + int_or_none, + unified_timestamp, + url_or_none, +) class DctpTvIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(#/)?filme/(?P.+?)/$' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P[^/?#&]+)' + _TESTS = [{ + # 4x3 'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/', - 'md5': '174dd4a8a6225cf5655952f969cfbe24', + 'md5': '3ffbd1556c3fe210724d7088fad723e3', 'info_dict': { 'id': '95eaa4f33dad413aa17b4ee613cccc6c', 'display_id': 'videoinstallation-fuer-eine-kaufhausfassade', - 'ext': 'mp4', + 'ext': 'm4v', 'title': 'Videoinstallation für eine Kaufhausfassade', 'description': 'Kurzfilm', + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 71.24, + 'timestamp': 1302172322, 'upload_date': '20110407', - 'thumbnail': 're:^https?://.*\.jpg$', }, - } + }, { + # 16x9 + 'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/', + 'only_matching': True, + }] + + _BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com' def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - object_id = self._html_search_meta('DC.identifier', webpage) - - servers_json = self._download_json( - 'http://www.dctp.tv/elastic_streaming_client/get_streaming_server/', - video_id, note='Downloading server list') - server = servers_json[0]['server'] - m3u8_path = self._search_regex( - r'\'([^\'"]+/playlist\.m3u8)"', webpage, 'm3u8 path') - formats = self._extract_m3u8_formats( - 'http://%s%s' % (server, m3u8_path), video_id, ext='mp4', - entry_protocol='m3u8_native') - - title = self._og_search_title(webpage) - description = self._html_search_meta('DC.description', webpage) - upload_date = unified_strdate( - self._html_search_meta('DC.date.created', webpage)) - thumbnail = self._og_search_thumbnail(webpage) + display_id = self._match_id(url) + + version = self._download_json( + '%s/version.json' % self._BASE_URL, display_id, + 'Downloading version JSON') + + restapi_base = '%s/%s/restapi' % ( + self._BASE_URL, version['version_name']) + + info = self._download_json( + '%s/slugs/%s.json' % (restapi_base, display_id), display_id, + 'Downloading video info JSON') + + media = self._download_json( + '%s/media/%s.json' % (restapi_base, compat_str(info['object_id'])), + display_id, 'Downloading media JSON') + + uuid = media['uuid'] + title = media['title'] + is_wide = media.get('is_wide') + formats = [] + + def add_formats(suffix): + templ = 'https://%%s/%s_dctp_%s.m4v' % (uuid, suffix) + formats.extend([{ + 'format_id': 'hls-' + suffix, + 'url': templ % 'cdn-segments.dctp.tv' + '/playlist.m3u8', + 'protocol': 'm3u8_native', + }, { + 'format_id': 's3-' + suffix, + 'url': templ % 'completed-media.s3.amazonaws.com', + }, { + 'format_id': 'http-' + suffix, + 'url': templ % 'cdn-media.dctp.tv', + }]) + + add_formats('0500_' + ('16x9' if is_wide else '4x3')) + if is_wide: + add_formats('720p') + + thumbnails = [] + images = media.get('images') + if isinstance(images, list): + for image in images: + if not isinstance(image, dict): + continue + image_url = url_or_none(image.get('url')) + if not image_url: + continue + thumbnails.append({ + 'url': image_url, + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + }) return { - 'id': object_id, + 'id': uuid, + 'display_id': display_id, 'title': title, + 'alt_title': media.get('subtitle'), + 'description': media.get('description') or media.get('teaser'), + 'timestamp': unified_timestamp(media.get('created')), + 'duration': float_or_none(media.get('duration_in_ms'), scale=1000), + 'thumbnails': thumbnails, 'formats': formats, - 'display_id': video_id, - 'description': description, - 'upload_date': upload_date, - 'thumbnail': thumbnail, }