X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/87a0165ca7e39af4dacb7ec637063b2cd35ae40b..8c72ec3993d7183e9c1fcd2cab445a3196ad8723:/youtube_dl/extractor/cbs.py diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index 58f258c..4a19a73 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -2,6 +2,7 @@ from __future__ import unicode_literals from .theplatform import ThePlatformFeedIE from ..utils import ( + ExtractorError, int_or_none, find_xpath_attr, xpath_element, @@ -12,13 +13,17 @@ from ..utils import ( class CBSBaseIE(ThePlatformFeedIE): def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'): - closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL') - return { - 'en': [{ - 'ext': 'ttml', - 'url': closed_caption_e.attrib['value'], - }] - } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else [] + subtitles = {} + for k, ext in [('sMPTE-TTCCURL', 'tt'), ('ClosedCaptionURL', 'ttml'), ('webVTTCaptionURL', 'vtt')]: + cc_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', k) + if cc_e is not None: + cc_url = cc_e.get('value') + if cc_url: + subtitles.setdefault(subtitles_lang, []).append({ + 'ext': ext, + 'url': cc_url, + }) + return subtitles class CBSIE(CBSBaseIE): @@ -49,21 +54,22 @@ class CBSIE(CBSBaseIE): 'only_matching': True, }] - def _extract_video_info(self, content_id): + def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517): items_data = self._download_xml( 'http://can.cbs.com/thunder/player/videoPlayerService.php', - content_id, query={'partner': 'cbs', 'contentId': content_id}) + content_id, query={'partner': site, 'contentId': content_id}) video_data = xpath_element(items_data, './/item') title = xpath_text(video_data, 'videoTitle', 'title', True) - tp_path = 'dJ5BDC/media/guid/2198311517/%s' % content_id + tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id) tp_release_url = 'http://link.theplatform.com/s/' + tp_path asset_types = [] subtitles = {} formats = [] + last_e = None for item in items_data.findall('.//item'): asset_type = xpath_text(item, 'assetType') - if not asset_type or asset_type in asset_types: + if not asset_type or asset_type in asset_types or 'HLS_FPS' in asset_type or 'DASH_CENC' in asset_type: continue asset_types.append(asset_type) query = { @@ -74,11 +80,17 @@ class CBSIE(CBSBaseIE): query['formats'] = 'MPEG4,M3U' elif asset_type in ('RTMP', 'WIFI', '3G'): query['formats'] = 'MPEG4,FLV' - tp_formats, tp_subtitles = self._extract_theplatform_smil( - update_url_query(tp_release_url, query), content_id, - 'Downloading %s SMIL data' % asset_type) + try: + tp_formats, tp_subtitles = self._extract_theplatform_smil( + update_url_query(tp_release_url, query), content_id, + 'Downloading %s SMIL data' % asset_type) + except ExtractorError as e: + last_e = e + continue formats.extend(tp_formats) subtitles = self._merge_subtitles(subtitles, tp_subtitles) + if last_e and not formats: + raise last_e self._sort_formats(formats) info = self._extract_theplatform_metadata(tp_path, content_id)