X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a070911bf98c751e4425008f1335cc57c86d30fc..432fd38466ba4fc2f31597488104ed0e729149a6:/youtube_dl/extractor/vevo.py diff --git a/youtube_dl/extractor/vevo.py b/youtube_dl/extractor/vevo.py index 43f6b02..4c0de35 100644 --- a/youtube_dl/extractor/vevo.py +++ b/youtube_dl/extractor/vevo.py @@ -1,14 +1,15 @@ from __future__ import unicode_literals import re -import xml.etree.ElementTree from .common import InfoExtractor from ..compat import ( + compat_etree_fromstring, compat_urllib_request, ) from ..utils import ( ExtractorError, + int_or_none, ) @@ -96,7 +97,7 @@ class VevoIE(InfoExtractor): if last_version['version'] == -1: raise ExtractorError('Unable to extract last version of the video') - renditions = xml.etree.ElementTree.fromstring(last_version['data']) + renditions = compat_etree_fromstring(last_version['data']) formats = [] # Already sorted from worst to best quality for rend in renditions.findall('rendition'): @@ -113,7 +114,7 @@ class VevoIE(InfoExtractor): def _formats_from_smil(self, smil_xml): formats = [] - smil_doc = xml.etree.ElementTree.fromstring(smil_xml.encode('utf-8')) + smil_doc = compat_etree_fromstring(smil_xml.encode('utf-8')) els = smil_doc.findall('.//{http://www.w3.org/2001/SMIL20/Language}video') for el in els: src = el.attrib['src'] @@ -192,9 +193,29 @@ class VevoIE(InfoExtractor): # Download via HLS API formats.extend(self._download_api_formats(video_id)) + # Download SMIL + smil_blocks = sorted(( + f for f in video_info['videoVersions'] + if f['sourceType'] == 13), + key=lambda f: f['version']) + smil_url = '%s/Video/V2/VFILE/%s/%sr.smil' % ( + self._SMIL_BASE_URL, video_id, video_id.lower()) + if smil_blocks: + smil_url_m = self._search_regex( + r'url="([^"]+)"', smil_blocks[-1]['data'], 'SMIL URL', + default=None) + if smil_url_m is not None: + smil_url = smil_url_m + if smil_url: + smil_xml = self._download_webpage( + smil_url, video_id, 'Downloading SMIL info', fatal=False) + if smil_xml: + formats.extend(self._formats_from_smil(smil_xml)) + self._sort_formats(formats) - timestamp_ms = int(self._search_regex( - r'/Date\((\d+)\)/', video_info['launchDate'], 'launch date')) + timestamp_ms = int_or_none(self._search_regex( + r'/Date\((\d+)\)/', + video_info['launchDate'], 'launch date', fatal=False)) return { 'id': video_id,