From 80dcee5cd5cbe623a53e0c582e3e3ae170c63e8d Mon Sep 17 00:00:00 2001 From: remitamine Date: Sat, 31 Oct 2015 04:02:49 +0100 Subject: [PATCH] [eitb] fix info extraction --- youtube_dl/extractor/eitb.py | 65 ++++++++++++++++++++++++------------ 1 file changed, 44 insertions(+), 21 deletions(-) diff --git a/youtube_dl/extractor/eitb.py b/youtube_dl/extractor/eitb.py index 2cba82532..fc8f15544 100644 --- a/youtube_dl/extractor/eitb.py +++ b/youtube_dl/extractor/eitb.py @@ -1,39 +1,62 @@ # encoding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor -from .brightcove import BrightcoveIE -from ..utils import ExtractorError +from ..compat import compat_urllib_request +from ..utils import ( + int_or_none, + unified_strdate, +) class EitbIE(InfoExtractor): IE_NAME = 'eitb.tv' - _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/(?P\d+)/(?P\d+)' + _VALID_URL = r'https?://www\.eitb\.tv/(eu/bideoa|es/video)/[^/]+/\d+/(?P\d+)' _TEST = { - 'add_ie': ['Brightcove'], - 'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/2677100210001/2743577154001/lasa-y-zabala-30-anos/', + 'url': 'http://www.eitb.tv/es/video/60-minutos-60-minutos-2013-2014/4104995148001/4090227752001/lasa-y-zabala-30-anos/', 'md5': 'edf4436247185adee3ea18ce64c47998', 'info_dict': { - 'id': '2743577154001', + 'id': '4090227752001', 'ext': 'mp4', 'title': '60 minutos (Lasa y Zabala, 30 aƱos)', - # All videos from eitb has this description in the brightcove info - 'description': '.', - 'uploader': 'Euskal Telebista', + 'description': '', + 'duration': 3996760, + 'upload_date': '20131014', }, } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - chapter_id = mobj.group('chapter_id') - webpage = self._download_webpage(url, chapter_id) - bc_url = BrightcoveIE._extract_brightcove_url(webpage) - if bc_url is None: - raise ExtractorError('Could not extract the Brightcove url') - # The BrightcoveExperience object doesn't contain the video id, we set - # it manually - bc_url += '&%40videoPlayer={0}'.format(chapter_id) - return self.url_result(bc_url, BrightcoveIE.ie_key()) + video_id = self._match_id(url) + video_data = self._download_json('http://mam.eitb.eus/mam/REST/ServiceMultiweb/Video/MULTIWEBTV/%s/' % video_id, video_id)['web_media'][0] + + formats = [] + for rendition in video_data['RENDITIONS']: + formats.append({ + 'url': rendition['PMD_URL'], + 'width': int_or_none(rendition.get('FRAME_WIDTH')), + 'height': int_or_none(rendition.get('FRAME_HEIGHT')), + 'tbr': int_or_none(rendition.get('ENCODING_RATE')), + }) + + # TODO: parse f4m manifest + request = compat_urllib_request.Request( + 'http://mam.eitb.eus/mam/REST/ServiceMultiweb/DomainRestrictedSecurity/TokenAuth/', + headers={'Referer': url}) + token_data = self._download_json(request, video_id, fatal=False) + if token_data: + m3u8_formats = self._extract_m3u8_formats('%s?hdnts=%s' % (video_data['HLS_SURL'], token_data['token']), video_id, m3u8_id='hls', fatal=False) + if m3u8_formats: + formats.extend(m3u8_formats) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': video_data['NAME_ES'], + 'description': video_data.get('SHORT_DESC_ES'), + 'thumbnail': video_data.get('STILL_URL'), + 'duration': int_or_none(video_data.get('LENGTH')), + 'upload_date': unified_strdate(video_data.get('BROADCST_DATE')), + 'formats': formats, + }