X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fwat.py;h=46b4d91330c50b5fb27917a42ef69d07a1a36ad9;hb=cc7fec5818254f4679896823c7de9d17f50201ca;hp=aba4159e9b856d627bc5fe4cff1f45a8259ca432;hpb=705f6f35bc13dd8d2ad3df9b1d22f6cad362afcc;p=youtube-dl diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py index aba4159e9..46b4d9133 100644 --- a/youtube_dl/extractor/wat.py +++ b/youtube_dl/extractor/wat.py @@ -1,48 +1,128 @@ -import json +# coding: utf-8 +from __future__ import unicode_literals + import re +import hashlib from .common import InfoExtractor - -from ..utils import ( - compat_urllib_parse, -) +from ..utils import unified_strdate class WatIE(InfoExtractor): - _VALID_URL=r'http://www.wat.tv/.*-(?P.*?)_.*?.html' + _VALID_URL = r'http://www\.wat\.tv/video/(?P.*)-(?P.*?)_.*?\.html' IE_NAME = 'wat.tv' - _TEST = { - u'url': u'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html', - u'file': u'6bv55.mp4', - u'md5': u'0a4fe7870f31eaeabb5e25fd8da8414a', - u'info_dict': { - u"title": u"World War Z - Philadelphia VOST" - } - } + _TESTS = [ + { + 'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html', + 'md5': 'ce70e9223945ed26a8056d413ca55dc9', + 'info_dict': { + 'id': '11713067', + 'display_id': 'soupe-figues-l-orange-aux-epices', + 'ext': 'mp4', + 'title': 'Soupe de figues à l\'orange et aux épices', + 'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.', + 'upload_date': '20140819', + 'duration': 120, + }, + }, + { + 'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html', + 'md5': 'fbc84e4378165278e743956d9c1bf16b', + 'info_dict': { + 'id': '11713075', + 'display_id': 'gregory-lemarchal-voix-ange', + 'ext': 'mp4', + 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)', + 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3', + 'upload_date': '20140816', + 'duration': 2910, + }, + }, + ] + + def download_video_info(self, real_id): + # 'contentv4' is used in the website, but it also returns the related + # videos, we don't need them + info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id) + return info['media'] def _real_extract(self, url): + def real_id_for_chapter(chapter): + return chapter['tc_start'].split('-')[0] mobj = re.match(self._VALID_URL, url) - short_id = mobj.group('shortID') - - player_data = compat_urllib_parse.urlencode({'shortVideoId': short_id, - 'html5': '1'}) - player_info = self._download_webpage('http://www.wat.tv/player?' + player_data, - short_id, u'Downloading player info') - player = json.loads(player_info)['player'] - html5_player = self._html_search_regex(r'iframe src="(.*?)"', player, - 'html5 player') - player_webpage = self._download_webpage(html5_player, short_id, - u'Downloading player webpage') - - video_url = self._search_regex(r'urlhtml5 : "(.*?)"', player_webpage, - 'video url') - title = self._search_regex(r'contentTitle : "(.*?)"', player_webpage, - 'title') - thumbnail = self._search_regex(r'previewMedia : "(.*?)"', player_webpage, - 'thumbnail') - return {'id': short_id, + short_id = mobj.group('short_id') + display_id = mobj.group('display_id') + webpage = self._download_webpage(url, display_id or short_id) + real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id') + + video_info = self.download_video_info(real_id) + + geo_list = video_info.get('geoList') + country = geo_list[0] if geo_list else '' + + chapters = video_info['chapters'] + first_chapter = chapters[0] + files = video_info['files'] + first_file = files[0] + + if real_id_for_chapter(first_chapter) != real_id: + self.to_screen('Multipart video detected') + chapter_urls = [] + for chapter in chapters: + chapter_id = real_id_for_chapter(chapter) + # Yes, when we this chapter is processed by WatIE, + # it will download the info again + chapter_info = self.download_video_info(chapter_id) + chapter_urls.append(chapter_info['url']) + entries = [self.url_result(chapter_url) for chapter_url in chapter_urls] + return self.playlist_result(entries, real_id, video_info['title']) + + upload_date = None + if 'date_diffusion' in first_chapter: + upload_date = unified_strdate(first_chapter['date_diffusion']) + # Otherwise we can continue and extract just one part, we have to use + # the short id for getting the video url + + formats = [{ + 'url': 'http://wat.tv/get/android5/%s.mp4' % real_id, + 'format_id': 'Mobile', + }] + + fmts = [('SD', 'web')] + if first_file.get('hasHD'): + fmts.append(('HD', 'webhd')) + + def compute_token(param): + timestamp = '%08x' % int(self._download_webpage( + 'http://www.wat.tv/servertime', real_id, + 'Downloading server time').split('|')[0]) + magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564' + return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp) + + for fmt in fmts: + webid = '/%s/%s' % (fmt[1], real_id) + video_url = self._download_webpage( + 'http://www.wat.tv/get%s?token=%s&getURL=1&country=%s' % (webid, compute_token(webid), country), + real_id, + 'Downloding %s video URL' % fmt[0], + 'Failed to download %s video URL' % fmt[0], + False) + if not video_url: + continue + formats.append({ 'url': video_url, 'ext': 'mp4', - 'title': title, - 'thumbnail': thumbnail, - } + 'format_id': fmt[0], + }) + + return { + 'id': real_id, + 'display_id': display_id, + 'title': first_chapter['title'], + 'thumbnail': first_chapter['preview'], + 'description': first_chapter['description'], + 'view_count': video_info['views'], + 'upload_date': upload_date, + 'duration': first_file['duration'], + 'formats': formats, + }