X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/feb5020b37d7d3ba4005a8bac6f4efece4ce4b8c..bcccf25821fdedf4d89a8f8847cfb5660a10da94:/youtube_dl/extractor/wat.py diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py index 0d1302c..de7d6b5 100644 --- a/youtube_dl/extractor/wat.py +++ b/youtube_dl/extractor/wat.py @@ -1,84 +1,128 @@ # coding: utf-8 +from __future__ import unicode_literals -import json import re from .common import InfoExtractor - +from ..compat import compat_str from ..utils import ( - compat_urllib_parse, + ExtractorError, unified_strdate, + HEADRequest, + float_or_none, ) class WatIE(InfoExtractor): - _VALID_URL=r'http://www.wat.tv/.*-(?P.*?)_.*?.html' + _VALID_URL = r'(?:wat:|https?://(?:www\.)?wat\.tv/video/.*-)(?P[0-9a-z]+)' IE_NAME = 'wat.tv' - _TEST = { - u'url': u'http://www.wat.tv/video/world-war-philadelphia-vost-6bv55_2fjr7_.html', - u'file': u'10631273.mp4', - u'md5': u'0a4fe7870f31eaeabb5e25fd8da8414a', - u'info_dict': { - u'title': u'World War Z - Philadelphia VOST', - u'description': u'La menace est partout. Que se passe-t-il à Philadelphia ?\r\nWORLD WAR Z, avec Brad Pitt, au cinéma le 3 juillet.\r\nhttp://www.worldwarz.fr', - } - } - - def download_video_info(self, real_id): + _TESTS = [ + { + 'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html', + 'md5': '83d882d9de5c9d97f0bb2c6273cde56a', + 'info_dict': { + 'id': '11713067', + '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', + 'ext': 'mp4', + 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)', + 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3', + 'upload_date': '20140816', + 'duration': 2910, + }, + 'skip': "Ce contenu n'est pas disponible pour l'instant.", + }, + ] + + def _real_extract(self, url): + video_id = self._match_id(url) + video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36)) + # 'contentv4' is used in the website, but it also returns the related # videos, we don't need them - info = self._download_webpage('http://www.wat.tv/interface/contentv3/' + real_id, real_id, 'Downloading video info') - info = json.loads(info) - return info['media'] + video_info = self._download_json( + 'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media'] + error_desc = video_info.get('error_desc') + if error_desc: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True) - 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') - webpage = self._download_webpage(url, short_id) - real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id') - - video_info = self.download_video_info(real_id) chapters = video_info['chapters'] first_chapter = chapters[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']) + def video_id_for_chapter(chapter): + return chapter['tc_start'].split('-')[0] + if video_id_for_chapter(first_chapter) != video_id: + self.to_screen('Multipart video detected') + entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters] + return self.playlist_result(entries, video_id, video_info['title']) # Otherwise we can continue and extract just one part, we have to use - # the short id for getting the video url - player_data = compat_urllib_parse.urlencode({'shortVideoId': short_id, - 'html5': '1'}) - player_info = self._download_webpage('http://www.wat.tv/player?' + player_data, - real_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, real_id, - u'Downloading player webpage') + # the video id for getting the video url - video_url = self._search_regex(r'urlhtml5 : "(.*?)"', player_webpage, - 'video url') - info = {'id': real_id, - 'url': video_url, - 'ext': 'mp4', - 'title': first_chapter['title'], - 'thumbnail': first_chapter['preview'], - 'description': first_chapter['description'], - 'view_count': video_info['views'], - } - if 'date_diffusion' in first_chapter: - info['upload_date'] = unified_strdate(first_chapter['date_diffusion']) + date_diffusion = first_chapter.get('date_diffusion') + upload_date = unified_strdate(date_diffusion) if date_diffusion else None - return info + def extract_url(path_template, url_type): + req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id) + head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type) + red_url = head.geturl() + if req_url == red_url: + raise ExtractorError( + '%s said: Sorry, this video is not available from your country.' % self.IE_NAME, + expected=True) + return red_url + + m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8') + http_url = extract_url('android5/%s.mp4', 'http') + + formats = [] + m3u8_formats = self._extract_m3u8_formats( + m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls') + formats.extend(m3u8_formats) + formats.extend(self._extract_f4m_formats( + m3u8_url.replace('ios.', 'web.').replace('.m3u8', '.f4m'), + video_id, f4m_id='hds', fatal=False)) + for m3u8_format in m3u8_formats: + mobj = re.search( + r'audio.*?%3D(\d+)(?:-video.*?%3D(\d+))?', m3u8_format['url']) + if not mobj: + continue + abr, vbr = mobj.groups() + abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000) + m3u8_format.update({ + 'vbr': vbr, + 'abr': abr, + }) + if not vbr or not abr: + continue + f = m3u8_format.copy() + f.update({ + 'url': re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url), + 'format_id': f['format_id'].replace('hls', 'http'), + 'protocol': 'http', + }) + formats.append(f) + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': first_chapter['title'], + 'thumbnail': first_chapter['preview'], + 'description': first_chapter['description'], + 'view_count': video_info['views'], + 'upload_date': upload_date, + 'duration': video_info['files'][0]['duration'], + 'formats': formats, + }