diff options
Diffstat (limited to 'youtube_dl/extractor/wat.py')
| -rw-r--r-- | youtube_dl/extractor/wat.py | 120 |
1 files changed, 49 insertions, 71 deletions
diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py index de7d6b559..b15e03768 100644 --- a/youtube_dl/extractor/wat.py +++ b/youtube_dl/extractor/wat.py @@ -1,15 +1,13 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import compat_str from ..utils import ( ExtractorError, + int_or_none, + try_get, unified_strdate, - HEADRequest, - float_or_none, ) @@ -19,7 +17,6 @@ class WatIE(InfoExtractor): _TESTS = [ { 'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html', - 'md5': '83d882d9de5c9d97f0bb2c6273cde56a', 'info_dict': { 'id': '11713067', 'ext': 'mp4', @@ -28,21 +25,27 @@ class WatIE(InfoExtractor): 'upload_date': '20140819', 'duration': 120, }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + 'expected_warnings': ['HTTP Error 404'], + 'skip': 'This content is no longer available', }, { 'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html', - 'md5': 'fbc84e4378165278e743956d9c1bf16b', + 'md5': 'b16574df2c3cd1a36ca0098f2a791925', '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.", + 'expected_warnings': ["Ce contenu n'est pas disponible pour l'instant."], + 'skip': 'This content is no longer available', }, ] + _GEO_BYPASS = False def _real_extract(self, url): video_id = self._match_id(url) @@ -50,79 +53,54 @@ class WatIE(InfoExtractor): # 'contentv4' is used in the website, but it also returns the related # videos, we don't need them - video_info = self._download_json( - 'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media'] + # video_data = self._download_json( + # 'http://www.wat.tv/interface/contentv4s/' + video_id, video_id) + video_data = self._download_json( + 'https://mediainfo.tf1.fr/mediainfocombo/' + video_id, + video_id, query={'context': 'MYTF1', 'pver': '4001000'}) + video_info = video_data['media'] error_desc = video_info.get('error_desc') if error_desc: - raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True) - - chapters = video_info['chapters'] - first_chapter = chapters[0] + if video_info.get('error_code') == 'GEOBLOCKED': + self.raise_geo_restricted(error_desc, video_info.get('geoList')) + raise ExtractorError(error_desc, expected=True) - def video_id_for_chapter(chapter): - return chapter['tc_start'].split('-')[0] + title = video_info['title'] - 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 video id for getting the video url - - date_diffusion = first_chapter.get('date_diffusion') - upload_date = unified_strdate(date_diffusion) if date_diffusion else None + formats = [] - 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 + def extract_formats(manifest_urls): + for f, f_url in manifest_urls.items(): + if not f_url: + continue + if f in ('dash', 'mpd'): + formats.extend(self._extract_mpd_formats( + f_url.replace('://das-q1.tf1.fr/', '://das-q1-ssl.tf1.fr/'), + video_id, mpd_id='dash', fatal=False)) + elif f == 'hls': + formats.extend(self._extract_m3u8_formats( + f_url, video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False)) - m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8') - http_url = extract_url('android5/%s.mp4', 'http') + delivery = video_data.get('delivery') or {} + extract_formats({delivery.get('format'): delivery.get('url')}) + if not formats: + if delivery.get('drm'): + raise ExtractorError('This video is DRM protected.', expected=True) + manifest_urls = self._download_json( + 'http://www.wat.tv/get/webhtml/' + video_id, video_id, fatal=False) + if manifest_urls: + extract_formats(manifest_urls) - 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'], + 'title': title, + 'thumbnail': video_info.get('preview'), + 'upload_date': unified_strdate(try_get( + video_data, lambda x: x['mediametrie']['chapters'][0]['estatS4'])), + 'duration': int_or_none(video_info.get('duration')), 'formats': formats, } |