- 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'])
-
- # 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')
-
- 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'])
-
- return info
+ if chapters:
+ first_chapter = chapters[0]
+
+ 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 video id for getting the video url
+ else:
+ first_chapter = video_info
+
+ title = first_chapter['title']
+
+ 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, fatal=False)
+ if head:
+ red_url = head.geturl()
+ if req_url != red_url:
+ return red_url
+ return None
+
+ def remove_bitrate_limit(manifest_url):
+ return re.sub(r'(?:max|min)_bitrate=\d+&?', '', manifest_url)
+
+ formats = []
+ try:
+ alt_urls = lambda manifest_url: [re.sub(r'(?:wdv|ssm)?\.ism/', repl + '.ism/', manifest_url) for repl in ('', 'ssm')]
+ manifest_urls = self._download_json(
+ 'http://www.wat.tv/get/webhtml/' + video_id, video_id)
+ m3u8_url = manifest_urls.get('hls')
+ if m3u8_url:
+ m3u8_url = remove_bitrate_limit(m3u8_url)
+ for m3u8_alt_url in alt_urls(m3u8_url):
+ formats.extend(self._extract_m3u8_formats(
+ m3u8_alt_url, video_id, 'mp4',
+ 'm3u8_native', m3u8_id='hls', fatal=False))
+ formats.extend(self._extract_f4m_formats(
+ m3u8_alt_url.replace('ios', 'web').replace('.m3u8', '.f4m'),
+ video_id, f4m_id='hds', fatal=False))
+ mpd_url = manifest_urls.get('mpd')
+ if mpd_url:
+ mpd_url = remove_bitrate_limit(mpd_url)
+ for mpd_alt_url in alt_urls(mpd_url):
+ formats.extend(self._extract_mpd_formats(
+ mpd_alt_url, video_id, mpd_id='dash', fatal=False))
+ self._sort_formats(formats)
+ except ExtractorError:
+ abr = 64
+ for vbr, width, height in self._FORMATS:
+ tbr = vbr + abr
+ format_id = 'http-%s' % tbr
+ fmt_url = 'http://dnl.adv.tf1.fr/2/USP-0x0/%s/%s/%s/ssm/%s-%s-64k.mp4' % (video_id[-4:-2], video_id[-2:], video_id, video_id, vbr)
+ if self._is_valid_url(fmt_url, video_id, format_id):
+ formats.append({
+ 'format_id': format_id,
+ 'url': fmt_url,
+ 'vbr': vbr,
+ 'abr': abr,
+ 'width': width,
+ 'height': height,
+ })
+
+ date_diffusion = first_chapter.get('date_diffusion') or video_data.get('configv4', {}).get('estatS4')
+ upload_date = unified_strdate(date_diffusion) if date_diffusion else None
+ duration = None
+ files = video_info['files']
+ if files:
+ duration = int_or_none(files[0].get('duration'))
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'thumbnail': first_chapter.get('preview'),
+ 'description': first_chapter.get('description'),
+ 'view_count': int_or_none(video_info.get('views')),
+ 'upload_date': upload_date,
+ 'duration': duration,
+ 'formats': formats,
+ }
Rapsys Git / youtubedl / blobdiff