Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/wat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     ExtractorError,
  10     unified_strdate,
  11     HEADRequest,
  12     int_or_none,
  13 )
  14
  15
  16 class WatIE(InfoExtractor):
  17     _VALID_URL = r'(?:wat:|https?://(?:www\.)?wat\.tv/video/.*-)(?P<id>[0-9a-z]+)'
  18     IE_NAME = 'wat.tv'
  19     _TESTS = [
  20         {
  21             'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
  22             'md5': '83d882d9de5c9d97f0bb2c6273cde56a',
  23             'info_dict': {
  24                 'id': '11713067',
  25                 'ext': 'mp4',
  26                 'title': 'Soupe de figues à l\'orange et aux épices',
  27                 'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
  28                 'upload_date': '20140819',
  29                 'duration': 120,
  30             },
  31         },
  32         {
  33             'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html',
  34             'md5': '34bdfa5ca9fd3c7eb88601b635b0424c',
  35             'info_dict': {
  36                 'id': '11713075',
  37                 'ext': 'mp4',
  38                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
  39                 'upload_date': '20140816',
  40             },
  41             'expected_warnings': ["Ce contenu n'est pas disponible pour l'instant."],
  42         },
  43     ]
  44
  45     _FORMATS = (
  46         (200, 416, 234),
  47         (400, 480, 270),
  48         (600, 640, 360),
  49         (1200, 640, 360),
  50         (1800, 960, 540),
  51         (2500, 1280, 720),
  52     )
  53
  54     def _real_extract(self, url):
  55         video_id = self._match_id(url)
  56         video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36))
  57
  58         # 'contentv4' is used in the website, but it also returns the related
  59         # videos, we don't need them
  60         video_data = self._download_json(
  61             'http://www.wat.tv/interface/contentv4s/' + video_id, video_id)
  62         video_info = video_data['media']
  63
  64         error_desc = video_info.get('error_desc')
  65         if error_desc:
  66             self.report_warning(
  67                 '%s returned error: %s' % (self.IE_NAME, error_desc))
  68
  69         chapters = video_info['chapters']
  70         if chapters:
  71             first_chapter = chapters[0]
  72
  73             def video_id_for_chapter(chapter):
  74                 return chapter['tc_start'].split('-')[0]
  75
  76             if video_id_for_chapter(first_chapter) != video_id:
  77                 self.to_screen('Multipart video detected')
  78                 entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters]
  79                 return self.playlist_result(entries, video_id, video_info['title'])
  80             # Otherwise we can continue and extract just one part, we have to use
  81             # the video id for getting the video url
  82         else:
  83             first_chapter = video_info
  84
  85         title = first_chapter['title']
  86
  87         def extract_url(path_template, url_type):
  88             req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
  89             head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type, fatal=False)
  90             if head:
  91                 red_url = head.geturl()
  92                 if req_url != red_url:
  93                     return red_url
  94             return None
  95
  96         def remove_bitrate_limit(manifest_url):
  97             return re.sub(r'(?:max|min)_bitrate=\d+&?', '', manifest_url)
  98
  99         formats = []
 100         try:
 101             manifest_urls = self._download_json(
 102                 'http://www.wat.tv/get/webhtml/' + video_id, video_id)
 103             m3u8_url = manifest_urls.get('hls')
 104             if m3u8_url:
 105                 m3u8_url = remove_bitrate_limit(m3u8_url)
 106                 m3u8_formats = self._extract_m3u8_formats(
 107                     m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
 108                 if m3u8_formats:
 109                     formats.extend(m3u8_formats)
 110                     formats.extend(self._extract_f4m_formats(
 111                         m3u8_url.replace('ios', 'web').replace('.m3u8', '.f4m'),
 112                         video_id, f4m_id='hds', fatal=False))
 113                     http_url = extract_url('android5/%s.mp4', 'http')
 114                     if http_url:
 115                         for m3u8_format in m3u8_formats:
 116                             vbr, abr = m3u8_format.get('vbr'), m3u8_format.get('abr')
 117                             if not vbr or not abr:
 118                                 continue
 119                             format_id = m3u8_format['format_id'].replace('hls', 'http')
 120                             fmt_url = re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url)
 121                             if self._is_valid_url(fmt_url, video_id, format_id):
 122                                 f = m3u8_format.copy()
 123                                 f.update({
 124                                     'url': fmt_url,
 125                                     'format_id': format_id,
 126                                     'protocol': 'http',
 127                                 })
 128                                 formats.append(f)
 129             mpd_url = manifest_urls.get('mpd')
 130             if mpd_url:
 131                 formats.extend(self._extract_mpd_formats(remove_bitrate_limit(
 132                     mpd_url), video_id, mpd_id='dash', fatal=False))
 133             self._sort_formats(formats)
 134         except ExtractorError:
 135             abr = 64
 136             for vbr, width, height in self._FORMATS:
 137                 tbr = vbr + abr
 138                 format_id = 'http-%s' % tbr
 139                 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)
 140                 if self._is_valid_url(fmt_url, video_id, format_id):
 141                     formats.append({
 142                         'format_id': format_id,
 143                         'url': fmt_url,
 144                         'vbr': vbr,
 145                         'abr': abr,
 146                         'width': width,
 147                         'height': height,
 148                     })
 149
 150         date_diffusion = first_chapter.get('date_diffusion') or video_data.get('configv4', {}).get('estatS4')
 151         upload_date = unified_strdate(date_diffusion) if date_diffusion else None
 152         duration = None
 153         files = video_info['files']
 154         if files:
 155             duration = int_or_none(files[0].get('duration'))
 156
 157         return {
 158             'id': video_id,
 159             'title': title,
 160             'thumbnail': first_chapter.get('preview'),
 161             'description': first_chapter.get('description'),
 162             'view_count': int_or_none(video_info.get('views')),
 163             'upload_date': upload_date,
 164             'duration': duration,
 165             'formats': formats,
 166         }