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             'info_dict': {
  23                 'id': '11713067',
  24                 'ext': 'mp4',
  25                 'title': 'Soupe de figues à l\'orange et aux épices',
  26                 'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
  27                 'upload_date': '20140819',
  28                 'duration': 120,
  29             },
  30             'params': {
  31                 # m3u8 download
  32                 'skip_download': True,
  33             },
  34             'expected_warnings': ['HTTP Error 404'],
  35         },
  36         {
  37             'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html',
  38             'md5': 'b16574df2c3cd1a36ca0098f2a791925',
  39             'info_dict': {
  40                 'id': '11713075',
  41                 'ext': 'mp4',
  42                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
  43                 'upload_date': '20140816',
  44             },
  45             'expected_warnings': ["Ce contenu n'est pas disponible pour l'instant."],
  46         },
  47     ]
  48
  49     _FORMATS = (
  50         (200, 416, 234),
  51         (400, 480, 270),
  52         (600, 640, 360),
  53         (1200, 640, 360),
  54         (1800, 960, 540),
  55         (2500, 1280, 720),
  56     )
  57
  58     def _real_extract(self, url):
  59         video_id = self._match_id(url)
  60         video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36))
  61
  62         # 'contentv4' is used in the website, but it also returns the related
  63         # videos, we don't need them
  64         video_data = self._download_json(
  65             'http://www.wat.tv/interface/contentv4s/' + video_id, video_id)
  66         video_info = video_data['media']
  67
  68         error_desc = video_info.get('error_desc')
  69         if error_desc:
  70             self.report_warning(
  71                 '%s returned error: %s' % (self.IE_NAME, error_desc))
  72
  73         chapters = video_info['chapters']
  74         if chapters:
  75             first_chapter = chapters[0]
  76
  77             def video_id_for_chapter(chapter):
  78                 return chapter['tc_start'].split('-')[0]
  79
  80             if video_id_for_chapter(first_chapter) != video_id:
  81                 self.to_screen('Multipart video detected')
  82                 entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters]
  83                 return self.playlist_result(entries, video_id, video_info['title'])
  84             # Otherwise we can continue and extract just one part, we have to use
  85             # the video id for getting the video url
  86         else:
  87             first_chapter = video_info
  88
  89         title = first_chapter['title']
  90
  91         def extract_url(path_template, url_type):
  92             req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
  93             head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type, fatal=False)
  94             if head:
  95                 red_url = head.geturl()
  96                 if req_url != red_url:
  97                     return red_url
  98             return None
  99
 100         def remove_bitrate_limit(manifest_url):
 101             return re.sub(r'(?:max|min)_bitrate=\d+&?', '', manifest_url)
 102
 103         formats = []
 104         try:
 105             alt_urls = lambda manifest_url: [re.sub(r'(?:wdv|ssm)?\.ism/', repl + '.ism/', manifest_url) for repl in ('', 'ssm')]
 106             manifest_urls = self._download_json(
 107                 'http://www.wat.tv/get/webhtml/' + video_id, video_id)
 108             m3u8_url = manifest_urls.get('hls')
 109             if m3u8_url:
 110                 m3u8_url = remove_bitrate_limit(m3u8_url)
 111                 for m3u8_alt_url in alt_urls(m3u8_url):
 112                     formats.extend(self._extract_m3u8_formats(
 113                         m3u8_alt_url, video_id, 'mp4',
 114                         'm3u8_native', m3u8_id='hls', fatal=False))
 115                     formats.extend(self._extract_f4m_formats(
 116                         m3u8_alt_url.replace('ios', 'web').replace('.m3u8', '.f4m'),
 117                         video_id, f4m_id='hds', fatal=False))
 118             mpd_url = manifest_urls.get('mpd')
 119             if mpd_url:
 120                 mpd_url = remove_bitrate_limit(mpd_url)
 121                 for mpd_alt_url in alt_urls(mpd_url):
 122                     formats.extend(self._extract_mpd_formats(
 123                         mpd_alt_url, video_id, mpd_id='dash', fatal=False))
 124             self._sort_formats(formats)
 125         except ExtractorError:
 126             abr = 64
 127             for vbr, width, height in self._FORMATS:
 128                 tbr = vbr + abr
 129                 format_id = 'http-%s' % tbr
 130                 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)
 131                 if self._is_valid_url(fmt_url, video_id, format_id):
 132                     formats.append({
 133                         'format_id': format_id,
 134                         'url': fmt_url,
 135                         'vbr': vbr,
 136                         'abr': abr,
 137                         'width': width,
 138                         'height': height,
 139                     })
 140
 141         date_diffusion = first_chapter.get('date_diffusion') or video_data.get('configv4', {}).get('estatS4')
 142         upload_date = unified_strdate(date_diffusion) if date_diffusion else None
 143         duration = None
 144         files = video_info['files']
 145         if files:
 146             duration = int_or_none(files[0].get('duration'))
 147
 148         return {
 149             'id': video_id,
 150             'title': title,
 151             'thumbnail': first_chapter.get('preview'),
 152             'description': first_chapter.get('description'),
 153             'view_count': int_or_none(video_info.get('views')),
 154             'upload_date': upload_date,
 155             'duration': duration,
 156             'formats': formats,
 157         }