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     float_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': 'fbc84e4378165278e743956d9c1bf16b',
  35             'info_dict': {
  36                 'id': '11713075',
  37                 'ext': 'mp4',
  38                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
  39                 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3',
  40                 'upload_date': '20140816',
  41                 'duration': 2910,
  42             },
  43             'skip': "Ce contenu n'est pas disponible pour l'instant.",
  44         },
  45     ]
  46
  47     def _real_extract(self, url):
  48         video_id = self._match_id(url)
  49         video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36))
  50
  51         # 'contentv4' is used in the website, but it also returns the related
  52         # videos, we don't need them
  53         video_info = self._download_json(
  54             'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media']
  55
  56         error_desc = video_info.get('error_desc')
  57         if error_desc:
  58             raise ExtractorError(
  59                 '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True)
  60
  61         chapters = video_info['chapters']
  62         first_chapter = chapters[0]
  63
  64         def video_id_for_chapter(chapter):
  65             return chapter['tc_start'].split('-')[0]
  66
  67         if video_id_for_chapter(first_chapter) != video_id:
  68             self.to_screen('Multipart video detected')
  69             entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters]
  70             return self.playlist_result(entries, video_id, video_info['title'])
  71         # Otherwise we can continue and extract just one part, we have to use
  72         # the video id for getting the video url
  73
  74         date_diffusion = first_chapter.get('date_diffusion')
  75         upload_date = unified_strdate(date_diffusion) if date_diffusion else None
  76
  77         def extract_url(path_template, url_type):
  78             req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
  79             head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type)
  80             red_url = head.geturl()
  81             if req_url == red_url:
  82                 raise ExtractorError(
  83                     '%s said: Sorry, this video is not available from your country.' % self.IE_NAME,
  84                     expected=True)
  85             return red_url
  86
  87         m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8')
  88         http_url = extract_url('android5/%s.mp4', 'http')
  89
  90         formats = []
  91         m3u8_formats = self._extract_m3u8_formats(
  92             m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
  93         formats.extend(m3u8_formats)
  94         formats.extend(self._extract_f4m_formats(
  95             m3u8_url.replace('ios.', 'web.').replace('.m3u8', '.f4m'),
  96             video_id, f4m_id='hds', fatal=False))
  97         for m3u8_format in m3u8_formats:
  98             mobj = re.search(
  99                 r'audio.*?%3D(\d+)(?:-video.*?%3D(\d+))?', m3u8_format['url'])
 100             if not mobj:
 101                 continue
 102             abr, vbr = mobj.groups()
 103             abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
 104             m3u8_format.update({
 105                 'vbr': vbr,
 106                 'abr': abr,
 107             })
 108             if not vbr or not abr:
 109                 continue
 110             f = m3u8_format.copy()
 111             f.update({
 112                 'url': re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url),
 113                 'format_id': f['format_id'].replace('hls', 'http'),
 114                 'protocol': 'http',
 115             })
 116             formats.append(f)
 117         self._sort_formats(formats)
 118
 119         return {
 120             'id': video_id,
 121             'title': first_chapter['title'],
 122             'thumbnail': first_chapter['preview'],
 123             'description': first_chapter['description'],
 124             'view_count': video_info['views'],
 125             'upload_date': upload_date,
 126             'duration': video_info['files'][0]['duration'],
 127             'formats': formats,
 128         }