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 import hashlib
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     unified_strdate,
  11 )
  12
  13
  14 class WatIE(InfoExtractor):
  15     _VALID_URL = r'http://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html'
  16     IE_NAME = 'wat.tv'
  17     _TESTS = [
  18         {
  19             'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
  20             'md5': 'ce70e9223945ed26a8056d413ca55dc9',
  21             'info_dict': {
  22                 'id': '11713067',
  23                 'display_id': 'soupe-figues-l-orange-aux-epices',
  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         },
  31         {
  32             'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html',
  33             'md5': 'fbc84e4378165278e743956d9c1bf16b',
  34             'info_dict': {
  35                 'id': '11713075',
  36                 'display_id': 'gregory-lemarchal-voix-ange',
  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 download_video_info(self, real_id):
  48         # 'contentv4' is used in the website, but it also returns the related
  49         # videos, we don't need them
  50         info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
  51         return info['media']
  52
  53     def _real_extract(self, url):
  54         def real_id_for_chapter(chapter):
  55             return chapter['tc_start'].split('-')[0]
  56         mobj = re.match(self._VALID_URL, url)
  57         short_id = mobj.group('short_id')
  58         display_id = mobj.group('display_id')
  59         webpage = self._download_webpage(url, display_id or short_id)
  60         real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
  61
  62         video_info = self.download_video_info(real_id)
  63
  64         error_desc = video_info.get('error_desc')
  65         if error_desc:
  66             raise ExtractorError(
  67                 '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True)
  68
  69         geo_list = video_info.get('geoList')
  70         country = geo_list[0] if geo_list else ''
  71
  72         chapters = video_info['chapters']
  73         first_chapter = chapters[0]
  74         files = video_info['files']
  75         first_file = files[0]
  76
  77         if real_id_for_chapter(first_chapter) != real_id:
  78             self.to_screen('Multipart video detected')
  79             chapter_urls = []
  80             for chapter in chapters:
  81                 chapter_id = real_id_for_chapter(chapter)
  82                 # Yes, when we this chapter is processed by WatIE,
  83                 # it will download the info again
  84                 chapter_info = self.download_video_info(chapter_id)
  85                 chapter_urls.append(chapter_info['url'])
  86             entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
  87             return self.playlist_result(entries, real_id, video_info['title'])
  88
  89         upload_date = None
  90         if 'date_diffusion' in first_chapter:
  91             upload_date = unified_strdate(first_chapter['date_diffusion'])
  92         # Otherwise we can continue and extract just one part, we have to use
  93         # the short id for getting the video url
  94
  95         formats = [{
  96             'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
  97             'format_id': 'Mobile',
  98         }]
  99
 100         fmts = [('SD', 'web')]
 101         if first_file.get('hasHD'):
 102             fmts.append(('HD', 'webhd'))
 103
 104         def compute_token(param):
 105             timestamp = '%08x' % int(self._download_webpage(
 106                 'http://www.wat.tv/servertime', real_id,
 107                 'Downloading server time').split('|')[0])
 108             magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
 109             return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
 110
 111         for fmt in fmts:
 112             webid = '/%s/%s' % (fmt[1], real_id)
 113             video_url = self._download_webpage(
 114                 'http://www.wat.tv/get%s?token=%s&getURL=1&country=%s' % (webid, compute_token(webid), country),
 115                 real_id,
 116                 'Downloding %s video URL' % fmt[0],
 117                 'Failed to download %s video URL' % fmt[0],
 118                 False)
 119             if not video_url:
 120                 continue
 121             formats.append({
 122                 'url': video_url,
 123                 'ext': 'mp4',
 124                 'format_id': fmt[0],
 125             })
 126
 127         return {
 128             'id': real_id,
 129             'display_id': display_id,
 130             'title': first_chapter['title'],
 131             'thumbnail': first_chapter['preview'],
 132             'description': first_chapter['description'],
 133             'view_count': video_info['views'],
 134             'upload_date': upload_date,
 135             'duration': first_file['duration'],
 136             'formats': formats,
 137         }