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 unified_strdate
   9
  10
  11 class WatIE(InfoExtractor):
  12     _VALID_URL = r'http://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html'
  13     IE_NAME = 'wat.tv'
  14     _TESTS = [
  15         {
  16             'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
  17             'md5': 'ce70e9223945ed26a8056d413ca55dc9',
  18             'info_dict': {
  19                 'id': '11713067',
  20                 'display_id': 'soupe-figues-l-orange-aux-epices',
  21                 'ext': 'mp4',
  22                 'title': 'Soupe de figues à l\'orange et aux épices',
  23                 'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
  24                 'upload_date': '20140819',
  25                 'duration': 120,
  26             },
  27         },
  28         {
  29             'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html',
  30             'md5': 'fbc84e4378165278e743956d9c1bf16b',
  31             'info_dict': {
  32                 'id': '11713075',
  33                 'display_id': 'gregory-lemarchal-voix-ange',
  34                 'ext': 'mp4',
  35                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
  36                 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3',
  37                 'upload_date': '20140816',
  38                 'duration': 2910,
  39             },
  40         },
  41     ]
  42
  43     def download_video_info(self, real_id):
  44         # 'contentv4' is used in the website, but it also returns the related
  45         # videos, we don't need them
  46         info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
  47         return info['media']
  48
  49     def _real_extract(self, url):
  50         def real_id_for_chapter(chapter):
  51             return chapter['tc_start'].split('-')[0]
  52         mobj = re.match(self._VALID_URL, url)
  53         short_id = mobj.group('short_id')
  54         display_id = mobj.group('display_id')
  55         webpage = self._download_webpage(url, display_id or short_id)
  56         real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
  57
  58         video_info = self.download_video_info(real_id)
  59
  60         geo_list = video_info.get('geoList')
  61         country = geo_list[0] if geo_list else ''
  62
  63         chapters = video_info['chapters']
  64         first_chapter = chapters[0]
  65         files = video_info['files']
  66         first_file = files[0]
  67
  68         if real_id_for_chapter(first_chapter) != real_id:
  69             self.to_screen('Multipart video detected')
  70             chapter_urls = []
  71             for chapter in chapters:
  72                 chapter_id = real_id_for_chapter(chapter)
  73                 # Yes, when we this chapter is processed by WatIE,
  74                 # it will download the info again
  75                 chapter_info = self.download_video_info(chapter_id)
  76                 chapter_urls.append(chapter_info['url'])
  77             entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
  78             return self.playlist_result(entries, real_id, video_info['title'])
  79
  80         upload_date = None
  81         if 'date_diffusion' in first_chapter:
  82             upload_date = unified_strdate(first_chapter['date_diffusion'])
  83         # Otherwise we can continue and extract just one part, we have to use
  84         # the short id for getting the video url
  85
  86         formats = [{
  87             'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
  88             'format_id': 'Mobile',
  89         }]
  90
  91         fmts = [('SD', 'web')]
  92         if first_file.get('hasHD'):
  93             fmts.append(('HD', 'webhd'))
  94
  95         def compute_token(param):
  96             timestamp = '%08x' % int(self._download_webpage(
  97                 'http://www.wat.tv/servertime', real_id,
  98                 'Downloading server time').split('|')[0])
  99             magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
 100             return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
 101
 102         for fmt in fmts:
 103             webid = '/%s/%s' % (fmt[1], real_id)
 104             video_url = self._download_webpage(
 105                 'http://www.wat.tv/get%s?token=%s&getURL=1&country=%s' % (webid, compute_token(webid), country),
 106                 real_id,
 107                 'Downloding %s video URL' % fmt[0],
 108                 'Failed to download %s video URL' % fmt[0],
 109                 False)
 110             if not video_url:
 111                 continue
 112             formats.append({
 113                 'url': video_url,
 114                 'ext': 'mp4',
 115                 'format_id': fmt[0],
 116             })
 117
 118         return {
 119             'id': real_id,
 120             'display_id': display_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': first_file['duration'],
 127             'formats': formats,
 128         }