youtube_dl/extractor/wat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6 import hashlib
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  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     _TEST = {
  18         'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
  19         'md5': 'ce70e9223945ed26a8056d413ca55dc9',
  20         'info_dict': {
  21             'id': '11713067',
  22             'display_id': 'soupe-figues-l-orange-aux-epices',
  23             'ext': 'mp4',
  24             'title': 'Soupe de figues à l\'orange et aux épices',
  25             'description': 'Retrouvez l\'émission "Petits plats en équilibre", diffusée le 18 août 2014.',
  26             'upload_date': '20140819',
  27             'duration': 120,
  28         },
  29     }
  30
  31     def download_video_info(self, real_id):
  32         # 'contentv4' is used in the website, but it also returns the related
  33         # videos, we don't need them
  34         info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
  35         return info['media']
  36
  37     def _real_extract(self, url):
  38         def real_id_for_chapter(chapter):
  39             return chapter['tc_start'].split('-')[0]
  40         mobj = re.match(self._VALID_URL, url)
  41         short_id = mobj.group('short_id')
  42         display_id = mobj.group('display_id')
  43         webpage = self._download_webpage(url, display_id or short_id)
  44         real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
  45
  46         video_info = self.download_video_info(real_id)
  47
  48         if video_info.get('geolock'):
  49             self.report_warning(
  50                 'This content is marked as not available in your area. Trying anyway ..')
  51
  52         chapters = video_info['chapters']
  53         first_chapter = chapters[0]
  54         files = video_info['files']
  55         first_file = files[0]
  56
  57         if real_id_for_chapter(first_chapter) != real_id:
  58             self.to_screen('Multipart video detected')
  59             chapter_urls = []
  60             for chapter in chapters:
  61                 chapter_id = real_id_for_chapter(chapter)
  62                 # Yes, when we this chapter is processed by WatIE,
  63                 # it will download the info again
  64                 chapter_info = self.download_video_info(chapter_id)
  65                 chapter_urls.append(chapter_info['url'])
  66             entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
  67             return self.playlist_result(entries, real_id, video_info['title'])
  68
  69         upload_date = None
  70         if 'date_diffusion' in first_chapter:
  71             upload_date = unified_strdate(first_chapter['date_diffusion'])
  72         # Otherwise we can continue and extract just one part, we have to use
  73         # the short id for getting the video url
  74
  75         formats = [{
  76             'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
  77             'format_id': 'Mobile',
  78         }]
  79
  80         fmts = [('SD', 'web')]
  81         if first_file.get('hasHD'):
  82             fmts.append(('HD', 'webhd'))
  83
  84         def compute_token(param):
  85             timestamp = '%08x' % int(time.time())
  86             magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
  87             return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
  88
  89         for fmt in fmts:
  90             webid = '/%s/%s' % (fmt[1], real_id)
  91             video_url = self._download_webpage(
  92                 'http://www.wat.tv/get%s?token=%s&getURL=1' % (webid, compute_token(webid)),
  93                 real_id,
  94                 'Downloding %s video URL' % fmt[0],
  95                 'Failed to download %s video URL' % fmt[0],
  96                 False)
  97             if not video_url:
  98                 continue
  99             formats.append({
 100                 'url': video_url,
 101                 'ext': 'mp4',
 102                 'format_id': fmt[0],
 103             })
 104
 105         return {
 106             'id': real_id,
 107             'display_id': display_id,
 108             'title': first_chapter['title'],
 109             'thumbnail': first_chapter['preview'],
 110             'description': first_chapter['description'],
 111             'view_count': video_info['views'],
 112             'upload_date': upload_date,
 113             'duration': first_file['duration'],
 114             'formats': formats,
 115         }