_ Git - youtube-dl/blob - 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     ExtractorError,
  11     unified_strdate,
  12 )
  13
  14
  15 class WatIE(InfoExtractor):
  16     _VALID_URL = r'http://www\.wat\.tv/video/(?P<display_id>.*)-(?P<short_id>.*?)_.*?\.html'
  17     IE_NAME = 'wat.tv'
  18     _TEST = {
  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     def download_video_info(self, real_id):
  33         # 'contentv4' is used in the website, but it also returns the related
  34         # videos, we don't need them
  35         info = self._download_json('http://www.wat.tv/interface/contentv3/' + real_id, real_id)
  36         return info['media']
  37
  38     def _real_extract(self, url):
  39         def real_id_for_chapter(chapter):
  40             return chapter['tc_start'].split('-')[0]
  41         mobj = re.match(self._VALID_URL, url)
  42         short_id = mobj.group('short_id')
  43         display_id = mobj.group('display_id')
  44         webpage = self._download_webpage(url, display_id or short_id)
  45         real_id = self._search_regex(r'xtpage = ".*-(.*?)";', webpage, 'real id')
  46
  47         video_info = self.download_video_info(real_id)
  48
  49         if video_info.get('geolock'):
  50             self.report_warning(
  51                 'This content is marked as not available in your area. Trying anyway ..')
  52
  53         chapters = video_info['chapters']
  54         first_chapter = chapters[0]
  55         files = video_info['files']
  56         first_file = files[0]
  57
  58         if real_id_for_chapter(first_chapter) != real_id:
  59             self.to_screen('Multipart video detected')
  60             chapter_urls = []
  61             for chapter in chapters:
  62                 chapter_id = real_id_for_chapter(chapter)
  63                 # Yes, when we this chapter is processed by WatIE,
  64                 # it will download the info again
  65                 chapter_info = self.download_video_info(chapter_id)
  66                 chapter_urls.append(chapter_info['url'])
  67             entries = [self.url_result(chapter_url) for chapter_url in chapter_urls]
  68             return self.playlist_result(entries, real_id, video_info['title'])
  69
  70         upload_date = None
  71         if 'date_diffusion' in first_chapter:
  72             upload_date = unified_strdate(first_chapter['date_diffusion'])
  73         # Otherwise we can continue and extract just one part, we have to use
  74         # the short id for getting the video url
  75
  76         formats = [{
  77             'url': 'http://wat.tv/get/android5/%s.mp4' % real_id,
  78             'format_id': 'Mobile',
  79         }]
  80
  81         fmts = [('SD', 'web')]
  82         if first_file.get('hasHD'):
  83             fmts.append(('HD', 'webhd'))
  84
  85         def compute_token(param):
  86             timestamp = '%08x' % int(time.time())
  87             magic = '9b673b13fa4682ed14c3cfa5af5310274b514c4133e9b3a81e6e3aba009l2564'
  88             return '%s/%s' % (hashlib.md5((magic + param + timestamp).encode('ascii')).hexdigest(), timestamp)
  89
  90         for fmt in fmts:
  91             webid = '/%s/%s' % (fmt[1], real_id)
  92             video_url = self._download_webpage(
  93                 'http://www.wat.tv/get%s?token=%s&getURL=1' % (webid, compute_token(webid)),
  94                 real_id,
  95                 'Downloding %s video URL' % fmt[0],
  96                 'Failed to download %s video URL' % fmt[0],
  97                 False)
  98             if not video_url:
  99                 continue
 100             formats.append({
 101                 'url': video_url,
 102                 'ext': 'mp4',
 103                 'format_id': fmt[0],
 104             })
 105
 106         return {
 107             'id': real_id,
 108             'display_id': display_id,
 109             'title': first_chapter['title'],
 110             'thumbnail': first_chapter['preview'],
 111             'description': first_chapter['description'],
 112             'view_count': video_info['views'],
 113             'upload_date': upload_date,
 114             'duration': first_file['duration'],
 115             'formats': formats,
 116         }