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 )
  13
  14
  15 class WatIE(InfoExtractor):
  16     _VALID_URL = r'(?:wat:|https?://(?:www\.)?wat\.tv/video/.*-)(?P<id>[0-9a-z]+)'
  17     IE_NAME = 'wat.tv'
  18     _TESTS = [
  19         {
  20             'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
  21             'md5': '83d882d9de5c9d97f0bb2c6273cde56a',
  22             'info_dict': {
  23                 'id': '11713067',
  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                 'ext': 'mp4',
  37                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
  38                 'description': 'md5:b7a849cf16a2b733d9cd10c52906dee3',
  39                 'upload_date': '20140816',
  40                 'duration': 2910,
  41             },
  42             'skip': "Ce contenu n'est pas disponible pour l'instant.",
  43         },
  44     ]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48         video_id = video_id if video_id.isdigit() and len(video_id) > 6 else compat_str(int(video_id, 36))
  49
  50         # 'contentv4' is used in the website, but it also returns the related
  51         # videos, we don't need them
  52         video_info = self._download_json(
  53             'http://www.wat.tv/interface/contentv3/' + video_id, video_id)['media']
  54
  55         error_desc = video_info.get('error_desc')
  56         if error_desc:
  57             raise ExtractorError(
  58                 '%s returned error: %s' % (self.IE_NAME, error_desc), expected=True)
  59
  60         chapters = video_info['chapters']
  61         first_chapter = chapters[0]
  62
  63         def video_id_for_chapter(chapter):
  64             return chapter['tc_start'].split('-')[0]
  65
  66         if video_id_for_chapter(first_chapter) != video_id:
  67             self.to_screen('Multipart video detected')
  68             entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters]
  69             return self.playlist_result(entries, video_id, video_info['title'])
  70         # Otherwise we can continue and extract just one part, we have to use
  71         # the video id for getting the video url
  72
  73         date_diffusion = first_chapter.get('date_diffusion')
  74         upload_date = unified_strdate(date_diffusion) if date_diffusion else None
  75
  76         def extract_url(path_template, url_type):
  77             req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
  78             head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type)
  79             red_url = head.geturl()
  80             if req_url == red_url:
  81                 raise ExtractorError(
  82                     '%s said: Sorry, this video is not available from your country.' % self.IE_NAME,
  83                     expected=True)
  84             return red_url
  85
  86         m3u8_url = extract_url('ipad/%s.m3u8', 'm3u8')
  87         http_url = extract_url('android5/%s.mp4', 'http')
  88
  89         formats = []
  90         m3u8_formats = self._extract_m3u8_formats(
  91             m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
  92         formats.extend(m3u8_formats)
  93         formats.extend(self._extract_f4m_formats(
  94             m3u8_url.replace('ios.', 'web.').replace('.m3u8', '.f4m'),
  95             video_id, f4m_id='hds', fatal=False))
  96         for m3u8_format in m3u8_formats:
  97             vbr, abr = m3u8_format.get('vbr'), m3u8_format.get('abr')
  98             if not vbr or not abr:
  99                 continue
 100             f = m3u8_format.copy()
 101             f.update({
 102                 'url': re.sub(r'%s-\d+00-\d+' % video_id, '%s-%d00-%d' % (video_id, round(vbr / 100), round(abr)), http_url),
 103                 'format_id': f['format_id'].replace('hls', 'http'),
 104                 'protocol': 'http',
 105             })
 106             formats.append(f)
 107         self._sort_formats(formats)
 108
 109         return {
 110             'id': video_id,
 111             'title': first_chapter['title'],
 112             'thumbnail': first_chapter['preview'],
 113             'description': first_chapter['description'],
 114             'view_count': video_info['views'],
 115             'upload_date': upload_date,
 116             'duration': video_info['files'][0]['duration'],
 117             'formats': formats,
 118         }