youtube_dlc/extractor/wat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     unified_strdate,
   8     HEADRequest,
   9     int_or_none,
  10 )
  11
  12
  13 class WatIE(InfoExtractor):
  14     _VALID_URL = r'(?:wat:|https?://(?:www\.)?wat\.tv/video/.*-)(?P<id>[0-9a-z]+)'
  15     IE_NAME = 'wat.tv'
  16     _TESTS = [
  17         {
  18             'url': 'http://www.wat.tv/video/soupe-figues-l-orange-aux-epices-6z1uz_2hvf7_.html',
  19             'info_dict': {
  20                 'id': '11713067',
  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             'params': {
  28                 # m3u8 download
  29                 'skip_download': True,
  30             },
  31             'expected_warnings': ['HTTP Error 404'],
  32         },
  33         {
  34             'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html',
  35             'md5': 'b16574df2c3cd1a36ca0098f2a791925',
  36             'info_dict': {
  37                 'id': '11713075',
  38                 'ext': 'mp4',
  39                 'title': 'Grégory Lemarchal, une voix d\'ange depuis 10 ans (1/3)',
  40                 'upload_date': '20140816',
  41             },
  42             'expected_warnings': ["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_data = self._download_json(
  53             'http://www.wat.tv/interface/contentv4s/' + video_id, video_id)
  54         video_info = video_data['media']
  55
  56         error_desc = video_info.get('error_desc')
  57         if error_desc:
  58             self.report_warning(
  59                 '%s returned error: %s' % (self.IE_NAME, error_desc))
  60
  61         chapters = video_info['chapters']
  62         if chapters:
  63             first_chapter = chapters[0]
  64
  65             def video_id_for_chapter(chapter):
  66                 return chapter['tc_start'].split('-')[0]
  67
  68             if video_id_for_chapter(first_chapter) != video_id:
  69                 self.to_screen('Multipart video detected')
  70                 entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters]
  71                 return self.playlist_result(entries, video_id, video_info['title'])
  72             # Otherwise we can continue and extract just one part, we have to use
  73             # the video id for getting the video url
  74         else:
  75             first_chapter = video_info
  76
  77         title = first_chapter['title']
  78
  79         def extract_url(path_template, url_type):
  80             req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id)
  81             head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type, fatal=False)
  82             if head:
  83                 red_url = head.geturl()
  84                 if req_url != red_url:
  85                     return red_url
  86             return None
  87
  88         formats = []
  89         manifest_urls = self._download_json(
  90             'http://www.wat.tv/get/webhtml/' + video_id, video_id)
  91         m3u8_url = manifest_urls.get('hls')
  92         if m3u8_url:
  93             formats.extend(self._extract_m3u8_formats(
  94                 m3u8_url, video_id, 'mp4',
  95                 'm3u8_native', m3u8_id='hls', fatal=False))
  96         mpd_url = manifest_urls.get('mpd')
  97         if mpd_url:
  98             formats.extend(self._extract_mpd_formats(
  99                 mpd_url.replace('://das-q1.tf1.fr/', '://das-q1-ssl.tf1.fr/'),
 100                 video_id, mpd_id='dash', fatal=False))
 101         self._sort_formats(formats)
 102
 103         date_diffusion = first_chapter.get('date_diffusion') or video_data.get('configv4', {}).get('estatS4')
 104         upload_date = unified_strdate(date_diffusion) if date_diffusion else None
 105         duration = None
 106         files = video_info['files']
 107         if files:
 108             duration = int_or_none(files[0].get('duration'))
 109
 110         return {
 111             'id': video_id,
 112             'title': title,
 113             'thumbnail': first_chapter.get('preview'),
 114             'description': first_chapter.get('description'),
 115             'view_count': int_or_none(video_info.get('views')),
 116             'upload_date': upload_date,
 117             'duration': duration,
 118             'formats': formats,
 119         }