youtube_dl/extractor/pornovoisines.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import random
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     int_or_none,
  10     float_or_none,
  11     unified_strdate,
  12 )
  13
  14
  15 class PornoVoisinesIE(InfoExtractor):
  16     _VALID_URL = r'http://(?:www\.)?pornovoisines\.com/showvideo/(?P<id>\d+)/(?P<display_id>[^/]+)'
  17
  18     _VIDEO_URL_TEMPLATE = 'http://stream%d.pornovoisines.com' \
  19         '/static/media/video/transcoded/%s-640x360-1000-trscded.mp4'
  20
  21     _SERVER_NUMBERS = (1, 2)
  22
  23     _TEST = {
  24         'url': 'http://www.pornovoisines.com/showvideo/1285/recherche-appartement/',
  25         'md5': '5ac670803bc12e9e7f9f662ce64cf1d1',
  26         'info_dict': {
  27             'id': '1285',
  28             'display_id': 'recherche-appartement',
  29             'ext': 'mp4',
  30             'title': 'Recherche appartement',
  31             'description': 'md5:819ea0b785e2a04667a1a01cdc89594e',
  32             'thumbnail': 're:^https?://.*\.jpg$',
  33             'upload_date': '20140925',
  34             'duration': 120,
  35             'view_count': int,
  36             'average_rating': float,
  37             'categories': ['Débutantes', 'Scénario', 'Sodomie'],
  38             'age_limit': 18,
  39         }
  40     }
  41
  42     @classmethod
  43     def build_video_url(cls, num):
  44         return cls._VIDEO_URL_TEMPLATE % (random.choice(cls._SERVER_NUMBERS), num)
  45
  46     def _real_extract(self, url):
  47         mobj = re.match(self._VALID_URL, url)
  48         video_id = mobj.group('id')
  49         display_id = mobj.group('display_id')
  50
  51         webpage = self._download_webpage(url, video_id)
  52
  53         video_url = self.build_video_url(video_id)
  54
  55         title = self._html_search_regex(
  56             r'<h1>(.+?)</h1>', webpage, 'title', flags=re.DOTALL)
  57         description = self._html_search_regex(
  58             r'<article id="descriptif">(.+?)</article>',
  59             webpage, "description", fatal=False, flags=re.DOTALL)
  60
  61         thumbnail = self._search_regex(
  62             r'<div id="mediaspace%s">\s*<img src="/?([^"]+)"' % video_id,
  63             webpage, 'thumbnail', fatal=False)
  64         if thumbnail:
  65             thumbnail = 'http://www.pornovoisines.com/%s' % thumbnail
  66
  67         upload_date = unified_strdate(self._search_regex(
  68             r'Publié le ([\d-]+)', webpage, 'upload date', fatal=False))
  69         duration = int_or_none(self._search_regex(
  70             'Durée (\d+)', webpage, 'duration', fatal=False))
  71         view_count = int_or_none(self._search_regex(
  72             r'(\d+) vues', webpage, 'view count', fatal=False))
  73         average_rating = self._search_regex(
  74             r'Note\s*:\s*(\d+(?:,\d+)?)', webpage, 'average rating', fatal=False)
  75         if average_rating:
  76             average_rating = float_or_none(average_rating.replace(',', '.'))
  77
  78         categories = self._html_search_meta(
  79             'keywords', webpage, 'categories', fatal=False)
  80         if categories:
  81             categories = [category.strip() for category in categories.split(',')]
  82
  83         return {
  84             'id': video_id,
  85             'display_id': display_id,
  86             'url': video_url,
  87             'title': title,
  88             'description': description,
  89             'thumbnail': thumbnail,
  90             'upload_date': upload_date,
  91             'duration': duration,
  92             'view_count': view_count,
  93             'average_rating': average_rating,
  94             'categories': categories,
  95             'age_limit': 18,
  96         }