Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/lifenews.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     unified_strdate
  10 )
  11
  12
  13 class LifeNewsIE(InfoExtractor):
  14     IE_NAME = 'lifenews'
  15     IE_DESC = 'LIFE | NEWS'
  16     _VALID_URL = r'http://lifenews\.ru/(?:mobile/)?news/(?P<id>\d+)'
  17
  18     _TEST = {
  19         'url': 'http://lifenews.ru/news/126342',
  20         'md5': 'e1b50a5c5fb98a6a544250f2e0db570a',
  21         'info_dict': {
  22             'id': '126342',
  23             'ext': 'mp4',
  24             'title': 'МВД разыскивает мужчин, оставивших в IKEA сумку с автоматом',
  25             'description': 'Камеры наблюдения гипермаркета зафиксировали троих мужчин, спрятавших оружейный арсенал в камере хранения.',
  26             'thumbnail': 'http://lifenews.ru/static/posts/2014/1/126342/.video.jpg',
  27             'upload_date': '20140130',
  28         }
  29     }
  30
  31     def _real_extract(self, url):
  32         mobj = re.match(self._VALID_URL, url)
  33         video_id = mobj.group('id')
  34
  35         webpage = self._download_webpage('http://lifenews.ru/mobile/news/%s' % video_id, video_id, 'Downloading page')
  36
  37         video_url = self._html_search_regex(
  38             r'<video.*?src="([^"]+)".*?></video>', webpage, 'video URL')
  39
  40         thumbnail = self._html_search_regex(
  41             r'<video.*?poster="([^"]+)".*?"></video>', webpage, 'video thumbnail')
  42
  43         title = self._og_search_title(webpage)
  44         TITLE_SUFFIX = ' - Первый по срочным новостям — LIFE | NEWS'
  45         if title.endswith(TITLE_SUFFIX):
  46             title = title[:-len(TITLE_SUFFIX)]
  47
  48         description = self._og_search_description(webpage)
  49
  50         view_count = self._html_search_regex(
  51             r'<div class=\'views\'>(\d+)</div>', webpage, 'view count', fatal=False)
  52         comment_count = self._html_search_regex(
  53             r'<div class=\'comments\'>(\d+)</div>', webpage, 'comment count', fatal=False)
  54
  55         upload_date = self._html_search_regex(
  56             r'<time datetime=\'([^\']+)\'>', webpage, 'upload date',fatal=False)
  57         if upload_date is not None:
  58             upload_date = unified_strdate(upload_date)
  59
  60         return {
  61             'id': video_id,
  62             'url': video_url,
  63             'thumbnail': thumbnail,
  64             'title': title,
  65             'description': description,
  66             'view_count': int_or_none(view_count),
  67             'comment_count': int_or_none(comment_count),
  68             'upload_date': upload_date,
  69         }