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     ExtractorError,
  11 )
  12
  13
  14 class LifeNewsIE(InfoExtractor):
  15     IE_NAME = 'lifenews'
  16     IE_DESC = 'LIFE | NEWS'
  17     _VALID_URL = r'http://lifenews\.ru/(?:mobile/)?news/(?P<id>\d+)'
  18
  19     _TEST = {
  20         'url': 'http://lifenews.ru/news/126342',
  21         'md5': 'e1b50a5c5fb98a6a544250f2e0db570a',
  22         'info_dict': {
  23             'id': '126342',
  24             'ext': 'mp4',
  25             'title': 'МВД разыскивает мужчин, оставивших в IKEA сумку с автоматом',
  26             'description': 'Камеры наблюдения гипермаркета зафиксировали троих мужчин, спрятавших оружейный арсенал в камере хранения.',
  27             'thumbnail': 're:http://.*\.jpg',
  28             'upload_date': '20140130',
  29         }
  30     }
  31
  32     def _real_extract(self, url):
  33         mobj = re.match(self._VALID_URL, url)
  34         video_id = mobj.group('id')
  35
  36         webpage = self._download_webpage('http://lifenews.ru/news/%s' % video_id, video_id, 'Downloading page')
  37
  38         videos = re.findall(r'<video.*?poster="(?P<poster>[^"]+)".*?src="(?P<video>[^"]+)".*?></video>', webpage)
  39         if not videos:
  40             raise ExtractorError('No media links available for %s' % video_id)
  41
  42         title = self._og_search_title(webpage)
  43         TITLE_SUFFIX = ' - Первый по срочным новостям — LIFE | NEWS'
  44         if title.endswith(TITLE_SUFFIX):
  45             title = title[:-len(TITLE_SUFFIX)]
  46
  47         description = self._og_search_description(webpage)
  48
  49         view_count = self._html_search_regex(
  50             r'<div class=\'views\'>(\d+)</div>', webpage, 'view count', fatal=False)
  51         comment_count = self._html_search_regex(
  52             r'<div class=\'comments\'>\s*<span class=\'counter\'>(\d+)</span>', webpage, 'comment count', fatal=False)
  53
  54         upload_date = self._html_search_regex(
  55             r'<time datetime=\'([^\']+)\'>', webpage, 'upload date', fatal=False)
  56         if upload_date is not None:
  57             upload_date = unified_strdate(upload_date)
  58
  59         def make_entry(video_id, media, video_number=None):
  60             return {
  61                 'id': video_id,
  62                 'url': media[1],
  63                 'thumbnail': media[0],
  64                 'title': title if video_number is None else '%s-video%s' % (title, video_number),
  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             }
  70
  71         if len(videos) == 1:
  72             return make_entry(video_id, videos[0])
  73         else:
  74             return [make_entry(video_id, media, video_number + 1) for video_number, media in enumerate(videos)]