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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     parse_duration,
   8     remove_end,
   9 )
  10
  11
  12 class LRTIE(InfoExtractor):
  13     IE_NAME = 'lrt.lt'
  14     _VALID_URL = r'https?://(?:www\.)?lrt\.lt/mediateka/irasas/(?P<id>[0-9]+)'
  15     _TEST = {
  16         'url': 'http://www.lrt.lt/mediateka/irasas/54391/',
  17         'info_dict': {
  18             'id': '54391',
  19             'ext': 'mp4',
  20             'title': 'Septynios Kauno dienos',
  21             'description': 'md5:24d84534c7dc76581e59f5689462411a',
  22             'duration': 1783,
  23             'view_count': int,
  24             'like_count': int,
  25         },
  26         'params': {
  27             'skip_download': True,  # m3u8 download
  28         },
  29     }
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33         webpage = self._download_webpage(url, video_id)
  34
  35         title = remove_end(self._og_search_title(webpage), ' - LRT')
  36         m3u8_url = self._search_regex(
  37             r'file\s*:\s*(["\'])(?P<url>.+?)\1\s*\+\s*location\.hash\.substring\(1\)',
  38             webpage, 'm3u8 url', group='url')
  39         formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
  40         self._sort_formats(formats)
  41
  42         thumbnail = self._og_search_thumbnail(webpage)
  43         description = self._og_search_description(webpage)
  44         duration = parse_duration(self._search_regex(
  45             r'var\s+record_len\s*=\s*(["\'])(?P<duration>[0-9]+:[0-9]+:[0-9]+)\1',
  46             webpage, 'duration', default=None, group='duration'))
  47
  48         view_count = int_or_none(self._html_search_regex(
  49             r'<div[^>]+class=(["\']).*?record-desc-seen.*?\1[^>]*>(?P<count>.+?)</div>',
  50             webpage, 'view count', fatal=False, group='count'))
  51         like_count = int_or_none(self._search_regex(
  52             r'<span[^>]+id=(["\'])flikesCount.*?\1>(?P<count>\d+)<',
  53             webpage, 'like count', fatal=False, group='count'))
  54
  55         return {
  56             'id': video_id,
  57             'title': title,
  58             'formats': formats,
  59             'thumbnail': thumbnail,
  60             'description': description,
  61             'duration': duration,
  62             'view_count': view_count,
  63             'like_count': like_count,
  64         }