youtube_dl/extractor/lrt.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     determine_ext,
   9     int_or_none,
  10     parse_duration,
  11     remove_end,
  12 )
  13
  14
  15 class LRTIE(InfoExtractor):
  16     IE_NAME = 'lrt.lt'
  17     _VALID_URL = r'https?://(?:www\.)?lrt\.lt/mediateka/irasas/(?P<id>[0-9]+)'
  18     _TESTS = [{
  19         # m3u8 download
  20         'url': 'http://www.lrt.lt/mediateka/irasas/54391/',
  21         'md5': 'fe44cf7e4ab3198055f2c598fc175cb0',
  22         'info_dict': {
  23             'id': '54391',
  24             'ext': 'mp4',
  25             'title': 'Septynios Kauno dienos',
  26             'description': 'md5:24d84534c7dc76581e59f5689462411a',
  27             'duration': 1783,
  28             'view_count': int,
  29             'like_count': int,
  30         },
  31     }, {
  32         # direct mp3 download
  33         'url': 'http://www.lrt.lt/mediateka/irasas/1013074524/',
  34         'md5': '389da8ca3cad0f51d12bed0c844f6a0a',
  35         'info_dict': {
  36             'id': '1013074524',
  37             'ext': 'mp3',
  38             'title': 'Kita tema 2016-09-05 15:05',
  39             'description': 'md5:1b295a8fc7219ed0d543fc228c931fb5',
  40             'duration': 3008,
  41             'view_count': int,
  42             'like_count': int,
  43         },
  44     }]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48         webpage = self._download_webpage(url, video_id)
  49
  50         title = remove_end(self._og_search_title(webpage), ' - LRT')
  51
  52         formats = []
  53         for _, file_url in re.findall(
  54                 r'file\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage):
  55             ext = determine_ext(file_url)
  56             if ext not in ('m3u8', 'mp3'):
  57                 continue
  58             # mp3 served as m3u8 produces stuttered media file
  59             if ext == 'm3u8' and '.mp3' in file_url:
  60                 continue
  61             if ext == 'm3u8':
  62                 formats.extend(self._extract_m3u8_formats(
  63                     file_url, video_id, 'mp4', entry_protocol='m3u8_native',
  64                     fatal=False))
  65             elif ext == 'mp3':
  66                 formats.append({
  67                     'url': file_url,
  68                     'vcodec': 'none',
  69                 })
  70         self._sort_formats(formats)
  71
  72         thumbnail = self._og_search_thumbnail(webpage)
  73         description = self._og_search_description(webpage)
  74         duration = parse_duration(self._search_regex(
  75             r'var\s+record_len\s*=\s*(["\'])(?P<duration>[0-9]+:[0-9]+:[0-9]+)\1',
  76             webpage, 'duration', default=None, group='duration'))
  77
  78         view_count = int_or_none(self._html_search_regex(
  79             r'<div[^>]+class=(["\']).*?record-desc-seen.*?\1[^>]*>(?P<count>.+?)</div>',
  80             webpage, 'view count', fatal=False, group='count'))
  81         like_count = int_or_none(self._search_regex(
  82             r'<span[^>]+id=(["\'])flikesCount.*?\1>(?P<count>\d+)<',
  83             webpage, 'like count', fatal=False, group='count'))
  84
  85         return {
  86             'id': video_id,
  87             'title': title,
  88             'formats': formats,
  89             'thumbnail': thumbnail,
  90             'description': description,
  91             'duration': duration,
  92             'view_count': view_count,
  93             'like_count': like_count,
  94         }