_ Git - youtube-dl/blob - youtube_dl/extractor/freesound.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     float_or_none,
   8     get_element_by_class,
   9     get_element_by_id,
  10     unified_strdate,
  11 )
  12
  13
  14 class FreesoundIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?freesound\.org/people/[^/]+/sounds/(?P<id>[^/]+)'
  16     _TEST = {
  17         'url': 'http://www.freesound.org/people/miklovan/sounds/194503/',
  18         'md5': '12280ceb42c81f19a515c745eae07650',
  19         'info_dict': {
  20             'id': '194503',
  21             'ext': 'mp3',
  22             'title': 'gulls in the city.wav',
  23             'description': 'the sounds of seagulls in the city',
  24             'duration': 130.233,
  25             'uploader': 'miklovan',
  26             'upload_date': '20130715',
  27             'tags': list,
  28         }
  29     }
  30
  31     def _real_extract(self, url):
  32         audio_id = self._match_id(url)
  33
  34         webpage = self._download_webpage(url, audio_id)
  35
  36         audio_url = self._og_search_property('audio', webpage, 'song url')
  37         title = self._og_search_property('audio:title', webpage, 'song title')
  38
  39         description = self._html_search_regex(
  40             r'(?s)id=["\']sound_description["\'][^>]*>(.+?)</div>',
  41             webpage, 'description', fatal=False)
  42
  43         duration = float_or_none(
  44             get_element_by_class('duration', webpage), scale=1000)
  45
  46         upload_date = unified_strdate(get_element_by_id('sound_date', webpage))
  47         uploader = self._og_search_property(
  48             'audio:artist', webpage, 'uploader', fatal=False)
  49
  50         channels = self._html_search_regex(
  51             r'Channels</dt><dd>(.+?)</dd>', webpage,
  52             'channels info', fatal=False)
  53
  54         tags_str = get_element_by_class('tags', webpage)
  55         tags = re.findall(r'<a[^>]+>([^<]+)', tags_str) if tags_str else None
  56
  57         audio_urls = [audio_url]
  58
  59         LQ_FORMAT = '-lq.mp3'
  60         if LQ_FORMAT in audio_url:
  61             audio_urls.append(audio_url.replace(LQ_FORMAT, '-hq.mp3'))
  62
  63         formats = [{
  64             'url': format_url,
  65             'format_note': channels,
  66             'quality': quality,
  67         } for quality, format_url in enumerate(audio_urls)]
  68         self._sort_formats(formats)
  69
  70         return {
  71             'id': audio_id,
  72             'title': title,
  73             'description': description,
  74             'duration': duration,
  75             'uploader': uploader,
  76             'upload_date': upload_date,
  77             'tags': tags,
  78             'formats': formats,
  79         }