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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urlparse
   8 from ..utils import (
   9     float_or_none,
  10     month_by_abbreviation,
  11 )
  12
  13
  14 class YamIE(InfoExtractor):
  15     _VALID_URL = r'http://mymedia.yam.com/m/(?P<id>\d+)'
  16
  17     _TESTS = [{
  18         # An audio hosted on Yam
  19         'url': 'http://mymedia.yam.com/m/2283921',
  20         'md5': 'c011b8e262a52d5473d9c2e3c9963b9c',
  21         'info_dict': {
  22             'id': '2283921',
  23             'ext': 'mp3',
  24             'title': '發現 - 趙薇 京華煙雲主題曲',
  25             'uploader_id': 'princekt',
  26             'upload_date': '20080807',
  27             'duration': 313.0,
  28         }
  29     }, {
  30         # An external video hosted on YouTube
  31         'url': 'http://mymedia.yam.com/m/3598173',
  32         'md5': '0238ceec479c654e8c2f1223755bf3e9',
  33         'info_dict': {
  34             'id': 'pJ2Deys283c',
  35             'ext': 'mp4',
  36             'upload_date': '20150202',
  37             'uploader': '新莊社大瑜伽社',
  38             'description': 'md5:f5cc72f0baf259a70fb731654b0d2eff',
  39             'uploader_id': '2323agoy',
  40             'title': '外婆的澎湖灣KTV-潘安邦',
  41         }
  42     }]
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46         page = self._download_webpage(url, video_id)
  47
  48         # Is it hosted externally on YouTube?
  49         youtube_url = self._html_search_regex(
  50             r'<embed src="(http://www.youtube.com/[^"]+)"',
  51             page, 'YouTube url', default=None)
  52         if youtube_url:
  53             return self.url_result(youtube_url, 'Youtube')
  54
  55         api_page = self._download_webpage(
  56             'http://mymedia.yam.com/api/a/?pID=' + video_id, video_id,
  57             note='Downloading API page')
  58         api_result_obj = compat_urlparse.parse_qs(api_page)
  59
  60         uploader_id = self._html_search_regex(
  61             r'<!-- 發表作者 -->：[\n ]+<a href="/([a-z]+)"',
  62             page, 'uploader id', fatal=False)
  63         mobj = re.search(r'<!-- 發表於 -->(?P<mon>[A-Z][a-z]{2})  ' +
  64                          r'(?P<day>\d{1,2}), (?P<year>\d{4})', page)
  65         if mobj:
  66             upload_date = '%s%02d%02d' % (
  67                 mobj.group('year'),
  68                 month_by_abbreviation(mobj.group('mon')),
  69                 int(mobj.group('day')))
  70         else:
  71             upload_date = None
  72         duration = float_or_none(api_result_obj['totaltime'][0], scale=1000)
  73
  74         return {
  75             'id': video_id,
  76             'url': api_result_obj['mp3file'][0],
  77             'title': self._html_search_meta('description', page),
  78             'duration': duration,
  79             'uploader_id': uploader_id,
  80             'upload_date': upload_date,
  81         }