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     ExtractorError,
  12     get_element_by_attribute,
  13 )
  14
  15
  16 class YamIE(InfoExtractor):
  17     IE_DESC = '蕃薯藤yam天空部落'
  18     _VALID_URL = r'https?://mymedia\.yam\.com/m/(?P<id>\d+)'
  19
  20     _TESTS = [{
  21         # An audio hosted on Yam
  22         'url': 'http://mymedia.yam.com/m/2283921',
  23         'md5': 'c011b8e262a52d5473d9c2e3c9963b9c',
  24         'info_dict': {
  25             'id': '2283921',
  26             'ext': 'mp3',
  27             'title': '發現 - 趙薇 京華煙雲主題曲',
  28             'description': '發現 - 趙薇 京華煙雲主題曲',
  29             'uploader_id': 'princekt',
  30             'upload_date': '20080807',
  31             'duration': 313.0,
  32         }
  33     }, {
  34         # An external video hosted on YouTube
  35         'url': 'http://mymedia.yam.com/m/3599430',
  36         'md5': '03127cf10d8f35d120a9e8e52e3b17c6',
  37         'info_dict': {
  38             'id': 'CNpEoQlrIgA',
  39             'ext': 'mp4',
  40             'upload_date': '20150306',
  41             'uploader': '新莊社大瑜伽社',
  42             'description': 'md5:11e2e405311633ace874f2e6226c8b17',
  43             'uploader_id': '2323agoy',
  44             'title': '20090412陽明山二子坪-1',
  45         },
  46         'skip': 'Video does not exist',
  47     }, {
  48         'url': 'http://mymedia.yam.com/m/3598173',
  49         'info_dict': {
  50             'id': '3598173',
  51             'ext': 'mp4',
  52         },
  53         'skip': 'cause Yam system error',
  54     }, {
  55         'url': 'http://mymedia.yam.com/m/3599437',
  56         'info_dict': {
  57             'id': '3599437',
  58             'ext': 'mp4',
  59         },
  60         'skip': 'invalid YouTube URL',
  61     }, {
  62         'url': 'http://mymedia.yam.com/m/2373534',
  63         'md5': '7ff74b91b7a817269d83796f8c5890b1',
  64         'info_dict': {
  65             'id': '2373534',
  66             'ext': 'mp3',
  67             'title': '林俊傑&蔡卓妍-小酒窩',
  68             'description': 'md5:904003395a0fcce6cfb25028ff468420',
  69             'upload_date': '20080928',
  70             'uploader_id': 'onliner2',
  71         }
  72     }]
  73
  74     def _real_extract(self, url):
  75         video_id = self._match_id(url)
  76         page = self._download_webpage(url, video_id)
  77
  78         # Check for errors
  79         system_msg = self._html_search_regex(
  80             r'系統訊息(?:<br>|\n|\r)*([^<>]+)<br>', page, 'system message',
  81             default=None)
  82         if system_msg:
  83             raise ExtractorError(system_msg, expected=True)
  84
  85         # Is it hosted externally on YouTube?
  86         youtube_url = self._html_search_regex(
  87             r'<embed src="(http://www.youtube.com/[^"]+)"',
  88             page, 'YouTube url', default=None)
  89         if youtube_url:
  90             return self.url_result(youtube_url, 'Youtube')
  91
  92         title = self._html_search_regex(
  93             r'<h1[^>]+class="heading"[^>]*>\s*(.+)\s*</h1>', page, 'title')
  94
  95         api_page = self._download_webpage(
  96             'http://mymedia.yam.com/api/a/?pID=' + video_id, video_id,
  97             note='Downloading API page')
  98         api_result_obj = compat_urlparse.parse_qs(api_page)
  99
 100         info_table = get_element_by_attribute('class', 'info', page)
 101         uploader_id = self._html_search_regex(
 102             r'<!-- 發表作者 -->：[\n ]+<a href="/([a-z0-9]+)"',
 103             info_table, 'uploader id', fatal=False)
 104         mobj = re.search(r'<!-- 發表於 -->(?P<mon>[A-Z][a-z]{2})\s+' +
 105                          r'(?P<day>\d{1,2}), (?P<year>\d{4})', page)
 106         if mobj:
 107             upload_date = '%s%02d%02d' % (
 108                 mobj.group('year'),
 109                 month_by_abbreviation(mobj.group('mon')),
 110                 int(mobj.group('day')))
 111         else:
 112             upload_date = None
 113         duration = float_or_none(api_result_obj['totaltime'][0], scale=1000)
 114
 115         return {
 116             'id': video_id,
 117             'url': api_result_obj['mp3file'][0],
 118             'title': title,
 119             'description': self._html_search_meta('description', page),
 120             'duration': duration,
 121             'uploader_id': uploader_id,
 122             'upload_date': upload_date,
 123         }