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