You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

245 lines
8.7 KiB

  1. # coding: utf-8
  2. from __future__ import unicode_literals
  3. import random
  4. import time
  5. import re
  6. from .common import InfoExtractor
  7. from ..utils import (
  8. strip_jsonp,
  9. unescapeHTML,
  10. )
  11. from ..compat import compat_urllib_request
  12. class QQMusicIE(InfoExtractor):
  13. IE_NAME = 'qqmusic'
  14. _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  15. _TESTS = [{
  16. 'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  17. 'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  18. 'info_dict': {
  19. 'id': '004295Et37taLD',
  20. 'ext': 'mp3',
  21. 'title': '可惜没如果',
  22. 'upload_date': '20141227',
  23. 'creator': '林俊杰',
  24. 'description': 'md5:d327722d0361576fde558f1ac68a7065',
  25. }
  26. }]
  27. _FORMATS = {
  28. 'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  29. 'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  30. 'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  31. }
  32. # Reference: m_r_GetRUin() in top_player.js
  33. # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  34. @staticmethod
  35. def m_r_get_ruin():
  36. curMs = int(time.time() * 1000) % 1000
  37. return int(round(random.random() * 2147483647) * curMs % 1E10)
  38. def _real_extract(self, url):
  39. mid = self._match_id(url)
  40. detail_info_page = self._download_webpage(
  41. 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  42. mid, note='Download song detail info',
  43. errnote='Unable to get song detail info', encoding='gbk')
  44. song_name = self._html_search_regex(
  45. r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  46. publish_time = self._html_search_regex(
  47. r'发行时间:(\d{4}-\d{2}-\d{2})', detail_info_page,
  48. 'publish time', default=None)
  49. if publish_time:
  50. publish_time = publish_time.replace('-', '')
  51. singer = self._html_search_regex(
  52. r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  53. lrc_content = self._html_search_regex(
  54. r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  55. detail_info_page, 'LRC lyrics', default=None)
  56. if lrc_content:
  57. lrc_content = lrc_content.replace('\\n', '\n')
  58. guid = self.m_r_get_ruin()
  59. vkey = self._download_json(
  60. 'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  61. mid, note='Retrieve vkey', errnote='Unable to get vkey',
  62. transform_source=strip_jsonp)['key']
  63. formats = []
  64. for format_id, details in self._FORMATS.items():
  65. formats.append({
  66. 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
  67. % (details['prefix'], mid, details['ext'], vkey, guid),
  68. 'format': format_id,
  69. 'format_id': format_id,
  70. 'preference': details['preference'],
  71. 'abr': details.get('abr'),
  72. })
  73. self._sort_formats(formats)
  74. return {
  75. 'id': mid,
  76. 'formats': formats,
  77. 'title': song_name,
  78. 'upload_date': publish_time,
  79. 'creator': singer,
  80. 'description': lrc_content,
  81. }
  82. class QQPlaylistBaseIE(InfoExtractor):
  83. @staticmethod
  84. def qq_static_url(category, mid):
  85. return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
  86. @classmethod
  87. def get_entries_from_page(cls, page):
  88. entries = []
  89. for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
  90. song_mid = unescapeHTML(item).split('|')[-5]
  91. entries.append(cls.url_result(
  92. 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
  93. song_mid))
  94. return entries
  95. class QQMusicSingerIE(QQPlaylistBaseIE):
  96. IE_NAME = 'qqmusic:singer'
  97. _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
  98. _TEST = {
  99. 'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
  100. 'info_dict': {
  101. 'id': '001BLpXF2DyJe2',
  102. 'title': '林俊杰',
  103. 'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
  104. },
  105. 'playlist_count': 12,
  106. }
  107. def _real_extract(self, url):
  108. mid = self._match_id(url)
  109. singer_page = self._download_webpage(
  110. self.qq_static_url('singer', mid), mid, 'Download singer page')
  111. entries = self.get_entries_from_page(singer_page)
  112. singer_name = self._html_search_regex(
  113. r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
  114. default=None)
  115. singer_id = self._html_search_regex(
  116. r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
  117. default=None)
  118. singer_desc = None
  119. if singer_id:
  120. req = compat_urllib_request.Request(
  121. 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
  122. req.add_header(
  123. 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
  124. singer_desc_page = self._download_xml(
  125. req, mid, 'Donwload singer description XML')
  126. singer_desc = singer_desc_page.find('./data/info/desc').text
  127. return self.playlist_result(entries, mid, singer_name, singer_desc)
  128. class QQMusicAlbumIE(QQPlaylistBaseIE):
  129. IE_NAME = 'qqmusic:album'
  130. _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
  131. _TEST = {
  132. 'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
  133. 'info_dict': {
  134. 'id': '000gXCTb2AhRR1',
  135. 'title': '我们都是这样长大的',
  136. 'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
  137. },
  138. 'playlist_count': 4,
  139. }
  140. def _real_extract(self, url):
  141. mid = self._match_id(url)
  142. album_page = self._download_webpage(
  143. self.qq_static_url('album', mid), mid, 'Download album page')
  144. entries = self.get_entries_from_page(album_page)
  145. album_name = self._html_search_regex(
  146. r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
  147. default=None)
  148. album_detail = self._html_search_regex(
  149. r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
  150. album_page, 'album details', default=None)
  151. return self.playlist_result(entries, mid, album_name, album_detail)
  152. class QQMusicToplistIE(QQPlaylistBaseIE):
  153. IE_NAME = 'qqmusic:toplist'
  154. _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
  155. _TESTS = [{
  156. 'url': 'http://y.qq.com/#type=toplist&p=global_123',
  157. 'info_dict': {
  158. 'id': 'global_123',
  159. 'title': '美国iTunes榜',
  160. },
  161. 'playlist_count': 10,
  162. }, {
  163. 'url': 'http://y.qq.com/#type=toplist&p=top_3',
  164. 'info_dict': {
  165. 'id': 'top_3',
  166. 'title': 'QQ音乐巅峰榜·欧美',
  167. 'description': 'QQ音乐巅峰榜·欧美根据用户收听行为自动生成,集结当下最流行的欧美新歌!:更新时间:每周四22点|统'
  168. '计周期:一周(上周四至本周三)|统计对象:三个月内发行的欧美歌曲|统计数量:100首|统计算法:根据'
  169. '歌曲在一周内的有效播放次数,由高到低取前100名(同一歌手最多允许5首歌曲同时上榜)|有效播放次数:'
  170. '登录用户完整播放一首歌曲,记为一次有效播放;同一用户收听同一首歌曲,每天记录为1次有效播放'
  171. },
  172. 'playlist_count': 100,
  173. }, {
  174. 'url': 'http://y.qq.com/#type=toplist&p=global_106',
  175. 'info_dict': {
  176. 'id': 'global_106',
  177. 'title': '韩国Mnet榜',
  178. },
  179. 'playlist_count': 50,
  180. }]
  181. def _real_extract(self, url):
  182. list_id = self._match_id(url)
  183. list_type, num_id = list_id.split("_")
  184. toplist_json = self._download_json(
  185. 'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
  186. % (list_type, num_id),
  187. list_id, 'Download toplist page')
  188. entries = [
  189. self.url_result(
  190. 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
  191. ) for song in toplist_json['songlist']
  192. ]
  193. topinfo = toplist_json.get('topinfo', {})
  194. list_name = topinfo.get('ListName')
  195. list_description = topinfo.get('info')
  196. return self.playlist_result(entries, list_id, list_name, list_description)