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.

279 lines
9.9 KiB

10 years ago
  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. clean_html,
  11. )
  12. from ..compat import compat_urllib_request
  13. class QQMusicIE(InfoExtractor):
  14. IE_NAME = 'qqmusic'
  15. _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P<id>[0-9A-Za-z]+)'
  16. _TESTS = [{
  17. 'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD',
  18. 'md5': '9ce1c1c8445f561506d2e3cfb0255705',
  19. 'info_dict': {
  20. 'id': '004295Et37taLD',
  21. 'ext': 'mp3',
  22. 'title': '可惜没如果',
  23. 'upload_date': '20141227',
  24. 'creator': '林俊杰',
  25. 'description': 'md5:d327722d0361576fde558f1ac68a7065',
  26. }
  27. }]
  28. _FORMATS = {
  29. 'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320},
  30. 'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128},
  31. 'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10}
  32. }
  33. # Reference: m_r_GetRUin() in top_player.js
  34. # http://imgcache.gtimg.cn/music/portal_v3/y/top_player.js
  35. @staticmethod
  36. def m_r_get_ruin():
  37. curMs = int(time.time() * 1000) % 1000
  38. return int(round(random.random() * 2147483647) * curMs % 1E10)
  39. def _real_extract(self, url):
  40. mid = self._match_id(url)
  41. detail_info_page = self._download_webpage(
  42. 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_yqq_song_detail_info.fcg?songmid=%s&play=0' % mid,
  43. mid, note='Download song detail info',
  44. errnote='Unable to get song detail info', encoding='gbk')
  45. song_name = self._html_search_regex(
  46. r"songname:\s*'([^']+)'", detail_info_page, 'song name')
  47. publish_time = self._html_search_regex(
  48. r'发行时间:(\d{4}-\d{2}-\d{2})', detail_info_page,
  49. 'publish time', default=None)
  50. if publish_time:
  51. publish_time = publish_time.replace('-', '')
  52. singer = self._html_search_regex(
  53. r"singer:\s*'([^']+)", detail_info_page, 'singer', default=None)
  54. lrc_content = self._html_search_regex(
  55. r'<div class="content" id="lrc_content"[^<>]*>([^<>]+)</div>',
  56. detail_info_page, 'LRC lyrics', default=None)
  57. if lrc_content:
  58. lrc_content = lrc_content.replace('\\n', '\n')
  59. guid = self.m_r_get_ruin()
  60. vkey = self._download_json(
  61. 'http://base.music.qq.com/fcgi-bin/fcg_musicexpress.fcg?json=3&guid=%s' % guid,
  62. mid, note='Retrieve vkey', errnote='Unable to get vkey',
  63. transform_source=strip_jsonp)['key']
  64. formats = []
  65. for format_id, details in self._FORMATS.items():
  66. formats.append({
  67. 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0'
  68. % (details['prefix'], mid, details['ext'], vkey, guid),
  69. 'format': format_id,
  70. 'format_id': format_id,
  71. 'preference': details['preference'],
  72. 'abr': details.get('abr'),
  73. })
  74. self._sort_formats(formats)
  75. return {
  76. 'id': mid,
  77. 'formats': formats,
  78. 'title': song_name,
  79. 'upload_date': publish_time,
  80. 'creator': singer,
  81. 'description': lrc_content,
  82. }
  83. class QQPlaylistBaseIE(InfoExtractor):
  84. @staticmethod
  85. def qq_static_url(category, mid):
  86. return 'http://y.qq.com/y/static/%s/%s/%s/%s.html' % (category, mid[-2], mid[-1], mid)
  87. @classmethod
  88. def get_entries_from_page(cls, page):
  89. entries = []
  90. for item in re.findall(r'class="data"[^<>]*>([^<>]+)</', page):
  91. song_mid = unescapeHTML(item).split('|')[-5]
  92. entries.append(cls.url_result(
  93. 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic',
  94. song_mid))
  95. return entries
  96. class QQMusicSingerIE(QQPlaylistBaseIE):
  97. IE_NAME = 'qqmusic:singer'
  98. _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P<id>[0-9A-Za-z]+)'
  99. _TEST = {
  100. 'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2',
  101. 'info_dict': {
  102. 'id': '001BLpXF2DyJe2',
  103. 'title': '林俊杰',
  104. 'description': 'md5:2a222d89ba4455a3af19940c0481bb78',
  105. },
  106. 'playlist_count': 12,
  107. }
  108. def _real_extract(self, url):
  109. mid = self._match_id(url)
  110. singer_page = self._download_webpage(
  111. self.qq_static_url('singer', mid), mid, 'Download singer page')
  112. entries = self.get_entries_from_page(singer_page)
  113. singer_name = self._html_search_regex(
  114. r"singername\s*:\s*'([^']+)'", singer_page, 'singer name',
  115. default=None)
  116. singer_id = self._html_search_regex(
  117. r"singerid\s*:\s*'([0-9]+)'", singer_page, 'singer id',
  118. default=None)
  119. singer_desc = None
  120. if singer_id:
  121. req = compat_urllib_request.Request(
  122. 'http://s.plcloud.music.qq.com/fcgi-bin/fcg_get_singer_desc.fcg?utf8=1&outCharset=utf-8&format=xml&singerid=%s' % singer_id)
  123. req.add_header(
  124. 'Referer', 'http://s.plcloud.music.qq.com/xhr_proxy_utf8.html')
  125. singer_desc_page = self._download_xml(
  126. req, mid, 'Donwload singer description XML')
  127. singer_desc = singer_desc_page.find('./data/info/desc').text
  128. return self.playlist_result(entries, mid, singer_name, singer_desc)
  129. class QQMusicAlbumIE(QQPlaylistBaseIE):
  130. IE_NAME = 'qqmusic:album'
  131. _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P<id>[0-9A-Za-z]+)'
  132. _TEST = {
  133. 'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0',
  134. 'info_dict': {
  135. 'id': '000gXCTb2AhRR1',
  136. 'title': '我们都是这样长大的',
  137. 'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6',
  138. },
  139. 'playlist_count': 4,
  140. }
  141. def _real_extract(self, url):
  142. mid = self._match_id(url)
  143. album_page = self._download_webpage(
  144. self.qq_static_url('album', mid), mid, 'Download album page')
  145. entries = self.get_entries_from_page(album_page)
  146. album_name = self._html_search_regex(
  147. r"albumname\s*:\s*'([^']+)',", album_page, 'album name',
  148. default=None)
  149. album_detail = self._html_search_regex(
  150. r'<div class="album_detail close_detail">\s*<p>((?:[^<>]+(?:<br />)?)+)</p>',
  151. album_page, 'album details', default=None)
  152. return self.playlist_result(entries, mid, album_name, album_detail)
  153. class QQMusicToplistIE(QQPlaylistBaseIE):
  154. IE_NAME = 'qqmusic:toplist'
  155. _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P<id>(top|global)_[0-9]+)'
  156. _TESTS = [{
  157. 'url': 'http://y.qq.com/#type=toplist&p=global_123',
  158. 'info_dict': {
  159. 'id': 'global_123',
  160. 'title': '美国iTunes榜',
  161. },
  162. 'playlist_count': 10,
  163. }, {
  164. 'url': 'http://y.qq.com/#type=toplist&p=top_3',
  165. 'info_dict': {
  166. 'id': 'top_3',
  167. 'title': 'QQ音乐巅峰榜·欧美',
  168. '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'
  169. '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'
  170. '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:'
  171. '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'
  172. },
  173. 'playlist_count': 100,
  174. }, {
  175. 'url': 'http://y.qq.com/#type=toplist&p=global_106',
  176. 'info_dict': {
  177. 'id': 'global_106',
  178. 'title': '韩国Mnet榜',
  179. },
  180. 'playlist_count': 50,
  181. }]
  182. def _real_extract(self, url):
  183. list_id = self._match_id(url)
  184. list_type, num_id = list_id.split("_")
  185. toplist_json = self._download_json(
  186. 'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json'
  187. % (list_type, num_id),
  188. list_id, 'Download toplist page')
  189. entries = [
  190. self.url_result(
  191. 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid']
  192. ) for song in toplist_json['songlist']
  193. ]
  194. topinfo = toplist_json.get('topinfo', {})
  195. list_name = topinfo.get('ListName')
  196. list_description = topinfo.get('info')
  197. return self.playlist_result(entries, list_id, list_name, list_description)
  198. class QQMusicPlaylistIE(QQPlaylistBaseIE):
  199. IE_NAME = 'qqmusic:playlist'
  200. _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P<id>[0-9]+)'
  201. _TEST = {
  202. 'url': 'http://y.qq.com/#type=taoge&id=3462654915',
  203. 'info_dict': {
  204. 'id': '3462654915',
  205. 'title': '韩国5月新歌精选下旬',
  206. 'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4',
  207. },
  208. 'playlist_count': 40,
  209. }
  210. def _real_extract(self, url):
  211. list_id = self._match_id(url)
  212. list_json = self._download_json(
  213. 'http://i.y.qq.com/qzone-music/fcg-bin/fcg_ucc_getcdinfo_byids_cp.fcg?type=1&json=1&utf8=1&onlysong=0&disstid=%s'
  214. % list_id, list_id, 'Download list page',
  215. transform_source=strip_jsonp)['cdlist'][0]
  216. entries = [
  217. self.url_result(
  218. 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid']
  219. ) for song in list_json['songlist']
  220. ]
  221. list_name = list_json.get('dissname')
  222. list_description = clean_html(unescapeHTML(list_json.get('desc')))
  223. return self.playlist_result(entries, list_id, list_name, list_description)