X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fqqmusic.py;h=8ead04606e8ac8f1e334248b47916dcf550cf333;hb=b65e5bb72f48c57316ac0166141810bd1e7b2de8;hp=4b4ef499394b538447097e86dc303fba0b9f5178;hpb=55e5841f14131ab61359535fdcc44e1564d555b8;p=youtube-dl diff --git a/youtube_dl/extractor/qqmusic.py b/youtube_dl/extractor/qqmusic.py index 4b4ef4993..8ead04606 100644 --- a/youtube_dl/extractor/qqmusic.py +++ b/youtube_dl/extractor/qqmusic.py @@ -9,13 +9,14 @@ from .common import InfoExtractor from ..utils import ( strip_jsonp, unescapeHTML, - js_to_json, + clean_html, ) from ..compat import compat_urllib_request class QQMusicIE(InfoExtractor): IE_NAME = 'qqmusic' + IE_DESC = 'QQ音乐' _VALID_URL = r'http://y.qq.com/#type=song&mid=(?P[0-9A-Za-z]+)' _TESTS = [{ 'url': 'http://y.qq.com/#type=song&mid=004295Et37taLD', @@ -27,12 +28,26 @@ class QQMusicIE(InfoExtractor): 'upload_date': '20141227', 'creator': '林俊杰', 'description': 'md5:d327722d0361576fde558f1ac68a7065', + 'thumbnail': 're:^https?://.*\.jpg$', + } + }, { + 'note': 'There is no mp3-320 version of this song.', + 'url': 'http://y.qq.com/#type=song&mid=004MsGEo3DdNxV', + 'md5': 'fa3926f0c585cda0af8fa4f796482e3e', + 'info_dict': { + 'id': '004MsGEo3DdNxV', + 'ext': 'mp3', + 'title': '如果', + 'upload_date': '20050626', + 'creator': '李季美', + 'description': 'md5:46857d5ed62bc4ba84607a805dccf437', + 'thumbnail': 're:^https?://.*\.jpg$', } }] _FORMATS = { - 'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40}, - 'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30}, + 'mp3-320': {'prefix': 'M800', 'ext': 'mp3', 'preference': 40, 'abr': 320}, + 'mp3-128': {'prefix': 'M500', 'ext': 'mp3', 'preference': 30, 'abr': 128}, 'm4a': {'prefix': 'C200', 'ext': 'm4a', 'preference': 10} } @@ -43,6 +58,11 @@ class QQMusicIE(InfoExtractor): curMs = int(time.time() * 1000) % 1000 return int(round(random.random() * 2147483647) * curMs % 1E10) + def _filter_lrc(self, data): + lyrics_expr = r'(\[[0-9]{2}:[0-9]{2}\.[0-9]{2,}\][^\n]*|\[[^\]]*\])' + texts = re.findall(lyrics_expr, data) + return ''.join(i + "\n" for i in texts) + def _real_extract(self, url): mid = self._match_id(url) @@ -69,6 +89,14 @@ class QQMusicIE(InfoExtractor): if lrc_content: lrc_content = lrc_content.replace('\\n', '\n') + thumbnail_url = None + albummid = self._search_regex( + [r'albummid:\'([0-9a-zA-Z]+)\'', r'"albummid":"([0-9a-zA-Z]+)"'], + detail_info_page, 'album mid', default=None) + if albummid: + thumbnail_url = "http://i.gtimg.cn/music/photo/mid_album_500/%s/%s/%s.jpg" \ + % (albummid[-2:-1], albummid[-1], albummid) + guid = self.m_r_get_ruin() vkey = self._download_json( @@ -77,22 +105,37 @@ class QQMusicIE(InfoExtractor): transform_source=strip_jsonp)['key'] formats = [] - for k, sf in self._FORMATS.items(): + for format_id, details in self._FORMATS.items(): formats.append({ 'url': 'http://cc.stream.qqmusic.qq.com/%s%s.%s?vkey=%s&guid=%s&fromtag=0' - % (sf['prefix'], mid, sf['ext'], vkey, guid), - 'format': k, 'format_id': k, 'preference': sf['preference'] + % (details['prefix'], mid, details['ext'], vkey, guid), + 'format': format_id, + 'format_id': format_id, + 'preference': details['preference'], + 'abr': details.get('abr'), }) + self._check_formats(formats, mid) self._sort_formats(formats) - return { + actual_lrc_lyrics = self._filter_lrc(lrc_content) + + info_dict = { 'id': mid, 'formats': formats, 'title': song_name, 'upload_date': publish_time, 'creator': singer, 'description': lrc_content, + 'thumbnail': thumbnail_url } + if actual_lrc_lyrics: + info_dict['subtitles'] = { + 'origin': [{ + 'ext': 'lrc', + 'data': actual_lrc_lyrics, + }] + } + return info_dict class QQPlaylistBaseIE(InfoExtractor): @@ -115,6 +158,7 @@ class QQPlaylistBaseIE(InfoExtractor): class QQMusicSingerIE(QQPlaylistBaseIE): IE_NAME = 'qqmusic:singer' + IE_DESC = 'QQ音乐 - 歌手' _VALID_URL = r'http://y.qq.com/#type=singer&mid=(?P[0-9A-Za-z]+)' _TEST = { 'url': 'http://y.qq.com/#type=singer&mid=001BLpXF2DyJe2', @@ -159,96 +203,130 @@ class QQMusicSingerIE(QQPlaylistBaseIE): class QQMusicAlbumIE(QQPlaylistBaseIE): IE_NAME = 'qqmusic:album' + IE_DESC = 'QQ音乐 - 专辑' _VALID_URL = r'http://y.qq.com/#type=album&mid=(?P[0-9A-Za-z]+)' - _TEST = { - 'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1&play=0', + _TESTS = [{ + 'url': 'http://y.qq.com/#type=album&mid=000gXCTb2AhRR1', 'info_dict': { 'id': '000gXCTb2AhRR1', 'title': '我们都是这样长大的', - 'description': 'md5:d216c55a2d4b3537fe4415b8767d74d6', + 'description': 'md5:179c5dce203a5931970d306aa9607ea6', }, 'playlist_count': 4, - } + }, { + 'url': 'http://y.qq.com/#type=album&mid=002Y5a3b3AlCu3', + 'info_dict': { + 'id': '002Y5a3b3AlCu3', + 'title': '그리고...', + 'description': 'md5:a48823755615508a95080e81b51ba729', + }, + 'playlist_count': 8, + }] def _real_extract(self, url): mid = self._match_id(url) - album_page = self._download_webpage( - self.qq_static_url('album', mid), mid, 'Download album page') + album = self._download_json( + 'http://i.y.qq.com/v8/fcg-bin/fcg_v8_album_info_cp.fcg?albummid=%s&format=json' % mid, + mid, 'Download album page')['data'] - entries = self.get_entries_from_page(album_page) - - album_name = self._html_search_regex( - r"albumname\s*:\s*'([^']+)',", album_page, 'album name', - default=None) - - album_detail = self._html_search_regex( - r'
\s*

((?:[^<>]+(?:
)?)+)

', - album_page, 'album details', default=None) + entries = [ + self.url_result( + 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid'] + ) for song in album['list'] + ] + album_name = album.get('name') + album_detail = album.get('desc') + if album_detail is not None: + album_detail = album_detail.strip() return self.playlist_result(entries, mid, album_name, album_detail) class QQMusicToplistIE(QQPlaylistBaseIE): IE_NAME = 'qqmusic:toplist' + IE_DESC = 'QQ音乐 - 排行榜' _VALID_URL = r'http://y\.qq\.com/#type=toplist&p=(?P(top|global)_[0-9]+)' _TESTS = [{ - 'url': 'http://y.qq.com/#type=toplist&p=global_12', + 'url': 'http://y.qq.com/#type=toplist&p=global_123', 'info_dict': { - 'id': 'global_12', - 'title': 'itunes榜', + 'id': 'global_123', + 'title': '美国iTunes榜', }, 'playlist_count': 10, }, { - 'url': 'http://y.qq.com/#type=toplist&p=top_6', + 'url': 'http://y.qq.com/#type=toplist&p=top_3', 'info_dict': { - 'id': 'top_6', + 'id': 'top_3', 'title': 'QQ音乐巅峰榜·欧美', + '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' + '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' + '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：' + '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' }, 'playlist_count': 100, }, { - 'url': 'http://y.qq.com/#type=toplist&p=global_5', + 'url': 'http://y.qq.com/#type=toplist&p=global_106', 'info_dict': { - 'id': 'global_5', - 'title': '韩国mnet排行榜', + 'id': 'global_106', + 'title': '韩国Mnet榜', }, 'playlist_count': 50, }] - @staticmethod - def strip_qq_jsonp(code): - return js_to_json(re.sub(r'^MusicJsonCallback\((.*?)\)/\*.+?\*/$', r'\1', code)) - def _real_extract(self, url): list_id = self._match_id(url) list_type, num_id = list_id.split("_") - list_page = self._download_webpage( - "http://y.qq.com/y/static/toplist/index/%s.html" % list_id, + toplist_json = self._download_json( + 'http://i.y.qq.com/v8/fcg-bin/fcg_v8_toplist_cp.fcg?type=%s&topid=%s&format=json' + % (list_type, num_id), list_id, 'Download toplist page') - entries = [] - if list_type == 'top': - jsonp_url = "http://y.qq.com/y/static/toplist/json/top/%s/1.js" % num_id - else: - jsonp_url = "http://y.qq.com/y/static/toplist/json/global/%s/1_1.js" % num_id + entries = [ + self.url_result( + 'http://y.qq.com/#type=song&mid=' + song['data']['songmid'], 'QQMusic', song['data']['songmid'] + ) for song in toplist_json['songlist'] + ] - toplist_json = self._download_json( - jsonp_url, list_id, note='Retrieve toplist json', - errnote='Unable to get toplist json', transform_source=self.strip_qq_jsonp) + topinfo = toplist_json.get('topinfo', {}) + list_name = topinfo.get('ListName') + list_description = topinfo.get('info') + return self.playlist_result(entries, list_id, list_name, list_description) - for song in toplist_json['l']: - s = song['s'] - song_mid = s.split("|")[20] - entries.append(self.url_result( - 'http://y.qq.com/#type=song&mid=' + song_mid, 'QQMusic', - song_mid)) - list_name = self._html_search_regex( - r'

([^\']+)

', list_page, 'top list name', - default=None) +class QQMusicPlaylistIE(QQPlaylistBaseIE): + IE_NAME = 'qqmusic:playlist' + IE_DESC = 'QQ音乐 - 歌单' + _VALID_URL = r'http://y\.qq\.com/#type=taoge&id=(?P[0-9]+)' + + _TEST = { + 'url': 'http://y.qq.com/#type=taoge&id=3462654915', + 'info_dict': { + 'id': '3462654915', + 'title': '韩国5月新歌精选下旬', + 'description': 'md5:d2c9d758a96b9888cf4fe82f603121d4', + }, + 'playlist_count': 40, + } + + def _real_extract(self, url): + list_id = self._match_id(url) + + list_json = self._download_json( + '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' + % list_id, list_id, 'Download list page', + transform_source=strip_jsonp)['cdlist'][0] + + entries = [ + self.url_result( + 'http://y.qq.com/#type=song&mid=' + song['songmid'], 'QQMusic', song['songmid'] + ) for song in list_json['songlist'] + ] - return self.playlist_result(entries, list_id, list_name) + list_name = list_json.get('dissname') + list_description = clean_html(unescapeHTML(list_json.get('desc'))) + return self.playlist_result(entries, list_id, list_name, list_description)