X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fkuwo.py;h=cc5b2a1c177addf3465f2e9b2f65569f7617053b;hb=067aa17edf5a46a8cbc4d6b90864eddf051fa2bc;hp=6a96a1aa42121a9755cfeee5dcf7e88b866b5dd6;hpb=8f73e89ca0ecde0a8bbd1f1463e9a06a53c6b573;p=youtube-dl

diff --git a/youtube_dl/extractor/kuwo.py b/youtube_dl/extractor/kuwo.py
index 6a96a1aa4..cc5b2a1c1 100644
--- a/youtube_dl/extractor/kuwo.py
+++ b/youtube_dl/extractor/kuwo.py
@@ -2,19 +2,64 @@
 from __future__ import unicode_literals
 
 import re
-import itertools
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     get_element_by_id,
     clean_html,
     ExtractorError,
+    InAdvancePagedList,
+    remove_start,
 )
 
 
-class KuwoIE(InfoExtractor):
+class KuwoBaseIE(InfoExtractor):
+    _FORMATS = [
+        {'format': 'ape', 'ext': 'ape', 'preference': 100},
+        {'format': 'mp3-320', 'ext': 'mp3', 'br': '320kmp3', 'abr': 320, 'preference': 80},
+        {'format': 'mp3-192', 'ext': 'mp3', 'br': '192kmp3', 'abr': 192, 'preference': 70},
+        {'format': 'mp3-128', 'ext': 'mp3', 'br': '128kmp3', 'abr': 128, 'preference': 60},
+        {'format': 'wma', 'ext': 'wma', 'preference': 20},
+        {'format': 'aac', 'ext': 'aac', 'abr': 48, 'preference': 10}
+    ]
+
+    def _get_formats(self, song_id, tolerate_ip_deny=False):
+        formats = []
+        for file_format in self._FORMATS:
+            query = {
+                'format': file_format['ext'],
+                'br': file_format.get('br', ''),
+                'rid': 'MUSIC_%s' % song_id,
+                'type': 'convert_url',
+                'response': 'url'
+            }
+
+            song_url = self._download_webpage(
+                'http://antiserver.kuwo.cn/anti.s',
+                song_id, note='Download %s url info' % file_format['format'],
+                query=query, headers=self.geo_verification_headers(),
+            )
+
+            if song_url == 'IPDeny' and not tolerate_ip_deny:
+                raise ExtractorError('This song is blocked in this region', expected=True)
+
+            if song_url.startswith('http://') or song_url.startswith('https://'):
+                formats.append({
+                    'url': song_url,
+                    'format_id': file_format['format'],
+                    'format': file_format['format'],
+                    'preference': file_format['preference'],
+                    'abr': file_format.get('abr'),
+                })
+
+        return formats
+
+
+class KuwoIE(KuwoBaseIE):
     IE_NAME = 'kuwo:song'
-    _VALID_URL = r'http://www\.kuwo\.cn/yinyue/(?P<id>[0-9]+?)/'
+    IE_DESC = '酷我音乐'
+    _VALID_URL = r'https?://(?:www\.)?kuwo\.cn/yinyue/(?P<id>\d+)'
     _TESTS = [{
         'url': 'http://www.kuwo.cn/yinyue/635632/',
         'info_dict': {
@@ -25,78 +70,59 @@ class KuwoIE(InfoExtractor):
             'upload_date': '20080122',
             'description': 'md5:ed13f58e3c3bf3f7fd9fbc4e5a7aa75c'
         },
+        'skip': 'this song has been offline because of copyright issues',
     }, {
         'url': 'http://www.kuwo.cn/yinyue/6446136/',
         'info_dict': {
             'id': '6446136',
             'ext': 'mp3',
             'title': '心',
+            'description': 'md5:5d0e947b242c35dc0eb1d2fce9fbf02c',
             'creator': 'IU',
             'upload_date': '20150518',
         },
         'params': {
-            'format': 'mp3-320'
+            'format': 'mp3-320',
         },
+    }, {
+        'url': 'http://www.kuwo.cn/yinyue/3197154?catalog=yueku2016',
+        'only_matching': True,
     }]
-    _FORMATS = [
-        {'format': 'ape', 'ext': 'ape', 'preference': 100},
-        {'format': 'mp3-320', 'ext': 'mp3', 'br': '320kmp3', 'abr': 320, 'preference': 80},
-        {'format': 'mp3-192', 'ext': 'mp3', 'br': '192kmp3', 'abr': 192, 'preference': 70},
-        {'format': 'mp3-128', 'ext': 'mp3', 'br': '128kmp3', 'abr': 128, 'preference': 60},
-        {'format': 'wma', 'ext': 'wma', 'preference': 20},
-        {'format': 'aac', 'ext': 'aac', 'abr': 48, 'preference': 10}
-    ]
-
-    def _get_formats(self, song_id):
-        formats = []
-        for file_format in self._FORMATS:
-            song_url = self._download_webpage(
-                "http://antiserver.kuwo.cn/anti.s?format=%s&br=%s&rid=MUSIC_%s&type=convert_url&response=url" %
-                (file_format['ext'], file_format.get('br', ''), song_id),
-                song_id, note="Download %s url info" % file_format["format"],
-            )
-            if song_url.startswith('http://') or song_url.startswith('https://'):
-                formats.append({
-                    'url': song_url,
-                    'format_id': file_format['format'],
-                    'format': file_format['format'],
-                    'preference': file_format['preference'],
-                    'abr': file_format.get('abr'),
-                })
-        self._sort_formats(formats)
-        return formats
 
     def _real_extract(self, url):
         song_id = self._match_id(url)
-        webpage = self._download_webpage(
+        webpage, urlh = self._download_webpage_handle(
             url, song_id, note='Download song detail info',
             errnote='Unable to get song detail info')
+        if song_id not in urlh.geturl() or '对不起，该歌曲由于版权问题已被下线，将返回网站首页' in webpage:
+            raise ExtractorError('this song has been offline because of copyright issues', expected=True)
 
         song_name = self._html_search_regex(
-            r'<h1 title="(.+?)">', webpage, 'song name')
-        singer_name = self._html_search_regex(
-            r'<div class="s_img">.+?title="(.+?)".+?</div>', webpage, 'singer name',
-            flags=re.DOTALL, default=None)
-        lrc_content = clean_html(get_element_by_id("lrcContent", webpage))
+            r'<p[^>]+id="lrcName">([^<]+)</p>', webpage, 'song name')
+        singer_name = remove_start(self._html_search_regex(
+            r'<a[^>]+href="http://www\.kuwo\.cn/artist/content\?name=([^"]+)">',
+            webpage, 'singer name', fatal=False), '歌手')
+        lrc_content = clean_html(get_element_by_id('lrcContent', webpage))
         if lrc_content == '暂无':     # indicates no lyrics
             lrc_content = None
 
         formats = self._get_formats(song_id)
+        self._sort_formats(formats)
 
         album_id = self._html_search_regex(
-            r'<p class="album" title=".+?">.+?<a href="http://www\.kuwo\.cn/album/([0-9]+)/" ',
-            webpage, 'album id', default=None, fatal=False)
+            r'<a[^>]+href="http://www\.kuwo\.cn/album/(\d+)/"',
+            webpage, 'album id', fatal=False)
 
         publish_time = None
         if album_id is not None:
             album_info_page = self._download_webpage(
-                "http://www.kuwo.cn/album/%s/" % album_id, song_id,
+                'http://www.kuwo.cn/album/%s/' % album_id, song_id,
                 note='Download album detail info',
                 errnote='Unable to get album detail info')
 
             publish_time = self._html_search_regex(
                 r'发行时间：(\d{4}-\d{2}-\d{2})', album_info_page,
-                'publish time', default=None)
+                'publish time', fatal=False)
             if publish_time:
                 publish_time = publish_time.replace('-', '')
 
@@ -112,13 +138,14 @@ class KuwoIE(InfoExtractor):
 
 class KuwoAlbumIE(InfoExtractor):
     IE_NAME = 'kuwo:album'
-    _VALID_URL = r'http://www\.kuwo\.cn/album/(?P<id>[0-9]+?)/'
+    IE_DESC = '酷我音乐 - 专辑'
+    _VALID_URL = r'https?://(?:www\.)?kuwo\.cn/album/(?P<id>\d+?)/'
     _TEST = {
         'url': 'http://www.kuwo.cn/album/502294/',
         'info_dict': {
             'id': '502294',
-            'title': 'M',
-            'description': 'md5:6a7235a84cc6400ec3b38a7bdaf1d60c',
+            'title': 'Made\xa0Series\xa0《M》',
+            'description': 'md5:d463f0d8a0ff3c3ea3d6ed7452a9483f',
         },
         'playlist_count': 2,
     }
@@ -131,15 +158,15 @@ class KuwoAlbumIE(InfoExtractor):
             errnote='Unable to get album info')
 
         album_name = self._html_search_regex(
-            r'<div class="comm".+?<h1 title="(.+?)">.+?</h1>', webpage,
-            'album name', flags=re.DOTALL)
-        album_intro = clean_html(
-            re.sub(r'^.+简介：', '', get_element_by_id("intro", webpage).strip()))
+            r'<div[^>]+class="comm"[^<]+<h1[^>]+title="([^"]+)"', webpage,
+            'album name')
+        album_intro = remove_start(
+            clean_html(get_element_by_id('intro', webpage)),
+            '%s简介：' % album_name)
 
         entries = [
-            self.url_result("http://www.kuwo.cn/yinyue/%s/" % song_id, 'Kuwo', song_id)
-            for song_id in re.findall(
-                r'<p class="listen"><a href="http://www\.kuwo\.cn/yinyue/([0-9]+)/" target="_blank" title="试听.*?"></a></p>',
+            self.url_result(song_url, 'Kuwo') for song_url in re.findall(
+                r'<p[^>]+class="listen"><a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+/)"',
                 webpage)
         ]
         return self.playlist_result(entries, album_id, album_name, album_intro)
@@ -147,15 +174,14 @@ class KuwoAlbumIE(InfoExtractor):
 
 class KuwoChartIE(InfoExtractor):
     IE_NAME = 'kuwo:chart'
-    _VALID_URL = r'http://yinyue\.kuwo\.cn/billboard_(?P<id>.+?).htm'
+    IE_DESC = '酷我音乐 - 排行榜'
+    _VALID_URL = r'https?://yinyue\.kuwo\.cn/billboard_(?P<id>[^.]+).htm'
     _TEST = {
         'url': 'http://yinyue.kuwo.cn/billboard_香港中文龙虎榜.htm',
         'info_dict': {
             'id': '香港中文龙虎榜',
-            'title': '香港中文龙虎榜',
-            'description': 're:[0-9]{4}第[0-9]{2}期',
         },
-        'playlist_mincount': 10,
+        'playlist_mincount': 7,
     }
 
     def _real_extract(self, url):
@@ -164,31 +190,35 @@ class KuwoChartIE(InfoExtractor):
             url, chart_id, note='Download chart info',
             errnote='Unable to get chart info')
 
-        chart_name = self._html_search_regex(
-            r'<h1 class="unDis">(.+?)</h1>', webpage, 'chart name')
-
-        chart_desc = self._html_search_regex(
-            r'<p class="tabDef">([0-9]{4}第[0-9]{2}期)</p>', webpage, 'chart desc')
-
         entries = [
-            self.url_result("http://www.kuwo.cn/yinyue/%s/" % song_id, 'Kuwo', song_id)
-            for song_id in re.findall(
-                r'<a href="http://www\.kuwo\.cn/yinyue/([0-9]+)/" .+?>.+?</a>', webpage)
+            self.url_result(song_url, 'Kuwo') for song_url in re.findall(
+                r'<a[^>]+href="(http://www\.kuwo\.cn/yinyue/\d+)', webpage)
         ]
-        return self.playlist_result(entries, chart_id, chart_name, chart_desc)
+        return self.playlist_result(entries, chart_id)
 
 
 class KuwoSingerIE(InfoExtractor):
     IE_NAME = 'kuwo:singer'
-    _VALID_URL = r'http://www\.kuwo\.cn/mingxing/(?P<id>[^/]+?)/$'
-    _TEST = {
+    IE_DESC = '酷我音乐 - 歌手'
+    _VALID_URL = r'https?://(?:www\.)?kuwo\.cn/mingxing/(?P<id>[^/]+)'
+    _TESTS = [{
         'url': 'http://www.kuwo.cn/mingxing/bruno+mars/',
         'info_dict': {
             'id': 'bruno+mars',
-            'title': 'Bruno Mars',
+            'title': 'Bruno\xa0Mars',
         },
-        'playlist_count': 10,
-    }
+        'playlist_mincount': 329,
+    }, {
+        'url': 'http://www.kuwo.cn/mingxing/Ali/music.htm',
+        'info_dict': {
+            'id': 'Ali',
+            'title': 'Ali',
+        },
+        'playlist_mincount': 95,
+        'skip': 'Regularly stalls travis build',  # See https://travis-ci.org/ytdl-org/youtube-dl/jobs/78878540
+    }]
+
+    PAGE_SIZE = 15
 
     def _real_extract(self, url):
         singer_id = self._match_id(url)
@@ -197,59 +227,37 @@ class KuwoSingerIE(InfoExtractor):
             errnote='Unable to get singer info')
 
         singer_name = self._html_search_regex(
-            r'姓名：<span>(.+?)</span>', webpage, 'singer name')
-
-        entries = [
-            self.url_result("http://www.kuwo.cn/yinyue/%s/" % song_id, 'Kuwo', song_id)
-            for song_id in re.findall(
-                r'<a href="http://www\.kuwo\.cn/yinyue/([0-9]+)/" .+?>.+?</a>',
-                webpage, flags=re.DOTALL)
-        ]
-        return self.playlist_result(entries, singer_id, singer_name)
+            r'<h1>([^<]+)</h1>', webpage, 'singer name')
 
+        artist_id = self._html_search_regex(
+            r'data-artistid="(\d+)"', webpage, 'artist id')
 
-class KuwoSingerMusicIE(InfoExtractor):
-    IE_NAME = 'kuwo:singermusic'
-    _VALID_URL = r'http://www\.kuwo\.cn/mingxing/(?P<id>[^/]+?)/music(_[0-9]+)?.htm'
-    _TEST = {
-        'url': 'http://www.kuwo.cn/mingxing/Ali/music.htm',
-        'info_dict': {
-            'id': 'Ali',
-            'title': 'Ali的热门歌曲',
-        },
-        'playlist_mincount': 95,
-    }
-
-    def _real_extract(self, url):
-        singer_id = self._match_id(url)
+        page_count = int(self._html_search_regex(
+            r'data-page="(\d+)"', webpage, 'page count'))
 
-        list_name = None
-        entries = []
-        for page_num in itertools.count(1):
+        def page_func(page_num):
             webpage = self._download_webpage(
-                'http://www.kuwo.cn/mingxing/%s/music_%d.htm' % (singer_id, page_num),
-                singer_id, note='Download song list page #%d' % page_num,
-                errnote='Unable to get song list page #%d' % page_num)
-
-            if list_name is None:
-                list_name = self._html_search_regex(
-                    r'<h1>([^<>]+)<span>', webpage, 'list name')
-
-            entries.extend([
-                self.url_result("http://www.kuwo.cn/yinyue/%s/" % song_id, 'Kuwo', song_id)
-                for song_id in re.findall(
-                    r'<p class="m_name"><a href="http://www\.kuwo\.cn/yinyue/([0-9]+)/',
+                'http://www.kuwo.cn/artist/contentMusicsAjax',
+                singer_id, note='Download song list page #%d' % (page_num + 1),
+                errnote='Unable to get song list page #%d' % (page_num + 1),
+                query={'artistId': artist_id, 'pn': page_num, 'rn': self.PAGE_SIZE})
+
+            return [
+                self.url_result(compat_urlparse.urljoin(url, song_url), 'Kuwo')
+                for song_url in re.findall(
+                    r'<div[^>]+class="name"><a[^>]+href="(/yinyue/\d+)',
                     webpage)
-            ])
-            if not re.search(r'<a href="[^"]+">下一页</a>', webpage):
-                break
+            ]
+
+        entries = InAdvancePagedList(page_func, page_count, self.PAGE_SIZE)
 
-        return self.playlist_result(entries, singer_id, list_name)
+        return self.playlist_result(entries, singer_id, singer_name)
 
 
 class KuwoCategoryIE(InfoExtractor):
     IE_NAME = 'kuwo:category'
-    _VALID_URL = r'http://yinyue\.kuwo\.cn/yy/cinfo_(?P<id>[0-9]+?).htm'
+    IE_DESC = '酷我音乐 - 分类'
+    _VALID_URL = r'https?://yinyue\.kuwo\.cn/yy/cinfo_(?P<id>\d+?).htm'
     _TEST = {
         'url': 'http://yinyue.kuwo.cn/yy/cinfo_86375.htm',
         'info_dict': {
@@ -257,7 +265,7 @@ class KuwoCategoryIE(InfoExtractor):
             'title': '八十年代精选',
             'description': '这些都是属于八十年代的回忆！',
         },
-        'playlist_count': 30,
+        'playlist_mincount': 24,
     }
 
     def _real_extract(self, url):
@@ -267,36 +275,44 @@ class KuwoCategoryIE(InfoExtractor):
             errnote='Unable to get category info')
 
         category_name = self._html_search_regex(
-            r'<h1 title="([^<>]+?)">[^<>]+?</h1>', webpage, 'category name')
-        
-        category_desc = re.sub(
-            r'^.+简介：', '', get_element_by_id("intro", webpage).strip())
-        
+            r'<h1[^>]+title="([^<>]+?)">[^<>]+?</h1>', webpage, 'category name')
+
+        category_desc = remove_start(
+            get_element_by_id('intro', webpage).strip(),
+            '%s简介：' % category_name)
+        if category_desc == '暂无':
+            category_desc = None
+
         jsonm = self._parse_json(self._html_search_regex(
-            r'var jsonm = (\{.+?\});', webpage, 'category songs'), category_id)
+            r'var\s+jsonm\s*=\s*([^;]+);', webpage, 'category songs'), category_id)
 
         entries = [
-            self.url_result(
-                "http://www.kuwo.cn/yinyue/%s/" % song['musicrid'],
-                'Kuwo', song['musicrid'])
+            self.url_result('http://www.kuwo.cn/yinyue/%s/' % song['musicrid'], 'Kuwo')
             for song in jsonm['musiclist']
         ]
         return self.playlist_result(entries, category_id, category_name, category_desc)
 
 
-class KuwoMvIE(KuwoIE):
+class KuwoMvIE(KuwoBaseIE):
     IE_NAME = 'kuwo:mv'
-    _VALID_URL = r'http://www\.kuwo\.cn/mv/(?P<id>[0-9]+?)/'
-    _TESTS = [{
+    IE_DESC = '酷我音乐 - MV'
+    _VALID_URL = r'https?://(?:www\.)?kuwo\.cn/mv/(?P<id>\d+?)/'
+    _TEST = {
         'url': 'http://www.kuwo.cn/mv/6480076/',
         'info_dict': {
             'id': '6480076',
-            'ext': 'mkv',
-            'title': '我们家MV',
+            'ext': 'mp4',
+            'title': 'My HouseMV',
             'creator': '2PM',
         },
-    }]
-    _FORMATS = KuwoIE._FORMATS + [
+        # In this video, music URLs (anti.s) are blocked outside China and
+        # USA, while the MV URL (mvurl) is available globally, so force the MV
+        # URL for consistent results in different countries
+        'params': {
+            'format': 'mv',
+        },
+    }
+    _FORMATS = KuwoBaseIE._FORMATS + [
         {'format': 'mkv', 'ext': 'mkv', 'preference': 250},
         {'format': 'mp4', 'ext': 'mp4', 'preference': 200},
     ]
@@ -308,15 +324,25 @@ class KuwoMvIE(KuwoIE):
             errnote='Unable to get mv detail info: %s' % song_id)
 
         mobj = re.search(
-            r'<h1 title="(?P<song>.+?)">[^<>]+<span .+?title="(?P<singer>.+?)".+?>[^<>]+</span></h1>',
+            r'<h1[^>]+title="(?P<song>[^"]+)">[^<]+<span[^>]+title="(?P<singer>[^"]+)"',
             webpage)
         if mobj:
             song_name = mobj.group('song')
             singer_name = mobj.group('singer')
         else:
-            raise ExtractorError("Unable to find song or singer names")
+            raise ExtractorError('Unable to find song or singer names')
 
-        formats = self._get_formats(song_id)
+        formats = self._get_formats(song_id, tolerate_ip_deny=True)
+
+        mv_url = self._download_webpage(
+            'http://www.kuwo.cn/yy/st/mvurl?rid=MUSIC_%s' % song_id,
+            song_id, note='Download %s MV URL' % song_id)
+        formats.append({
+            'url': mv_url,
+            'format_id': 'mv',
+        })
+
+        self._sort_formats(formats)
 
         return {
             'id': song_id,