]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/mixcloud.py
Update README.md
[youtubedl] / youtube_dl / extractor / mixcloud.py
index 07d194562e77044a8d8d87138ed32205842a1a25..9759560f1bee5642c7e4f11c8aa8891fd4b3ed7d 100644 (file)
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 
+import itertools
 import re
 
 from .common import InfoExtractor
 from ..compat import (
 import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse,
+    compat_b64decode,
+    compat_chr,
+    compat_ord,
+    compat_str,
+    compat_urllib_parse_unquote,
+    compat_zip
 )
 from ..utils import (
 )
 from ..utils import (
-    ExtractorError,
-    HEADRequest,
     int_or_none,
     parse_iso8601,
     int_or_none,
     parse_iso8601,
+    strip_or_none,
+    try_get,
 )
 
 
 )
 
 
-class MixcloudIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)'
+class MixcloudBaseIE(InfoExtractor):
+    def _call_api(self, object_type, object_fields, display_id, username, slug=None):
+        lookup_key = object_type + 'Lookup'
+        return self._download_json(
+            'https://www.mixcloud.com/graphql', display_id, query={
+                'query': '''{
+  %s(lookup: {username: "%s"%s}) {
+    %s
+  }
+}''' % (lookup_key, username, ', slug: "%s"' % slug if slug else '', object_fields)
+            })['data'][lookup_key]
+
+
+class MixcloudIE(MixcloudBaseIE):
+    _VALID_URL = r'https?://(?:(?:www|beta|m)\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
     IE_NAME = 'mixcloud'
 
     IE_NAME = 'mixcloud'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
         'info_dict': {
         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
         'info_dict': {
-            'id': 'dholbach-cryptkeeper',
-            'ext': 'mp3',
+            'id': 'dholbach_cryptkeeper',
+            'ext': 'm4a',
             'title': 'Cryptkeeper',
             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
             'uploader': 'Daniel Holbach',
             'uploader_id': 'dholbach',
             'title': 'Cryptkeeper',
             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
             'uploader': 'Daniel Holbach',
             'uploader_id': 'dholbach',
-            'upload_date': '20111115',
+            'thumbnail': r're:https?://.*\.jpg',
+            'view_count': int,
             'timestamp': 1321359578,
             'timestamp': 1321359578,
-            'thumbnail': 're:https?://.*\.jpg',
+            'upload_date': '20111115',
+        },
+    }, {
+        'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
+        'info_dict': {
+            'id': 'gillespeterson_caribou-7-inch-vinyl-mix-chat',
+            'ext': 'mp3',
+            'title': 'Caribou 7 inch Vinyl Mix & Chat',
+            'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
+            'uploader': 'Gilles Peterson Worldwide',
+            'uploader_id': 'gillespeterson',
+            'thumbnail': 're:https?://.*',
             'view_count': int,
             'view_count': int,
-            'like_count': int,
+            'timestamp': 1422987057,
+            'upload_date': '20150203',
         },
         },
-    }
+    }, {
+        'url': 'https://beta.mixcloud.com/RedLightRadio/nosedrip-15-red-light-radio-01-18-2016/',
+        'only_matching': True,
+    }]
+    _DECRYPTION_KEY = 'IFYOUWANTTHEARTISTSTOGETPAIDDONOTDOWNLOADFROMMIXCLOUD'
 
 
-    def _get_url(self, track_id, template_url):
-        server_count = 30
-        for i in range(server_count):
-            url = template_url % i
-            try:
-                # We only want to know if the request succeed
-                # don't download the whole file
-                self._request_webpage(
-                    HEADRequest(url), track_id,
-                    'Checking URL %d/%d ...' % (i + 1, server_count + 1))
-                return url
-            except ExtractorError:
-                pass
-
-        return None
+    @staticmethod
+    def _decrypt_xor_cipher(key, ciphertext):
+        """Encrypt/Decrypt XOR cipher. Both ways are possible because it's XOR."""
+        return ''.join([
+            compat_chr(compat_ord(ch) ^ compat_ord(k))
+            for ch, k in compat_zip(ciphertext, itertools.cycle(key))])
 
     def _real_extract(self, url):
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        uploader = mobj.group(1)
-        cloudcast_name = mobj.group(2)
-        track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name)))
-
-        webpage = self._download_webpage(url, track_id)
-
-        preview_url = self._search_regex(
-            r'\s(?:data-preview-url|m-preview)="(.+?)"', webpage, 'preview url')
-        song_url = preview_url.replace('/previews/', '/c/originals/')
-        template_url = re.sub(r'(stream\d*)', 'stream%d', song_url)
-        final_song_url = self._get_url(track_id, template_url)
-        if final_song_url is None:
-            self.to_screen('Trying with m4a extension')
-            template_url = template_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
-            final_song_url = self._get_url(track_id, template_url)
-        if final_song_url is None:
-            raise ExtractorError('Unable to extract track url')
-
-        PREFIX = (
-            r'<span class="play-button[^"]*?"'
-            r'(?:\s+[a-zA-Z0-9-]+(?:="[^"]+")?)*?\s+')
-        title = self._html_search_regex(
-            PREFIX + r'm-title="([^"]+)"', webpage, 'title')
-        thumbnail = self._proto_relative_url(self._html_search_regex(
-            PREFIX + r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail',
-            fatal=False))
-        uploader = self._html_search_regex(
-            PREFIX + r'm-owner-name="([^"]+)"',
-            webpage, 'uploader', fatal=False)
-        uploader_id = self._search_regex(
-            r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
-        description = self._og_search_description(webpage)
-        like_count = int_or_none(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
-            webpage, 'like count', fatal=False))
-        view_count = int_or_none(self._search_regex(
-            r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
-            webpage, 'play count', fatal=False))
-        timestamp = parse_iso8601(self._search_regex(
-            r'<time itemprop="dateCreated" datetime="([^"]+)">',
-            webpage, 'upload date'))
+        username, slug = re.match(self._VALID_URL, url).groups()
+        username, slug = compat_urllib_parse_unquote(username), compat_urllib_parse_unquote(slug)
+        track_id = '%s_%s' % (username, slug)
+
+        cloudcast = self._call_api('cloudcast', '''audioLength
+    comments(first: 100) {
+      edges {
+        node {
+          comment
+          created
+          user {
+            displayName
+            username
+          }
+        }
+      }
+      totalCount
+    }
+    description
+    favorites {
+      totalCount
+    }
+    featuringArtistList
+    isExclusive
+    name
+    owner {
+      displayName
+      url
+      username
+    }
+    picture(width: 1024, height: 1024) {
+        url
+    }
+    plays
+    publishDate
+    reposts {
+      totalCount
+    }
+    streamInfo {
+      dashUrl
+      hlsUrl
+      url
+    }
+    tags {
+      tag {
+        name
+      }
+    }''', track_id, username, slug)
+
+        title = cloudcast['name']
+
+        stream_info = cloudcast['streamInfo']
+        formats = []
+
+        for url_key in ('url', 'hlsUrl', 'dashUrl'):
+            format_url = stream_info.get(url_key)
+            if not format_url:
+                continue
+            decrypted = self._decrypt_xor_cipher(
+                self._DECRYPTION_KEY, compat_b64decode(format_url))
+            if url_key == 'hlsUrl':
+                formats.extend(self._extract_m3u8_formats(
+                    decrypted, track_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif url_key == 'dashUrl':
+                formats.extend(self._extract_mpd_formats(
+                    decrypted, track_id, mpd_id='dash', fatal=False))
+            else:
+                formats.append({
+                    'format_id': 'http',
+                    'url': decrypted,
+                    'downloader_options': {
+                        # Mixcloud starts throttling at >~5M
+                        'http_chunk_size': 5242880,
+                    },
+                })
+
+        if not formats and cloudcast.get('isExclusive'):
+            self.raise_login_required()
+
+        self._sort_formats(formats)
+
+        comments = []
+        for edge in (try_get(cloudcast, lambda x: x['comments']['edges']) or []):
+            node = edge.get('node') or {}
+            text = strip_or_none(node.get('comment'))
+            if not text:
+                continue
+            user = node.get('user') or {}
+            comments.append({
+                'author': user.get('displayName'),
+                'author_id': user.get('username'),
+                'text': text,
+                'timestamp': parse_iso8601(node.get('created')),
+            })
+
+        tags = []
+        for t in cloudcast.get('tags'):
+            tag = try_get(t, lambda x: x['tag']['name'], compat_str)
+            if not tag:
+                tags.append(tag)
+
+        get_count = lambda x: int_or_none(try_get(cloudcast, lambda y: y[x]['totalCount']))
+
+        owner = cloudcast.get('owner') or {}
 
         return {
             'id': track_id,
             'title': title,
 
         return {
             'id': track_id,
             'title': title,
-            'url': final_song_url,
-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'uploader_id': uploader_id,
-            'timestamp': timestamp,
-            'view_count': view_count,
-            'like_count': like_count,
+            'formats': formats,
+            'description': cloudcast.get('description'),
+            'thumbnail': try_get(cloudcast, lambda x: x['picture']['url'], compat_str),
+            'uploader': owner.get('displayName'),
+            'timestamp': parse_iso8601(cloudcast.get('publishDate')),
+            'uploader_id': owner.get('username'),
+            'uploader_url': owner.get('url'),
+            'duration': int_or_none(cloudcast.get('audioLength')),
+            'view_count': int_or_none(cloudcast.get('plays')),
+            'like_count': get_count('favorites'),
+            'repost_count': get_count('reposts'),
+            'comment_count': get_count('comments'),
+            'comments': comments,
+            'tags': tags,
+            'artist': ', '.join(cloudcast.get('featuringArtistList') or []) or None,
         }
         }
+
+
+class MixcloudPlaylistBaseIE(MixcloudBaseIE):
+    def _get_cloudcast(self, node):
+        return node
+
+    def _get_playlist_title(self, title, slug):
+        return title
+
+    def _real_extract(self, url):
+        username, slug = re.match(self._VALID_URL, url).groups()
+        username = compat_urllib_parse_unquote(username)
+        if not slug:
+            slug = 'uploads'
+        else:
+            slug = compat_urllib_parse_unquote(slug)
+        playlist_id = '%s_%s' % (username, slug)
+
+        is_playlist_type = self._ROOT_TYPE == 'playlist'
+        playlist_type = 'items' if is_playlist_type else slug
+        list_filter = ''
+
+        has_next_page = True
+        entries = []
+        while has_next_page:
+            playlist = self._call_api(
+                self._ROOT_TYPE, '''%s
+    %s
+    %s(first: 100%s) {
+      edges {
+        node {
+          %s
+        }
+      }
+      pageInfo {
+        endCursor
+        hasNextPage
+      }
+    }''' % (self._TITLE_KEY, self._DESCRIPTION_KEY, playlist_type, list_filter, self._NODE_TEMPLATE),
+                playlist_id, username, slug if is_playlist_type else None)
+
+            items = playlist.get(playlist_type) or {}
+            for edge in items.get('edges', []):
+                cloudcast = self._get_cloudcast(edge.get('node') or {})
+                cloudcast_url = cloudcast.get('url')
+                if not cloudcast_url:
+                    continue
+                entries.append(self.url_result(
+                    cloudcast_url, MixcloudIE.ie_key(), cloudcast.get('slug')))
+
+            page_info = items['pageInfo']
+            has_next_page = page_info['hasNextPage']
+            list_filter = ', after: "%s"' % page_info['endCursor']
+
+        return self.playlist_result(
+            entries, playlist_id,
+            self._get_playlist_title(playlist[self._TITLE_KEY], slug),
+            playlist.get(self._DESCRIPTION_KEY))
+
+
+class MixcloudUserIE(MixcloudPlaylistBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<id>[^/]+)/(?P<type>uploads|favorites|listens|stream)?/?$'
+    IE_NAME = 'mixcloud:user'
+
+    _TESTS = [{
+        'url': 'http://www.mixcloud.com/dholbach/',
+        'info_dict': {
+            'id': 'dholbach_uploads',
+            'title': 'Daniel Holbach (uploads)',
+            'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
+        },
+        'playlist_mincount': 36,
+    }, {
+        'url': 'http://www.mixcloud.com/dholbach/uploads/',
+        'info_dict': {
+            'id': 'dholbach_uploads',
+            'title': 'Daniel Holbach (uploads)',
+            'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
+        },
+        'playlist_mincount': 36,
+    }, {
+        'url': 'http://www.mixcloud.com/dholbach/favorites/',
+        'info_dict': {
+            'id': 'dholbach_favorites',
+            'title': 'Daniel Holbach (favorites)',
+            'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
+        },
+        # 'params': {
+        #     'playlist_items': '1-100',
+        # },
+        'playlist_mincount': 396,
+    }, {
+        'url': 'http://www.mixcloud.com/dholbach/listens/',
+        'info_dict': {
+            'id': 'dholbach_listens',
+            'title': 'Daniel Holbach (listens)',
+            'description': 'md5:b60d776f0bab534c5dabe0a34e47a789',
+        },
+        # 'params': {
+        #     'playlist_items': '1-100',
+        # },
+        'playlist_mincount': 1623,
+        'skip': 'Large list',
+    }, {
+        'url': 'https://www.mixcloud.com/FirstEar/stream/',
+        'info_dict': {
+            'id': 'FirstEar_stream',
+            'title': 'First Ear (stream)',
+            'description': 'Curators of good music\r\n\r\nfirstearmusic.com',
+        },
+        'playlist_mincount': 271,
+    }]
+
+    _TITLE_KEY = 'displayName'
+    _DESCRIPTION_KEY = 'biog'
+    _ROOT_TYPE = 'user'
+    _NODE_TEMPLATE = '''slug
+          url'''
+
+    def _get_playlist_title(self, title, slug):
+        return '%s (%s)' % (title, slug)
+
+
+class MixcloudPlaylistIE(MixcloudPlaylistBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
+    IE_NAME = 'mixcloud:playlist'
+
+    _TESTS = [{
+        'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
+        'info_dict': {
+            'id': 'maxvibes_jazzcat-on-ness-radio',
+            'title': 'Ness Radio sessions',
+        },
+        'playlist_mincount': 59,
+    }]
+    _TITLE_KEY = 'name'
+    _DESCRIPTION_KEY = 'description'
+    _ROOT_TYPE = 'playlist'
+    _NODE_TEMPLATE = '''cloudcast {
+            slug
+            url
+          }'''
+
+    def _get_cloudcast(self, node):
+        return node.get('cloudcast') or {}