[mailru:music] Add extractor (closes #15618)

youtube_dl/extractor/mailru.py

@@ -1,12 +1,17 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import itertools
+import json
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     int_or_none,
+    parse_duration,
     remove_end,
+    try_get,
 )
 
 
@@ -157,3 +162,153 @@ class MailRuIE(InfoExtractor):
             'view_count': view_count,
             'formats': formats,
         }
+
+
+class MailRuMusicSearchBaseIE(InfoExtractor):
+    def _search(self, query, url, audio_id, limit=100, offset=0):
+        search = self._download_json(
+            'https://my.mail.ru/cgi-bin/my/ajax', audio_id,
+            'Downloading songs JSON page %d' % (offset // limit + 1),
+            headers={
+                'Referer': url,
+                'X-Requested-With': 'XMLHttpRequest',
+            }, query={
+                'xemail': '',
+                'ajax_call': '1',
+                'func_name': 'music.search',
+                'mna': '',
+                'mnb': '',
+                'arg_query': query,
+                'arg_extended': '1',
+                'arg_search_params': json.dumps({
+                    'music': {
+                        'limit': limit,
+                        'offset': offset,
+                    },
+                }),
+                'arg_limit': limit,
+                'arg_offset': offset,
+            })
+        return next(e for e in search if isinstance(e, dict))
+
+    @staticmethod
+    def _extract_track(t, fatal=True):
+        audio_url = t['URL'] if fatal else t.get('URL')
+        if not audio_url:
+            return
+
+        audio_id = t['File'] if fatal else t.get('File')
+        if not audio_id:
+            return
+
+        thumbnail = t.get('AlbumCoverURL') or t.get('FiledAlbumCover')
+        uploader = t.get('OwnerName') or t.get('OwnerName_Text_HTML')
+        uploader_id = t.get('UploaderID')
+        duration = int_or_none(t.get('DurationInSeconds')) or parse_duration(
+            t.get('Duration') or t.get('DurationStr'))
+        view_count = int_or_none(t.get('PlayCount') or t.get('PlayCount_hr'))
+
+        track = t.get('Name') or t.get('Name_Text_HTML')
+        artist = t.get('Author') or t.get('Author_Text_HTML')
+
+        if track:
+            title = '%s - %s' % (artist, track) if artist else track
+        else:
+            title = audio_id
+
+        return {
+            'extractor_key': MailRuMusicIE.ie_key(),
+            'id': audio_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'duration': duration,
+            'view_count': view_count,
+            'vcodec': 'none',
+            'abr': int_or_none(t.get('BitRate')),
+            'track': track,
+            'artist': artist,
+            'album': t.get('Album'),
+            'url': audio_url,
+        }
+
+
+class MailRuMusicIE(MailRuMusicSearchBaseIE):
+    IE_NAME = 'mailru:music'
+    IE_DESC = 'Музыка@Mail.Ru'
+    _VALID_URL = r'https?://my\.mail\.ru/music/songs/[^/?#&]+-(?P<id>[\da-f]+)'
+    _TESTS = [{
+        'url': 'https://my.mail.ru/music/songs/%D0%BC8%D0%BB8%D1%82%D1%85-l-a-h-luciferian-aesthetics-of-herrschaft-single-2017-4e31f7125d0dfaef505d947642366893',
+        'md5': '0f8c22ef8c5d665b13ac709e63025610',
+        'info_dict': {
+            'id': '4e31f7125d0dfaef505d947642366893',
+            'ext': 'mp3',
+            'title': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017 - М8Л8ТХ',
+            'uploader': 'Игорь Мудрый',
+            'uploader_id': '1459196328',
+            'duration': 280,
+            'view_count': int,
+            'vcodec': 'none',
+            'abr': 320,
+            'track': 'L.A.H. (Luciferian Aesthetics of Herrschaft) single, 2017',
+            'artist': 'М8Л8ТХ',
+        },
+    }]
+
+    def _real_extract(self, url):
+        audio_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, audio_id)
+
+        title = self._og_search_title(webpage)
+        music_data = self._search(title, url, audio_id)['MusicData']
+        t = next(t for t in music_data if t.get('File') == audio_id)
+
+        info = self._extract_track(t)
+        info['title'] = title
+        return info
+
+
+class MailRuMusicSearchIE(MailRuMusicSearchBaseIE):
+    IE_NAME = 'mailru:music:search'
+    IE_DESC = 'Музыка@Mail.Ru'
+    _VALID_URL = r'https?://my\.mail\.ru/music/search/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://my.mail.ru/music/search/black%20shadow',
+        'info_dict': {
+            'id': 'black shadow',
+        },
+        'playlist_mincount': 532,
+    }]
+
+    def _real_extract(self, url):
+        query = compat_urllib_parse_unquote(self._match_id(url))
+
+        entries = []
+
+        LIMIT = 100
+        offset = 0
+
+        for _ in itertools.count(1):
+            search = self._search(query, url, query, LIMIT, offset)
+
+            music_data = search.get('MusicData')
+            if not music_data or not isinstance(music_data, list):
+                break
+
+            for t in music_data:
+                track = self._extract_track(t, fatal=False)
+                if track:
+                    entries.append(track)
+
+            total = try_get(
+                search, lambda x: x['Results']['music']['Total'], int)
+
+            if total is not None:
+                if offset > total:
+                    break
+
+            offset += LIMIT
+
+        return self.playlist_result(entries, query)