[zingmp3] Fix extractor (#2889)

Authored by: hatienl0i261299

yt_dlp/extractor/zingmp3.py

@@ -1,22 +1,47 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import hashlib
+import hmac
+import urllib.parse
+
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
+    traverse_obj,
+    HEADRequest,
 )
 
 
 class ZingMp3BaseIE(InfoExtractor):
-    _VALID_URL_TMPL = r'https?://(?:mp3\.zing|zingmp3)\.vn/(?:%s)/[^/]+/(?P<id>\w+)\.html'
+    _VALID_URL_TMPL = r'https?://(?:mp3\.zing|zingmp3)\.vn/(?P<type>(?:%s))/[^/]+/(?P<id>\w+)(?:\.html|\?)'
     _GEO_COUNTRIES = ['VN']
+    _DOMAIN = 'https://zingmp3.vn'
+    _SLUG_API = {
+        'bai-hat': '/api/v2/page/get/song',
+        'embed': '/api/v2/page/get/song',
+        'video-clip': '/api/v2/page/get/video',
+        'playlist': '/api/v2/page/get/playlist',
+        'album': '/api/v2/page/get/playlist',
+        'lyric': '/api/v2/lyric/get/lyric',
+        'song_streaming': '/api/v2/song/get/streaming',
+    }
 
-    def _extract_item(self, item, fatal):
+    _API_KEY = '88265e23d4284f25963e6eedac8fbfa3'
-        item_id = item['id']
+    _SECRET_KEY = b'2aa2d1c561e809b267f3638c4a307aab'
-        title = item.get('name') or item['title']
+
+    def _extract_item(self, item, song_id, type_url, fatal):
+        item_id = item.get('encodeId') or song_id
+        title = item.get('title') or item.get('alias')
+
+        if type_url == 'video-clip':
+            source = item.get('streaming')
+        else:
+            api = self.get_api_with_signature(name_api=self._SLUG_API.get('song_streaming'), param={'id': item_id})
+            source = self._download_json(api, video_id=item_id).get('data')
 
         formats = []
-        for k, v in (item.get('source') or {}).items():
+        for k, v in (source or {}).items():
             if not v:
                 continue
             if k in ('mp4', 'hls'):
@@ -34,7 +59,9 @@ def _extract_item(self, item, fatal):
                             'height': int_or_none(self._search_regex(
                                 r'^(\d+)p', res, 'resolution', default=None)),
                         })
-            else:
+                continue
+            elif v == 'VIP':
+                continue
             formats.append({
                 'ext': 'mp3',
                 'format_id': k,
@@ -45,20 +72,22 @@ def _extract_item(self, item, fatal):
         if not formats:
             if not fatal:
                 return
-            msg = item['msg']
+            msg = item.get('msg')
             if msg == 'Sorry, this content is not available in your country.':
                 self.raise_geo_restricted(countries=self._GEO_COUNTRIES, metadata_available=True)
             self.raise_no_formats(msg, expected=True)
         self._sort_formats(formats)
 
-        subtitles = None
         lyric = item.get('lyric')
-        if lyric:
+        if not lyric:
+            api = self.get_api_with_signature(name_api=self._SLUG_API.get("lyric"), param={'id': item_id})
+            info_lyric = self._download_json(api, video_id=item_id)
+            lyric = traverse_obj(info_lyric, ('data', 'file'))
         subtitles = {
             'origin': [{
                 'url': lyric,
             }],
-            }
+        } if lyric else None
 
         album = item.get('album') or {}
 
@@ -66,30 +95,41 @@ def _extract_item(self, item, fatal):
             'id': item_id,
             'title': title,
             'formats': formats,
-            'thumbnail': item.get('thumbnail'),
+            'thumbnail': traverse_obj(item, 'thumbnail', 'thumbnailM'),
             'subtitles': subtitles,
             'duration': int_or_none(item.get('duration')),
             'track': title,
-            'artist': item.get('artists_names'),
+            'artist': traverse_obj(item, 'artistsNames', 'artists_names'),
-            'album': album.get('name') or album.get('title'),
+            'album': traverse_obj(album, 'name', 'title'),
-            'album_artist': album.get('artists_names'),
+            'album_artist': traverse_obj(album, 'artistsNames', 'artists_names'),
         }
 
+    def _real_initialize(self):
+        if not self.get_param('cookiefile') and not self.get_param('cookiesfrombrowser'):
+            self._request_webpage(HEADRequest(self._DOMAIN), None, note='Updating cookies')
+
     def _real_extract(self, url):
-        page_id = self._match_id(url)
+        song_id, type_url = self._match_valid_url(url).group('id', 'type')
-        webpage = self._download_webpage(
+
-            url.replace('://zingmp3.vn/', '://mp3.zing.vn/'),
+        api = self.get_api_with_signature(name_api=self._SLUG_API[type_url], param={'id': song_id})
-            page_id, query={'play_song': 1})
+
-        data_path = self._search_regex(
+        return self._process_data(self._download_json(api, song_id)['data'], song_id, type_url)
-            r'data-xml="([^"]+)', webpage, 'data path')
+
-        return self._process_data(self._download_json(
+    def get_api_with_signature(self, name_api, param):
-            'https://mp3.zing.vn/xhr' + data_path, page_id)['data'])
+        sha256 = hashlib.sha256(''.join(f'{k}={v}' for k, v in param.items()).encode('utf-8')).hexdigest()
+
+        data = {
+            'apiKey': self._API_KEY,
+            'sig': hmac.new(self._SECRET_KEY, f'{name_api}{sha256}'.encode('utf-8'), hashlib.sha512).hexdigest(),
+            **param,
+        }
+        return f'{self._DOMAIN}{name_api}?{urllib.parse.urlencode(data)}'
 
 
 class ZingMp3IE(ZingMp3BaseIE):
-    _VALID_URL = ZingMp3BaseIE._VALID_URL_TMPL % 'bai-hat|video-clip'
+    _VALID_URL = ZingMp3BaseIE._VALID_URL_TMPL % 'bai-hat|video-clip|embed'
     _TESTS = [{
-        'url': 'http://mp3.zing.vn/bai-hat/Xa-Mai-Xa-Bao-Thy/ZWZB9WAB.html',
+        'url': 'https://mp3.zing.vn/bai-hat/Xa-Mai-Xa-Bao-Thy/ZWZB9WAB.html',
         'md5': 'ead7ae13693b3205cbc89536a077daed',
         'info_dict': {
             'id': 'ZWZB9WAB',
@@ -108,7 +148,7 @@ class ZingMp3IE(ZingMp3BaseIE):
             'album_artist': 'Bảo Thy',
         },
     }, {
-        'url': 'https://mp3.zing.vn/video-clip/Suong-Hoa-Dua-Loi-K-ICM-RYO/ZO8ZF7C7.html',
+        'url': 'https://zingmp3.vn/video-clip/Suong-Hoa-Dua-Loi-K-ICM-RYO/ZO8ZF7C7.html',
         'md5': 'e9c972b693aa88301ef981c8151c4343',
         'info_dict': {
             'id': 'ZO8ZF7C7',
@@ -119,15 +159,18 @@ class ZingMp3IE(ZingMp3BaseIE):
             'track': 'Sương Hoa Đưa Lối',
             'artist': 'K-ICM, RYO',
         },
+    }, {
+        'url': 'https://zingmp3.vn/embed/song/ZWZEI76B?start=false',
+        'only_matching': True,
     }, {
         'url': 'https://zingmp3.vn/bai-hat/Xa-Mai-Xa-Bao-Thy/ZWZB9WAB.html',
         'only_matching': True,
     }]
     IE_NAME = 'zingmp3'
-    IE_DESC = 'mp3.zing.vn'
+    IE_DESC = 'zingmp3.vn'
 
-    def _process_data(self, data):
+    def _process_data(self, data, song_id, type_url):
-        return self._extract_item(data, True)
+        return self._extract_item(data, song_id, type_url, True)
 
 
 class ZingMp3AlbumIE(ZingMp3BaseIE):
@@ -139,7 +182,7 @@ class ZingMp3AlbumIE(ZingMp3BaseIE):
             'id': 'ZWZBWDAF',
             'title': 'Lâu Đài Tình Ái',
         },
-        'playlist_count': 10,
+        'playlist_count': 9,
     }, {
         'url': 'http://mp3.zing.vn/playlist/Duong-Hong-Loan-apollobee/IWCAACCB.html',
         'only_matching': True,
@@ -149,12 +192,12 @@ class ZingMp3AlbumIE(ZingMp3BaseIE):
     }]
     IE_NAME = 'zingmp3:album'
 
-    def _process_data(self, data):
+    def _process_data(self, data, song_id, type_url):
         def entries():
-            for item in (data.get('items') or []):
+            for item in traverse_obj(data, ('song', 'items')) or []:
-                entry = self._extract_item(item, False)
+                entry = self._extract_item(item, song_id, type_url, False)
                 if entry:
                     yield entry
-        info = data.get('info') or {}
+
-        return self.playlist_result(
+        return self.playlist_result(entries(), traverse_obj(data, 'id', 'encodeId'),
-            entries(), info.get('id'), info.get('name') or info.get('title'))
+                                    traverse_obj(data, 'name', 'title'))