From 33b8c411bc240fb7860ff2220344e9cbd4989933 Mon Sep 17 00:00:00 2001
From: Ha Tien Loi <loiht.b17vt220@stu.ptit.edu.vn>
Date: Thu, 10 Mar 2022 04:54:26 +0700
Subject: [PATCH] [MangoTV] Improve extractor (#2971)

Authored by: hatienl0i261299
---
 yt_dlp/extractor/mgtv.py | 59 ++++++++++++++++++++++++++++++++++++++--
 1 file changed, 56 insertions(+), 3 deletions(-)

diff --git a/yt_dlp/extractor/mgtv.py b/yt_dlp/extractor/mgtv.py
index cab3aa045..4ac70ea57 100644
--- a/yt_dlp/extractor/mgtv.py
+++ b/yt_dlp/extractor/mgtv.py
@@ -13,12 +13,15 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
+    try_get,
+    url_or_none,
 )
 
 
 class MGTVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:w(?:ww)?\.)?mgtv\.com/(v|b)/(?:[^/]+/)*(?P<id>\d+)\.html'
     IE_DESC = '芒果TV'
+    IE_NAME = 'MangoTV'
 
     _TESTS = [{
         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
@@ -30,6 +33,32 @@ class MGTVIE(InfoExtractor):
             'duration': 7461,
             'thumbnail': r're:^https?://.*\.jpg$',
         },
+    }, {
+        'url': 'https://w.mgtv.com/b/427837/15588271.html',
+        'info_dict': {
+            'id': '15588271',
+            'ext': 'mp4',
+            'title': '春日迟迟再出发 沉浸版',
+            'description': 'md5:a7a05a05b1aa87bd50cae619b19bbca6',
+            'thumbnail': r're:^https?://.+\.jpg',
+            'duration': 4026,
+        },
+    }, {
+        'url': 'https://w.mgtv.com/b/333652/7329822.html',
+        'info_dict': {
+            'id': '7329822',
+            'ext': 'mp4',
+            'title': '拜托，请你爱我',
+            'description': 'md5:cd81be6499bafe32e4d143abd822bf9c',
+            'thumbnail': r're:^https?://.+\.jpg',
+            'duration': 2656,
+        },
+    }, {
+        'url': 'https://w.mgtv.com/b/427837/15591647.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://w.mgtv.com/b/388252/15634192.html?fpa=33318&fpos=4&lastp=ch_home',
+        'only_matching': True,
     }, {
         'url': 'http://www.mgtv.com/b/301817/3826653.html',
         'only_matching': True,
@@ -40,12 +69,14 @@ class MGTVIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        tk2 = base64.urlsafe_b64encode(b'did=%s|pno=1030|ver=0.3.0301|clit=%d' % (compat_str(uuid.uuid4()).encode(), time.time()))[::-1]
+        tk2 = base64.urlsafe_b64encode(
+            f'did={compat_str(uuid.uuid4()).encode()}|pno=1030|ver=0.3.0301|clit={int(time.time())}'.encode())[::-1]
         try:
             api_data = self._download_json(
                 'https://pcweb.api.mgtv.com/player/video', video_id, query={
                     'tk2': tk2,
                     'video_id': video_id,
+                    'type': 'pch5'
                 }, headers=self.geo_verification_headers())['data']
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
@@ -61,6 +92,7 @@ class MGTVIE(InfoExtractor):
                 'pm2': api_data['atc']['pm2'],
                 'tk2': tk2,
                 'video_id': video_id,
+                'src': 'intelmgtv',
             }, headers=self.geo_verification_headers())['data']
         stream_domain = stream_data['stream_domain'][0]
 
@@ -71,7 +103,7 @@ class MGTVIE(InfoExtractor):
                 continue
             format_data = self._download_json(
                 stream_domain + stream_path, video_id,
-                note='Download video info for format #%d' % idx)
+                note=f'Download video info for format #{idx}')
             format_url = format_data.get('info')
             if not format_url:
                 continue
@@ -79,7 +111,7 @@ class MGTVIE(InfoExtractor):
                 r'_(\d+)_mp4/', format_url, 'tbr', default=None))
             formats.append({
                 'format_id': compat_str(tbr or idx),
-                'url': format_url,
+                'url': url_or_none(format_url),
                 'ext': 'mp4',
                 'tbr': tbr,
                 'protocol': 'm3u8_native',
@@ -97,4 +129,25 @@ class MGTVIE(InfoExtractor):
             'description': info.get('desc'),
             'duration': int_or_none(info.get('duration')),
             'thumbnail': info.get('thumb'),
+            'subtitles': self.extract_subtitles(video_id, stream_domain),
         }
+
+    def _get_subtitles(self, video_id, domain):
+        info = self._download_json(f'https://pcweb.api.mgtv.com/video/title?videoId={video_id}',
+                                   video_id, fatal=False) or {}
+        subtitles = {}
+        for sub in try_get(info, lambda x: x['data']['title']) or []:
+            url_sub = sub.get('url')
+            if not url_sub:
+                continue
+            locale = sub.get('captionCountrySimpleName')
+            sub = self._download_json(f'{domain}{url_sub}', video_id, fatal=False,
+                                      note=f'Download subtitle for locale {sub.get("name")} ({locale})') or {}
+            sub_url = url_or_none(sub.get('info'))
+            if not sub_url:
+                continue
+            subtitles.setdefault(locale or 'en', []).append({
+                'url': sub_url,
+                'ext': 'srt'
+            })
+        return subtitles