Merge 3091daa834 into a96a45b2cd

youtube_dl/extractor/newgrounds.py

@@ -5,7 +5,6 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     extract_attributes,
-    int_or_none,
     parse_duration,
     parse_filesize,
     unified_timestamp,
@@ -27,15 +26,15 @@ class NewgroundsIE(InfoExtractor):
             'duration': 143,
         },
     }, {
-        'url': 'https://www.newgrounds.com/portal/view/673111',
+        'url': 'https://www.newgrounds.com/portal/view/850292',
-        'md5': '3394735822aab2478c31b1004fe5e5bc',
+        'md5': 'bb7cacf45e1b4d648e2dac2d79284d67',
         'info_dict': {
-            'id': '673111',
+            'id': '850292',
             'ext': 'mp4',
-            'title': 'Dancin',
+            'title': 'Timeless (2021)',
-            'uploader': 'Squirrelman82',
+            'uploader': 'Kevuhn',
-            'timestamp': 1460256780,
+            'timestamp': 1657896960,
-            'upload_date': '20160410',
+            'upload_date': '20220715',
         },
     }, {
         # source format unavailable, additional mp4 formats
@@ -44,7 +43,7 @@ class NewgroundsIE(InfoExtractor):
             'id': '689400',
             'ext': 'mp4',
             'title': 'ZTV News Episode 8',
-            'uploader': 'BennettTheSage',
+            'uploader': 'ZONE-SAMA',
             'timestamp': 1487965140,
             'upload_date': '20170224',
         },
@@ -58,40 +57,46 @@ class NewgroundsIE(InfoExtractor):
 
         webpage = self._download_webpage(url, media_id)
 
-        title = self._html_search_regex(
+        paths = url.split('/')
-            r'<title>([^>]+)</title>', webpage, 'title')
+        if paths[-3] == 'audio':
+            isAudio = True
+        else:
+            isAudio = False
 
-        media_url = self._parse_json(self._search_regex(
+        if isAudio:
-            r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
+            media_url = self._parse_json(self._search_regex(
+                r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
 
-        formats = [{
+            uploader = self._html_search_regex(
-            'url': media_url,
+                (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
-            'format_id': 'source',
+                 r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
-            'quality': 1,
+                fatal=False)
-        }]
 
-        max_resolution = int_or_none(self._search_regex(
+            formats = [{
-            r'max_resolution["\']\s*:\s*(\d+)', webpage, 'max resolution',
+                'url': media_url,
-            default=None))
+                'format_id': 'source',
-        if max_resolution:
+                'quality': 1,
-            url_base = media_url.rpartition('.')[0]
+            }]
-            for resolution in (360, 720, 1080):
-                if resolution > max_resolution:
-                    break
-                formats.append({
-                    'url': '%s.%dp.mp4' % (url_base, resolution),
-                    'format_id': '%dp' % resolution,
-                    'height': resolution,
-                })
 
-        self._check_formats(formats, media_id)
+        else:
-        self._sort_formats(formats)
+            media_url = 'https://www.newgrounds.com/portal/video/' + media_id
+            media = self._download_json(media_url, media_id, headers={'X-Requested-With': 'XMLHttpRequest'})
+
+            uploader = media['author']
+
+            sources = media['sources']
 
-        uploader = self._html_search_regex(
+            formats = []
-            (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
+            for source in sources:
-             r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
+                for i in range(len(sources[source])):
-            fatal=False)
+                    formats.append({
+                        'url': sources[source][i]['src'],
+                        'format_id': source,
+                        'height': int(source[:-2])  # 1080p -> 1080
+                    })
 
+        title = self._html_search_regex(
+            r'<title>([^>]+)</title>', webpage, 'title')
         timestamp = unified_timestamp(self._html_search_regex(
             (r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+</dd>\s*<dd>[^<]+)',
              r'<dt>\s*Uploaded\s*</dt>\s*<dd>([^<]+)'), webpage, 'timestamp',
@@ -99,7 +104,6 @@ class NewgroundsIE(InfoExtractor):
         duration = parse_duration(self._search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>.+?</dd>\s*<dd>([^<]+)', webpage,
             'duration', default=None))
-
         filesize_approx = parse_filesize(self._html_search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>(.+?)</dd>', webpage, 'filesize',
             default=None))
@@ -109,6 +113,9 @@ class NewgroundsIE(InfoExtractor):
         if '<dd>Song' in webpage:
             formats[0]['vcodec'] = 'none'
 
+        self._check_formats(formats, media_id)
+        self._sort_formats(formats)
+
         return {
             'id': media_id,
             'title': title,