Merge b1699d4752 into e1b3fa242c

- added thumbnail extractor
- added upload date
- added description
- added view count
- fix video extraction
- fix playlist extraction

youtube_dl/extractor/newgrounds.py

@@ -8,12 +8,14 @@ from ..utils import (
     int_or_none,
     parse_duration,
     parse_filesize,
+    parse_count,
+    parse_resolution,
     unified_timestamp,
 )
 
 
 class NewgroundsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:audio/listen|portal/view)/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:audio/listen|portal/view)/(?P<id>[0-9]+)(?:/format/flash)?'
     _TESTS = [{
         'url': 'https://www.newgrounds.com/audio/listen/549479',
         'md5': 'fe6033d297591288fa1c1f780386f07a',
@@ -25,17 +27,18 @@ class NewgroundsIE(InfoExtractor):
             'timestamp': 1378878540,
             'upload_date': '20130911',
             'duration': 143,
+            'description': 'Sonic blahblahblah I\'m late again SEPTEMBER 11TH: Busmode-The story of ants on a log. ',
         },
     }, {
-        'url': 'https://www.newgrounds.com/portal/view/673111',
+        'url': 'https://www.newgrounds.com/portal/view/297383',
-        'md5': '3394735822aab2478c31b1004fe5e5bc',
         'info_dict': {
-            'id': '673111',
+            'id': '297383',
             'ext': 'mp4',
-            'title': 'Dancin',
+            'title': 'Metal Gear Awesome',
-            'uploader': 'Squirrelman82',
+            'uploader': 'Egoraptor',
-            'timestamp': 1460256780,
+            'timestamp': 1140663240,
-            'upload_date': '20160410',
+            'upload_date': '20060223',
+            'description': 'Metal Gear is awesome is so is this movie.',
         },
     }, {
         # source format unavailable, additional mp4 formats
@@ -44,9 +47,10 @@ class NewgroundsIE(InfoExtractor):
             'id': '689400',
             'ext': 'mp4',
             'title': 'ZTV News Episode 8',
-            'uploader': 'BennettTheSage',
+            'uploader': 'ZONE-SAMA',
             'timestamp': 1487965140,
             'upload_date': '20170224',
+            'description': 'ZTV News Episode 8 (February 2017)',
         },
         'params': {
             'skip_download': True,
@@ -55,51 +59,65 @@ class NewgroundsIE(InfoExtractor):
 
     def _real_extract(self, url):
         media_id = self._match_id(url)
+        formats = []
+        uploader = None
 
         webpage = self._download_webpage(url, media_id)
 
         title = self._html_search_regex(
-            r'<title>([^>]+)</title>', webpage, 'title')
+            r'<title>(.+?)</title>', webpage, 'title')
 
-        media_url = self._parse_json(self._search_regex(
+        media_url_string = self._search_regex(r'"url"\s*:\s*("[^"]+"),', webpage, 'media url string', fatal=False, default=None)
-            r'"url"\s*:\s*("[^"]+"),', webpage, ''), media_id)
 
-        formats = [{
+        if media_url_string:
-            'url': media_url,
+            media_url = self._parse_json(media_url_string, media_id)
-            'format_id': 'source',
+            formats = [{
-            'quality': 1,
+                'url': media_url,
-        }]
+                'format_id': 'source',
+                'quality': 1,
+            }]
+        else:
+            json_data = self._download_json('https://www.newgrounds.com/portal/video/' + media_id, media_id, headers={
+                'Accept': 'application/json, text/javascript, */*; q=0.01',
+                'Accept-Encoding': 'gzip, deflate, br',
+                'X-Requested-With': 'XMLHttpRequest',
+            })
 
-        max_resolution = int_or_none(self._search_regex(
+            uploader = json_data.get('author')
-            r'max_resolution["\']\s*:\s*(\d+)', webpage, 'max resolution',
+            media_formats = json_data['sources']
-            default=None))
+            for media_format in media_formats:
-        if max_resolution:
+                media_sources = media_formats[media_format]
-            url_base = media_url.rpartition('.')[0]
+                for source in media_sources:
-            for resolution in (360, 720, 1080):
+                    format = {
-                if resolution > max_resolution:
+                        'format_id': media_format,
-                    break
+                        'url': source.get('src'),
-                formats.append({
+                    }
-                    'url': '%s.%dp.mp4' % (url_base, resolution),
+                    format.update(parse_resolution(media_format))
-                    'format_id': '%dp' % resolution,
+                    format['quality'] = format.get('height')
-                    'height': resolution,
+                    formats.append(format)
-                })
 
-        self._check_formats(formats, media_id)
+        if not uploader:
-        self._sort_formats(formats)
+            uploader = self._html_search_regex(
-
+                (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
-        uploader = self._html_search_regex(
+                 r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
-            (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
+                fatal=False)
-             r'(?:Author|Writer)\s*<a[^>]+>([^<]+)'), webpage, 'uploader',
-            fatal=False)
 
         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',
             default=None))
+
+        thumbnail = self._og_search_thumbnail(webpage)
+
         duration = parse_duration(self._search_regex(
             r'(?s)<dd>\s*Song\s*</dd>\s*<dd>.+?</dd>\s*<dd>([^<]+)', webpage,
             'duration', default=None))
 
+        description = self._og_search_description(webpage)
+
+        view_count = parse_count(self._html_search_regex(r'(?s)<dt>\s*Views\s*</dt>\s*<dd>([\d\.,]+)</dd>', webpage,
+                                                         'view_count', fatal=False, 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 +127,9 @@ 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,
@@ -116,6 +137,9 @@ class NewgroundsIE(InfoExtractor):
             'timestamp': timestamp,
             'duration': duration,
             'formats': formats,
+            'description': description,
+            'thumbnail': thumbnail,
+            'view_count': view_count,
         }
 
 
@@ -127,14 +151,14 @@ class NewgroundsPlaylistIE(InfoExtractor):
             'id': 'cats',
             'title': 'Cats',
         },
-        'playlist_mincount': 46,
+        'playlist_mincount': 45,
     }, {
-        'url': 'http://www.newgrounds.com/portal/search/author/ZONE-SAMA',
+        'url': 'https://www.newgrounds.com/collection/dogs',
         'info_dict': {
-            'id': 'ZONE-SAMA',
+            'id': 'dogs',
-            'title': 'Portal Search: ZONE-SAMA',
+            'title': 'Dogs',
         },
-        'playlist_mincount': 47,
+        'playlist_mincount': 25,
     }, {
         'url': 'http://www.newgrounds.com/audio/search/title/cats',
         'only_matching': True,
@@ -155,14 +179,14 @@ class NewgroundsPlaylistIE(InfoExtractor):
 
         entries = []
         for a, path, media_id in re.findall(
-                r'(<a[^>]+\bhref=["\']/?((?:portal/view|audio/listen)/(\d+))[^>]+>)',
+                r'(<a[^>]+href="https?://[^/]+/(audio/listen|portal/view)/([0-9]+)"[^>]+>)',
                 webpage):
             a_class = extract_attributes(a).get('class')
             if a_class not in ('item-portalsubmission', 'item-audiosubmission'):
                 continue
             entries.append(
                 self.url_result(
-                    'https://www.newgrounds.com/%s' % path,
+                    'https://www.newgrounds.com/%s/%s' % (path, media_id),
                     ie=NewgroundsIE.ie_key(), video_id=media_id))
 
         return self.playlist_result(entries, playlist_id, title)