+
+ max_resolution = int_or_none(self._search_regex(
+ r'max_resolution["\']\s*:\s*(\d+)', webpage, 'max resolution',
+ default=None))
+ if max_resolution:
+ 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)
+ self._sort_formats(formats)
+
+ uploader = self._html_search_regex(
+ (r'(?s)<h4[^>]*>(.+?)</h4>.*?<em>\s*Author\s*</em>',
+ 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))
+ 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))
+ if len(formats) == 1:
+ formats[0]['filesize_approx'] = filesize_approx
+
+ if '<dd>Song' in webpage:
+ formats[0]['vcodec'] = 'none'
+
+ return {
+ 'id': media_id,
+ 'title': title,
+ 'uploader': uploader,
+ 'timestamp': timestamp,
+ 'duration': duration,
+ 'formats': formats,
+ }
+
+
+class NewgroundsPlaylistIE(InfoExtractor):
+ _VALID_URL = r'https?://(?:www\.)?newgrounds\.com/(?:collection|[^/]+/search/[^/]+)/(?P<id>[^/?#&]+)'
+ _TESTS = [{
+ 'url': 'https://www.newgrounds.com/collection/cats',
+ 'info_dict': {
+ 'id': 'cats',
+ 'title': 'Cats',
+ },
+ 'playlist_mincount': 46,
+ }, {
+ 'url': 'http://www.newgrounds.com/portal/search/author/ZONE-SAMA',
+ 'info_dict': {
+ 'id': 'ZONE-SAMA',
+ 'title': 'Portal Search: ZONE-SAMA',
+ },
+ 'playlist_mincount': 47,
+ }, {
+ 'url': 'http://www.newgrounds.com/audio/search/title/cats',
+ 'only_matching': True,
+ }]
+
+ def _real_extract(self, url):
+ playlist_id = self._match_id(url)
+
+ webpage = self._download_webpage(url, playlist_id)
+
+ title = self._search_regex(
+ r'<title>([^>]+)</title>', webpage, 'title', default=None)
+
+ # cut left menu
+ webpage = self._search_regex(
+ r'(?s)<div[^>]+\bclass=["\']column wide(.+)',
+ webpage, 'wide column', default=webpage)
+
+ entries = []
+ for a, path, media_id in re.findall(
+ r'(<a[^>]+\bhref=["\']/?((?:portal/view|audio/listen)/(\d+))[^>]+>)',
+ 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,
+ ie=NewgroundsIE.ie_key(), video_id=media_id))
+
+ return self.playlist_result(entries, playlist_id, title)