[mailru] fixed extraction

improved valid_url regex to match more urls like
>https://my.mail.ru/mail/7bw7/video/embed/_myvideo/6778
>https://videoapi.my.mail.ru/videos/embed/mail/7bw7/_myvideo/6778.html
previously not matched.

Fixed json extraction to download metadata and video url.
This commit is contained in:
nixxo 2020-09-07 14:21:21 +02:00 committed by GitHub
parent d51e23d9fc
commit d53fec697e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -20,10 +20,10 @@ class MailRuIE(InfoExtractor):
IE_DESC = 'Видео@Mail.Ru' IE_DESC = 'Видео@Mail.Ru'
_VALID_URL = r'''(?x) _VALID_URL = r'''(?x)
https?:// https?://
(?:(?:www|m)\.)?my\.mail\.ru/+ (?:(?:www|m|videoapi)\.)?my\.mail\.ru/+
(?: (?:
video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)| video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|
(?:(?P<idv2prefix>(?:[^/]+/+){2})video/(?P<idv2suffix>[^/]+/\d+))\.html| (?:videos/embed/)?(?:(?P<idv2prefix>(?:[^/]+/+){2})(?:video/(?:embed/)?)?(?P<idv2suffix>[^/]+/\d+))(?:\.html)?|
(?:video/embed|\+/video/meta)/(?P<metaid>\d+) (?:video/embed|\+/video/meta)/(?P<metaid>\d+)
) )
''' '''
@ -108,15 +108,21 @@ class MailRuIE(InfoExtractor):
if not video_id: if not video_id:
video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix') video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
page_config = self._parse_json(self._search_regex( page_config = self._parse_json(self._search_regex([
r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>', r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
r'(?s)"video":\s*(\{.+?\}),'],
webpage, 'page config', default='{}'), video_id, fatal=False) webpage, 'page config', default='{}'), video_id, fatal=False)
if page_config: if page_config:
meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') or page_config.get('metadataUrl')
else: else:
meta_url = None meta_url = None
video_data = None video_data = None
# fix meta_url if missing the host address
if re.match(r'^\/\+\/', meta_url):
meta_url = 'https://my.mail.ru' + meta_url
if meta_url: if meta_url:
video_data = self._download_json( video_data = self._download_json(
meta_url, video_id or meta_id, 'Downloading video meta JSON', meta_url, video_id or meta_id, 'Downloading video meta JSON',