dirkf
commented
2 years ago I believe that must be so.
This patch gets a video on the free pages but I have no idea about the new subscription pages:
--- old/youtube-dl/youtube_dl/extractor/raywenderlich.py
+++ new/youtube-dl/youtube_dl/extractor/raywenderlich.py
@@ -7,11 +7,17 @@
from ..compat import compat_str
from ..utils import (
ExtractorError,
+ clean_html,
+ get_element_by_attribute,
+ get_element_by_class,
int_or_none,
merge_dicts,
+ parse_duration,
+ str_or_none,
try_get,
unescapeHTML,
unified_timestamp,
+ unified_strdate,
urljoin,
)
@@ -82,7 +88,10 @@
webpage, default=None) or self._html_search_meta(
'twitter:image', webpage, 'thumbnail')
- if '>Subscribe to unlock' in webpage:
+ if self._search_regex(
+ ('>Subscribe to unlock',
+ r'<a\s[^>]*?\bdata-subscription-required\s*=\s*(?:\'|"|\b)true(?:\'|"|\b)'),
+ webpage, 'subscription needed', default=False, group=0):
raise ExtractorError(
'This content is only available for subscribers',
expected=True)
@@ -91,8 +100,25 @@
'thumbnail': thumbnail,
}
- vimeo_id = self._search_regex(
- r'data-vimeo-id=["\'](\d+)', webpage, 'vimeo id', default=None)
+ vimeo_url = self._og_search_property('video:url', webpage, fatal=False)
+ if vimeo_url:
+ info.update({
+ '_type': 'url_transparent',
+ 'title': self._og_search_title(webpage),
+ 'description': clean_html(self._og_search_description(webpage)),
+ 'duration': parse_duration(get_element_by_class('c-box-list__item-duration', webpage)),
+ 'upload_date': unified_strdate(
+ re.split(r'[^\w\s]',
+ get_element_by_class('c-written-tutorial__meta', webpage))[0].strip()),
+ 'creator': clean_html(str_or_none(get_element_by_class('c-content-author__title', webpage)))
+ })
+
+ if not vimeo_url:
+ vimeo_id = self._search_regex(
+ r'data-vimeo-id=["\'](\d+)', webpage, 'vimeo id', default=None)
+ else:
+ vimeo_id = self._search_regex(
+ r'https?://player\.vimeo\.com/external/(\d+)', vimeo_url, 'vimeo id', default=None)
if not vimeo_id:
data = self._parse_json(
Diodor7
Checklist
Verbose log
Description
It looks like the site has changed the output of the video id and now the extractor\raywenderlich.py does not work correctly. Can someone help?