You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Dec 28, 2017. It is now read-only.
Basically, content-type's of files uploaded through django FileFields get improperly escaped by the version of S3BotoStorage in athumb because unicode Content-type headers get sent through boto, which escapes them, which aws doesn't unescape. So you end up with a content-type metadata in aws looking something like application%2fpdf . Aside from being a bad content-type in and of itself, this causes Chrome's pdf reader to get hung up.
TBH, I'm not actually using athumb in my project at the moment, but was inadvertently using the S3BotoStorage import from it instead of the django-storages one (careless IDE auto-import), so I'm probably not a great one to test it out the branch with any degree of thoroughness, although i could certainly check to see if it solves this particular problem. I just switched the import to django-storages and things started working.
Basically, content-type's of files uploaded through django FileFields get improperly escaped by the version of S3BotoStorage in athumb because unicode Content-type headers get sent through boto, which escapes them, which aws doesn't unescape. So you end up with a content-type metadata in aws looking something like
application%2fpdf
. Aside from being a bad content-type in and of itself, this causes Chrome's pdf reader to get hung up.See boto/boto#1669
Current version of django-storages doesn't suffer from this issue, so I assume that implementing #21 would fix this.
The text was updated successfully, but these errors were encountered: