Philip Woolford
Philip Woolford
@askor yeah, I should have thought of that... I think this is still OK though as it can bubble up JWT format/validity errors separate from JWT presence.
@kaplanelad Can you provide a little more guidance? There are no existing JWT tests in that folder, are you OK for me to just have at it? I have made...
@kaplanelad sorry for the ping, I don't know if you get one when I put an update. I fixed the style CI fail.
Hi @rawhuul, it would require a redesign of the high level and low level APIs to support the reporting flags that send `fanotify_event_info_fid` and `fanotify_event_info_pidfd`. Same with supporting error reporting...
There aren't currently an plans for API changes, though I am happy to work with you to add features to the crate
Actually, I had a look at the fanotify API again today and it seemed OK to implement as long as I hacked around the DST in the `fanotify_event_info_fid ` struct....
@rawhuul I don't know if you started looking, but I just pushed some more code to properly handle `FAN_EVENT_INFO_TYPE_DFID_NAME` returning a string with the name of the file being operated...
Yeah, unfortunately even the proposed patches aren't sufficient as the fid struct is added with the [u8;0] instead of properly modelling the DST behaviour.
If anyone sees this in the future and is wondering why I don't want to use DSTs, please have a look at the kind of stuff you need to do...
Yeah, no worries at all. I was debugging an issue where fanotify seems to break in weird ways when used on an nfsv4 mount with kerberos auth, and I figured...