See CTFd/CTFd#2157 for details

Way too long here

This needs the location hash functionality that exists in regular core that lets users directly link to a specific challenge.

See vitejs/vite#2375 (comment)

We do not seem to be releasing the SSE connection with each tab. We need to release the connection if we have too many tabs because users will be unable to interact with CTFd if they have too many tabs open.

We should make sure that our behavior is the same as core when scores are hidden/admins only. I'm not sure if we're supposed to show the user's own score or not.

It looks like we are missing the paused color response in the challenge response alert.

This should probably be yellow.

Title is self explanatory, not sure what the cause is.

The scoreboard page of the current core theme loads a replacement text ("No solves yet") if there are no solves. This probably got lost or was deleted intentionally in this rewrite.

A simple solution for this would be to either implement something similar to the core theme or implement a check of the data in scoreboard.js that changes some variable and updates the page via Alpine.js structures in the template (for example x-if).

The second option seems cleaner to me since it separates the template from the logic part. I implemented this version locally and could push or paste it somewhere if it helps. However, there are many possible solutions for this. I simply wanted to bring this to your attention so that it is not forgotten.

Local site

My site at ddomdel.ctfd.io

There is no legend on the score graph, we should bring it back.

We should currently be using Vite 2 but we should update to Vite 3 if possible: https://vitejs.dev/blog/announcing-vite3.html#new-documentation.

• Is there a place where the selection of alpine over vue was discussed?

After running UIUCTF 2023, I felt that alpine was insufficient for our needs, and would have greatly preferred a Vue base. I am considering doing a full rewrite of core-beta in Vue, but would be happy to talk more about it with you first.

as implemented in async loadChallenge(challengeId), core-beta invokes displayChallenge in CTFd.js which downloads view.js for a challenge and does this:

https://github.com/CTFd/CTFd.js/blob/65df4e05fa039b6e464d50e62423e4536580f318/pages/challenge.js#L18-L33

before core-beta, postRender is invoked after the html is actually rendered, yet core-beta uses alpinejs, so show() the modal takes effect after the entire displayChallenge, which breaks the logic of a plugin which relied on postRender to get the challenge id with jquery($('#challenge-id')) and load the challenge info.

I think what I really need is CTFd._internal.challenge.data.id, yet I still wonder should postRender keep the previous behavior when virtual dom is actually rendered? otherwise what's the point of the postRender hook?

Show a notification count in the title to make it clearer when people should go read a notification.

This needs to be synced across all CTFd tabs so it's a little tricky.

Someone pointed out to me that the alignment of the Team header is off and could be better positioned.

Truncate filenames that end in ellipsis (...) should show the full file name when you hover over them or something similar.

I do not think we are incrementing the attempts counter after a submission. We should look at seeing what the best strategy is for this.

CTFd will frequently throw a FileNotFoundError while yarn/vite is building because the manifest file hasn't been generated.

This is something we should investigate on fixing.

Here is a comparison.

Mac

Linux

Windows