bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#63469: 29.0.90; project.el doesn't add menu-bar entries


From: Dmitry Gutov
Subject: bug#63469: 29.0.90; project.el doesn't add menu-bar entries
Date: Fri, 26 May 2023 01:19:25 +0300
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.11.0

On 25/05/2023 21:41, Juri Linkov wrote:
-(cl-defmethod project-name ((_project (head vc)))
-  (or project-vc-name
+(cl-defmethod project-name ((project (head vc)))
+  (or (project--value-in-dir 'project-vc-name (project-root project))
         (cl-call-next-method)))
Thanks, I confirm this uses project-vc-name from dir-locals in non-file
buffers.
If you had a chance to verify that the performance is good enough (for
using the result in the mode-line), that would be great.

Either way...

;;;###autoload(put 'compilation-directory 'safe-local-variable 'stringp)
Huh, so the attribute :safe only makes proper sense for preloaded packages
and autoloaded defcustom forms (which are discouraged anyway).

Do you want to write a patch for all those vars in project.el? It will
probably make sense to push it to emacs-29.
With this patch the problem is solved.  I can push it to emacs-29.
...this patch LGTM too. Please go ahead and install it, or even both
patches.
I pushed ###autoload to emacs-29.

Thanks!

But not sure about your patch
since currently I have no idea how to verify its performance.
Maybe it should be pushed to master so that everyone can test it?

I think the performance requirements for 'project-name' are currently rather modest: the name is only used when selecting the project (in a custom config) and when generating some buffer names. For that, the new logic should be fine.

With your feature, however, project-name could be called during redisplay, and that might require better performance. So I thought you would test it together with the patch for your feature request.

But I suppose if you're not working on it right this moment, I could push the patch to emacs-29, and then we could refine the performance (if necessary) on master.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]