cancelling yequake-org-capture toggles buffer in other frame

[emacsomancer]

Cancelling an org-capture (with q) seems to switch the buffer in the non-org-capture frame.

[alphapapa]

Congratulations, sir, you have found the golden ticket!

Haha, seriously though... Yeah, it does, and I haven't been able to figure out why or how to fix it. I didn't think it was worth delaying release for.

org-capture does a lot of tricky stuff with windows and buffers and restoring state. I already had to override one function to get the correct behavior. It'll probably require more hackyness to fix this.

I don't know when or if I'll be able to fix this. Help would be appreciated. :)

[emacsomancer]

Of course this is sort of a minor annoyance, since presumably 90% of the time a user opens a yequake-org-capture frame, they will actually not be cancelling the capture. And it's also still preferable to how my org-capture frames have worked up until this point - which either didn't automatically close, or else also closed an additional open emacs frame.

Do you think it would be possible to somehow use something like a "cancel-org-capture-hook" to pre-emptively toggle the buffer in the other frame, so that whatever weirdness cancelling org-capture does just ends up restoring the original buffer to that frame? (Or – since such a hook presumably doesn't actually exist – maybe over-riding q to toggle the buffer in the other frame and then call whatever function is org-capture has bound to q?)

[alphapapa]

I suppose if one piled enough hacks upon hacks, it could work. But I'd rather figure out why it's happening in the first place. It might actually be a bug in Org itself. I think what we need to do is find the code that's causing it, then we can figure out how to fix it.

[emacsomancer]

Interestingly, cancelling with C-c C-k once one has started a note doesn't trigger this.

[claytharrison]

If I do e.g. (yequake-org-capture nil "n") to trigger a specific template, that template actually does not open (I still get the template selection frame), and the key that I specified becomes the filename of the new buffer in the other frame (with my config folder as the path, so for me ~/.dotfiles/n).

That's all if I use your emacsclient ... command to trigger the capture. If I eval the same command from within Emacs, my chosen capture template indeed pops up in the current frame, and then after finalizing the buffer (either by committing or quitting), a yequake frame pops up and the behavior described in the previous paragraph occurs, with the new buffer popping up in the frame I originally ran yequake-org-capture from.

Not sure what the implications of this are, but thought it might be a useful clue.

[alphapapa]

Hi @claytharrison , thanks for the report. Can you provide the specific steps to reproduce the problem, and details about your Emacs version and platform?

[claytharrison]

GNU Emacs 29.1 (build 3, x86_64-pc-linux-gnu, GTK+ Version 3.24.33, cairo version 1.16.0) of 2023-12-31

Org mode version 9.6.6 (release_9.6.6)

Platform - Linux (Pop!_OS 22.04)

Minimum init.el:

(unless (package-installed-p 'quelpa)
  (with-temp-buffer
    (url-insert-file-contents "https://raw.githubusercontent.com/quelpa/quelpa/master/quelpa.el")
    (eval-buffer)
    (quelpa-self-upgrade)))

(quelpa
 '(quelpa-use-package
   :fetcher git
   :url "https://github.com/quelpa/quelpa-use-package.git"))

(require 'quelpa-use-package)

(use-package yequake
  :quelpa (yequake :fetcher github :repo "alphapapa/yequake")
  :custom
  (yequake-frames
   '(("org-capture"
      (buffer-fns . (yequake-org-capture nil "t"))
      (width . 0.75)
      (height . 0.5)
      (alpha . 0.85)
      (frame-parameters . ((undecorated . t)
                           (skip-taskbar . t)
                           (sticky . t)
                             ))))))

Steps to reproduce:

1. Start emacs daemon
2. Try running emacsclient -n -e '(yequake-toggle "org-capture")'
3. Notice it does not do anything.
4. Open an emacsclient frame
5. Run the command from 2) again
6. Notice a new frame pops up, but that despite the code to call yequake-org-capture with the KEYS argument "t", the frame still prompts you for a template selection.
7. Choose "q" to abort.
8. Notice that the first frame you opened now has an empty buffer named "t" (this name will be whichever key was passed to yequake-org-capture).
9. Close that buffer if you like
10. Now try using 'M-:' to eval (yequake-org-capture nil "t")
11. Notice that we go directly to the Task template we asked for.
12. Exit with C-c, C-k, or C-w, all show basically the same behavior.
13. Notice that now a yequake frame pops up, prompting to choose an org-capture template
14. Choose "q" to abort.
15. See 8).
16. If you had instead chosen a template in step 14), you would have gotten the "t" buffer in the pop-up frame instead, with the message Clipboard pasted as level 2 subtree.

Hope this is useful!

[alphapapa]

@claytharrison Thanks, that is very thorough. I wonder if anything has changed in org-capture related to this. My Emacs workflows have changed since I wrote Yequake, and I haven't been using this feature regularly, so I might not have noticed if something had changed.

[claytharrison]

Now that you mention it, I did get a warning from yequake about `org-switch-to-buffer-other-window` being missing or something when I first started up a new emacs profile with the init.el above, but on subsequent startups I didn't get it anymore so I forgot to mention it. I see in 9.7 "org-switch-to-buffer-other-window and org-no-popups are now deprecated," but not sure why that would be a problem in 9.6.

…

On Wed, Nov 13, 2024, 06:47 Adam Porter ***@***.***> wrote: @claytharrison <https://github.com/claytharrison> Thanks, that is very thorough. I wonder if anything has changed in org-capture related to this. My Emacs workflows have changed since I wrote Yequake, and I haven't been using this feature regularly, so I might not have noticed if something had changed. — Reply to this email directly, view it on GitHub <#6 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALPH3DMWFV4V2JZNJDME37T2ALRWNAVCNFSM6AAAAABRSROCKKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZSGQ2DSNBWHA> . You are receiving this because you were mentioned.Message ID: ***@***.***>