Cursor colour no longer changes to match hiragana/katakana mode.

[JordanAnthonyKing]

Hi, after a recent bump to this mode by doom-emacs the cursor colour no longer changes to match the input mode. I don't believe that this is custom behaviour implemented by Doom.

Can you confirm if this was deliberate?

[conao3]

Maybe your issue comes from #189. Please test clean emacs (emacs -Q) and show minimum init.el, reproduce steps.

[JordanAnthonyKing]

Here is a screenshot of a fresh emacs install:

I have:

• Added melpa via customise
• Installed ddskk
• Opened the scratch buffer
• Run M-x skk-mode

Toggling hiragana and kana via q does not change the colour of the cursor. Candidate text does match the colour shown in the modeline however.

Back to my doom config: I have configured straight to pull the master branch of this repo and the same is true:

Thanks

[JordanAnthonyKing]

An extra note, when opening emacs after updating ddskk this message is shown in the minibuffer:

Error in post-command-hook (ccc-update-buffer-local-frame-params): (void-function facement-color-equal)

So yes, it would seem that this was introduced by #189

[conao3]

Confirmed. cursor color now does not change. your message, Error in post-command-hook (ccc-update-buffer-local-frame-params): (void-function facement-color-equal) is directly indicate this issue. (I cannot get this message) But there're no facemenu-color-equal I think...? (Stopped using facemenu functions and removed its dependencies in #189.)

If we get backtrace, this debugging could go forward. If you eval (setq debug-on-error t) before that message, your Emacs show some backtrace?

[JordanAnthonyKing]

Although I can't seem to get any backtrace when emacs loads, if I call ccc-set-buffer-local-cursor-color manually I get this:

Debugger entered--Lisp error: (void-function facemenu-read-color)
  facemenu-read-color("Cursor color: ")
  byte-code("\301\30\302\10!)C\207" [prompt "Cursor color: " facemenu-read-color] 2)
  call-interactively(ccc-set-buffer-local-cursor-color record nil)
  command-execute(ccc-set-buffer-local-cursor-color record)
  counsel-M-x-action("ccc-set-buffer-local-cursor-color")
  #f(compiled-function (x) #<bytecode -0x16d723f3d35b6b80>)("ccc-set-buffer-local-cursor-color")
  ivy-call()

[conao3]

I confirmed ddskk cursor no longer changes its color but my ccc-set-buffer-local-cursor-color works well. I tested like below step. Please try it.

1. Save below file as ~/.debug.emacs.d/ddskk/init.el

   
   ;; ~/.debug.emacs.d/ddskk/init.el

;; you can run like 'emacs -q -l ~/.debug.emacs.d/ddskk/init.el' (when load-file-name (setq user-emacs-directory (expand-file-name (file-name-directory load-file-name))))

(prog1 "leaf" (custom-set-variables '(package-archives '(("org" . "https://orgmode.org/elpa/") ("melpa" . "https://melpa.org/packages/") ("gnu" . "https://elpa.gnu.org/packages/")))) (package-initialize) (unless (package-installed-p 'leaf) (package-refresh-contents) (package-install 'leaf)))

(leaf *initialize-emacs :custom ((debug-on-error . t) (init-file-debug . t)))

(leaf skk :ensure ddskk :bind (("C-\" . skk-mode)) :custom ((default-input-method . "japanese-skk")))

2. Run Emacs with loading that file
    `emacs -q -l ~/.debug.emacs.d/ddskk/init.el`
3. After emacs launched, `M-x ccc-set-buffer-local-cursor-color red`
4. I confirmed cursor color change into red without any errors.

[JordanAnthonyKing]

I'm unsure how you're running ccc-set-buffer-local-cursor-color from M-x with a parameter. Is there a prompt after doing M-x ccc-set-buffer-local-cursor-color RET? Running M-x eval-expression and then (ccc-set-buffer-local-cursor-color red) also doesn't work.

If these functions require a parameter they shouldn't really be invokable from M-x anyway right?

[conao3]

Yes, actual key sequence, M-x ccc-set-buffer-local-cursor-color RET red. And it makes my cursor red. Please make sure remove ~/.debug.emacs.d/ddskk/elpa directory if you have already.

[JordanAnthonyKing]

Apologies, I was testing on doom emacs with ivy. On vanilla emacs M-x ccc-set-buffer-local-cursor-color RET red does indeed work as expected.

[conao3]

OK, we can go forward next step. Why ccc-set-buffer-local-cursor-color works well, ddskk cursor no longer changes its color? I'll look into this but I don't have much time on this issue...

For now, I'll revert #189, please test the reverted version ddskk. (maybe after 2 or 3 hours, MELPA will give you that version)

[JordanAnthonyKing]

Good to hear 🙂 I'm about to perform a fresh install of doom. I will let you know.

[JordanAnthonyKing]

Has #189 been reverted? I still see it in the commit history?

[conao3]

I miss clicked Close pull-request! Reopen and merged! :) Please test after 2 or 3 hours later.

[JordanAnthonyKing]

Hi, I don't believe reverting #189 has fixed the issue. Looking at doom it seems that the issue was introduced some time before commit cec99365fb3ad725c44514b918696ab153596b8d

[tats]

A current workaround is evaluating (require 'facemenu), because recent Emacs 28 no longer preload facemenu.el.

The fix by https://github.com/skk-dev/ddskk/pull/189 looks correct, so I believe reverting is unneeded.

Please note that ccc.el is provided by the ccc package instead of the ddskk package, so take care of the version of the ccc package.

[JordanAnthonyKing]

I agree reverting was probably unnecessary, however I was running emacs 28 with the latest version of this entire repo, ddskk, ccc etc. and was not seeing the desired behaviour.

Shall I try again after the revert has been reverted?

[tats]

The current code needs (require 'facemenu) for Emacs 28 because of reverting https://github.com/skk-dev/ddskk/pull/189.

Also note that the inline functions are expanded into the caller, so you need to recompile ddskk with the updated ccc if the inline functions in ccc.el are updated.

[JordanAnthonyKing]

I can confirm this works now.