Closed wofwofwof closed 3 years ago
ok, found my problem, just cleaned up my old dependencies in the packages and now iit works. No problem with xah-fly-mode. Sorry for the noise
lol, reopened again
its still open after the cleanup, but I've got confused because the keys with C- mapping are working. And there are lots of triggers that get one back into the fly-spell mode, some examples are listed:
is it possible that some part of xah now needs an interactive environment and that on daemon mode that doesen't work?
It sound to me like possible config conflict? What do you think?
Have you tried to reproduce the problem with a minimal config (i.e., emacs --no-init-file
or -q
)?
Hello TRSx80,
thanks for your support. I've tried to make a minimal config where the problem is still there. This is my .emacs.d/init.el:
(require 'package)
(add-to-list 'package-archives '("melpa" . "https://melpa.org/packages/"))
(package-initialize)
(require 'xah-fly-keys)
(xah-fly-keys-set-layout "qwertz")
(xah-fly-keys 1)
Thank you very much
alex
I don't know why it happens but this fixed it for me.
(defun my/server-fix-up()
"Make sure 'xah-fly-keys' is starting in command-mode.
https://github.com/xahlee/xah-fly-keys/issues/103
https://github.com/daviwil/emacs-from-scratch/blob/master/show-notes/Emacs-Tips-08.org#configuring-the-ui-for-new-frames"
(xah-fly-keys-set-layout "dvorak")
(xah-fly-keys t))
(if (daemonp)
(add-hook 'server-after-make-frame-hook 'my/server-fix-up))
Edit: If you define keys under 'use-package' (maybe also just 'define-key') Take out: (xah-fly-keys-set-layout "dvorak").
CelestialNebula, thanks a lot, that fixed the problem for me, too.
thanks guys. thanks @CelestialNebula . i added that to customization page http://ergoemacs.org/misc/xah-fly-keys_customization.html
If i use this fix, in the daemon mode, some calendar bindings seem to have been damaged (in insert mode), for instance (after M-x calendar) and switch to insert-mode:
i d -> diary-insert-entry i b -> diary-insert-block-entry
Does anybody has an idea of what is going on here?
What do you mean damaged? Those appear to be the default keys.
Can we reopen this issue? It is still a major issue - the mode refuses to work, its very unituitive. Any clue as to what is causing this?
;; Use xah-fly-keys a modal keybinding scheme
(require 'xah-fly-keys)
(xah-fly-keys-set-layout "qwerty")
(if (daemonp)
(add-hook 'server-after-make-frame-hook #'(lambda ()
(unless xah-fly-keys
(xah-fly-keys 1))))
(xah-fly-keys 1))
I have to resort to this.
(when (daemonp)
(add-hook 'server-after-make-frame-hook #'(lambda ()
(when (and xah-fly-keys
(not xah-fly-insert-state-p))
(set-transient-map xah-fly-command-map (lambda () t))))))
set-transient-map is failing when set during init. Minimal solution that works.
maybe open a new issue and reference to this one. i'll look into. meanwhile, i don't recommend running daemon. it creates a lot complications. is confusing in concept, also a lot technical complexities. also more complexities if you are on windows.
instead, setup ur emacs so that u can open multiple instances without emacs ever bugging u to ask questions during launch (e.g. last emacs crashed). so you can always work in mulitple intances in parallel, launch new anytime. if u code emacs, u have to resart emacs often, sometimes multiple times a day, anyway.
maybe open a new issue and reference to this one.
Will do, thanks. Need to collect my findings more concisely; daemon introduces two regressions - one this and the other setting the cursor colour/shape.
meanwhile, i don't recommend running daemon.
You're correct in this - I have found many bugs while running daemon - for ex.; kill-emacs-hook
has some quirks - I am sure there are other such quirks throughout.
I can set up my init such that the difference in opening frame times is less than half a second daemon or not daemon; but I have chosen to not do this - despite having to annoyingly restart the daemon over and over some days to reset temporary code tests - because its quite fast -- the latency in opening subsequent files is substantial even for 0.3~0.4 seconds. The bottleneck is elisp's GC threshold - when GC runs collinearly, execution time is trivially doubled. The annoyances of daemon has not reached the threshold for me to consider refactoring my init and tweak GC, mostly I have learnt to live around it.
But I agree with you, running daemon is not without its share of problems.
When I use my config with emacs everything works fine, when I use it with emac --daemon mode the xah-fly-key mode is activated but ignored until I
After this everything works fine. I use version xah-fly-keys-20200707.1310.
At the moment I have no clue whats the reason for this behaviour. On xah-fly-keys-20190223.716 that configuration works. Can somebody just try if its my configuration or a general problem with xah-fly-mode?
Thanks for that fantastic mode
cheers
alex