-
-
Notifications
You must be signed in to change notification settings - Fork 3.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Spell-Fu marks every word as incorrect #4009
Comments
Do you have aspell or flyspell enabled? |
I have aspell enabled and installed. The line in my init.el file looks like |
Yeah I'm having the same issue |
spell-fu caches words in ~/.emacs.d/.local/etc/spell-fu |
Perfect, that fixes it for me. Thanks! |
I deleted the files in ~/.emacs.d/.local/etc/spell-fu, but the problem remained. It had the same behavior with the files gone before I restarted emacs as well as after I restarted and the files were re-created. As a note, the two files I found in the directory were words_en.el.data, and words_en.txt. Any ideas as to how I could debug this? |
Usually, if deleting those files does not help, you can look inside the files and see errors, that |
Sounds smart. Since I don't know what these files are supposed to contain, could you tell me what is in yours? The first few lines of mine are: ~/.emacs.d/.local/etc/spell-fu/words_en.txt
~/.emacs.d/.local/etc/spell-fu/words_en.el.data
|
Looks like the same as mine but with |
I've done some more testing and have found the following:
I should also note that I tried each of these on a base installation of doom without any private configurations besides addding (spell +aspell) to init.el. So the problem remains and, as far as I can tell, is something to do with spell-fu on windows or how doom handles it. |
I finally discovered it was the carriage returns placed in the words_en.el.data file on Windows. Manually deleting all of them from the file brought back normal behavior. |
So the issue is that spell-fu generates the word list in the wrong encoding? (i.e. wrong newlines) Should this be reported upstream (to spell-fu)? |
That sounds like the right action to me. A few other notes about functionality if anyone cares:
I guess these are probably intended features or side-effects of spell-fu functionality and should be referred to the upstream repository as well. |
@ngharrison thanks for documenting your work around. Did you report this bug upstream? I'd like to follow it but can't find it. @hlissner you're right the line encoding for |
@shombando Thanks for the reminder to file it upstream. I guess you tend to forget about things when they start working again. It is now up: https://gitlab.com/ideasman42/emacs-spell-fu/-/issues/13 |
This might related to the issue here. I also got most words labeled wrong. Here is the post and solution that works for me. OPAnyone using The directory of
Certainly (Related: #4009) SolutionRelated packages: On arch, I installed packages Use
|
Main Problem
When I enable Spell-Fu (spell-fu-mode using SPC t s), every word in a buffer is underlined in red.
When jumping to the next spelling error (+spell/next-error using ] s), it jumps to every word.
If the word is checked (+spell/correct using z =), it does correctly tell if it is spelled correctly or gives suggestions if incorrect.
I'm on Windows 10 and just recently updated to Emacs 27.1 and the latest release of Doom. The problem didn't occur before that.
Examples
With Spell-Fu enabled:
Type z = with point on "sentence", and it prints out:
Type z = with point on "sentance", and it prints out:
System information:
The text was updated successfully, but these errors were encountered: