I have a problem with my emacs config...
When trying to insert dead chars like ', ~ and " when in emacs-lisp-mode, the only thing that happens is the minibuffer that show "Emacs-Lisp:" and my char isn't inserted.
Note that it doesn't happend when in C/C++ mode :/
My keyboard config is "us_intl" with dead keys.
Emacs version: 24.3.1
My config:
https://gist.github.com/anonymous/61880d721172c0cf0db1
Related
vscode > Preferences: Open Settings (JSON)
{
"vim.normalModeKeyBindingsNonRecursive": [
{
"before": ["t", "s"],
"commands": ["python.sortImports"],
}
]
}
I set it up as above, but it doesn't work. What could be wrong?
I'm typing 't' and 's' in vim-normal mode.
ctrl+alt+s is woring. but i wanna have custom vim shortcuts
Try reinstalling the vim extension. If this fails, you could try the following:
Are your configurations correct?
Adjust the extension's logging level to 'debug', restart VS Code. As each > remapped configuration is loaded, it is outputted to console. In the Developer > Tools console, do you see any errors?
debug: Remapper: normalModeKeyBindingsNonRecursive. before=0. after=^.
debug: Remapper: insertModeKeyBindings. before=j,j. after=<Esc>.
error: Remapper: insertModeKeyBindings. Invalid configuration. Missing 'after' key or 'command'. before=j,k.
Misconfigured configurations are ignored.
Does the extension handle the keys you are trying to remap?
VSCodeVim explicitly instructs VS Code which key events we care about through > the package.json. If the key you are trying to remap is a key in which vim/vscodevim generally does not handle, then it's most likely that this extension does not receive those key events from VS Code. With logging level adjusted to 'debug', as you press keys, you should see output similar to:
debug: ModeHandler: handling key=A.
debug: ModeHandler: handling key=l.
debug: ModeHandler: handling key=<BS>.
debug: ModeHandler: handling key=<C-a>.
As you press the key that you are trying to remap, do you see it outputted here? If not, it means we don't subscribe to those key events. It is still possible to remap those keys by using VSCode's keybindings.json.
Source: https://marketplace.visualstudio.com/items?itemName=vscodevim.vim#debugging-remappings
I installed the latest branch of coc but have been having trouble using it.
For some reason, whenever I enter Insert mode, this error keeps flashing:
[coc.nvim] error: at NeovimClient.emitNotification (/Users/carlosgrijalva/.vim/plugged/coc.nvim/build/index.js:1
I also attempted to install certain extensions like coc-tsserver and got the following error:
[coc.nvim] Error on install coc-tsserver: Error: coc-tsserver 1.6.3 requires coc.nvim >= >=0.0.79, please update coc
.nvim.
Not sure what's going on since i'm using the latest release. Below is my .vimrc:
" PLUGINS
" ===================================
call plug#begin('~/.vim/plugged')
Plug 'morhetz/gruvbox'
Plug 'neoclide/coc.nvim', {'branch': 'release'}
Plug 'vim-airline/vim-airline'
Plug 'vim-airline/vim-airline-themes'
Plug 'junegunn/fzf', { 'do': { -> fzf#install()} }
Plug 'junegunn/fzf.vim'
call plug#end()
" VIM General
" ===================================
set nocompatible
syntax on
set showmode " Shows mode at bottom of window
set relativenumber " Relative number line
set showmatch " Show matching brackets.
set smartcase " Do smart case matching
set smartindent " Do smart indenting
set incsearch " Incremental search
nnoremap <Leader><space> :noh<Enter>
set number " Show line numbers
set noswapfile " Disables creating swap files
set undodir=~/.vim/undodir
set tabstop=2
set shiftwidth=2
let g:gruvbox_contrast_dark='hard'
set termguicolors
colorscheme gruvbox
set background=dark
imap jj <Esc>
let mapleader=","
You're not using the latest coc.nvim release, try to delete ~/.vim/plugged/coc.nvim and install again.
The following is an example of how my flycheck errors show up on emacs:
Method name "createQATask" doesn't conform to
'[a-z_][a-z0-9_]{2,30}$' pattern [invalid-name]
Here are the checkers I’m running (checked through C-c ! v):
Syntax checkers for buffer __manifest__.py in python-mode:
First checker to run:
python-flake8
- may enable: yes
- executable: Found at /usr/bin/python3
- configuration file: Not found
- `flake8' module: Found at "/home/devdesk4/.local/lib/python3.5/site-packages/flake8/__init__.py"
- next checkers: python-pylint, python-mypy
Checkers that may run as part of the first checker's chain:
python-pylint
- may enable: yes
- executable: Found at /usr/bin/python3
- configuration file: Found at "/home/devdesk4/.pylintrc"
- `pylint' module: Found at "/home/devdesk4/.local/lib/python3.5/site-packages/pylint/__init__.py"
- next checkers: python-mypy
Checkers that could run if selected:
python-pycompile select
- may enable: yes
- executable: Found at /usr/bin/python3
- next checkers: python-mypy
I’ve tried disabling python-pylint’s configuration file, but the behavior still remains. I’ve also tried a minimal .emacs configuration which only contained the following:
(setq package-archives
'(("gnu" . "http://elpa.gnu.org/packages/")
("marmalade" . "http://marmalade-repo.org/packages/")
("melpa" . "http://melpa.milkbox.net/packages/")
("melpa-stable" . "https://stable.melpa.org/packages/")))
(use-package flycheck
:ensure t
:init
(global-flycheck-mode t))
Are those symbols (", ') not being displayed properly, or is this some sort of default flycheck configuration that I can override?
April 25, 2020 Update (Possible Solution)
Weirdly, I executed pip install --upgrade pylint just to check if I really had the latest version of pylint, and it upgraded from 2.3.0 to 2.4.4, and that fixed the issue.
However, this solution conflicts with using https://pypi.org/project/pylint-odoo/, because it reverts me back to version 2.3.0 which has those html-escape sequences.
Same-day update
It’s confirmed to be an upstream bug in Pylint.
The following is an example of how my flycheck errors show up on emacs:
Method name "createQATask" doesn't conform to '[a-z_][a-z0-9_]{2,30}$' pattern [invalid-name]
The linter evidentially believes it should be producing HTML output.
By the looks of it you are running all of python-flake8, python-pylint, and python-mypy. I suggest that you firstly test them one at a time to establish which one is producing that output, and then look at the documentation for that tool to find out how to prevent it from generating HTML.
I’ve been using org-mode for my GTD system for a little while and now that I’m happy with it I decided to set it up on all my computers. My main machine for the last year or so has been on Archlinux and that’s where I did my setup originally now I’ve added a windows computer and a Pixelbook running Emacs in crostini. I’m syncing my org files with Dropbox and that works fine. I’ve copied my .emacs file to each computer and changed some file paths for my windows machine. Almost everything seems to be working fine except for something strange with my Capture Templates.
On the Pixelbook the Templates using file+olp+datetree throws an invalid capture target specification ie:
("m" "Meeting Note" entry
(file+olp+datetree "~/Dropbox/org/Journal.org" "Meetings")
"* Description: %^{Description|Meeting} %T
- Participants: %^{Participants}
- Location: %^{Location}
** Notes:
- %?
** Next Steps: :ReFile:
- [ ] ")
On the windows computer my Inbox ToDo capture throws the same error in the minibuffer:
("t" "New TODO to Inbox" entry
(file "C:/Users/jgallo/Dropbox/org/GTD.org" "Inbox")
"* TODO %?" :prepend :time-prompt t)
Anyone have any ideas what I might be missing?
The file+olp+datetree target expects just one argument, the filename. The same is true of the file target in your Windows configuration. If you want to use a headline for the target, you can use file+headline.
The details of these can be seen in the Org manual at Org capture template elements
I cannot make emacs to jump to next-error, previous-error, first-error in vhdl-mode.
I am using FSF Emacs 23.3.1 with recent vhdl-mode 3.33.28 under ubuntu.
I can compile with Modelsim and I get compiler error list:
-- Loading package standard
-- ...
-- Compiling entity foo
** Error: path/foo.vhd(22): (vcom-1136) Unknown identifier "std_olgic".
I tried to adapt error regexp (in compiler-setup) to the simplest one:
\*\* Error: \([a-zA-Z\/_.]*\)(\([0-9]*\)).*
When I use it this way I can see that it parses errors correctly:
sed "s/\*\* Error: \([a-zA-Z\/_.]*\)(\([0-9]*\)).*/\1 \2/" ...
path/foo.vhd 22
I changed "File subexp index" and "Line subexp index" respectively to 1 and 2 but still I cannot jump around errors.
The following config resolves this bug for me
'(vhdl-compile-use-local-error-regexp t)
(add-to-list 'compilation-error-regexp-alist '("** Error: \\(.+\\)(\\([0-9]*\\)):" 1 2))
In Emacs regexps, you need to double-escape parens in a string (explanation). Try something like this: \*\* Error: \\([a-zA-Z0-9/_.]+\\)(\\([0-9]+\\)).*
Sorry to bump an old topic but I just ran into this issue and got things to work out for me.
Here are the settings I used to get it to work:
Regexp:
\(ERROR\|WARNING\|\*\* Error\|\*\* Warning\)[^:]*:\( *[[0-9]+]\)? \(.+\)(\([0-9]+\)):
File subexp index: 3
Line subexp index: 4
Vhdl Compile Use Local Error Regexp (under Vhdl Compile group): Off
And here's my story about it:
http://www.velocityreviews.com/forums/t957495-emacs-vhdl-mode-next-error-previous-error-and-first-error-are-not-working.html
:P
Hope this helps!