summary refs log tree commit diff
path: root/emacs/custom/my-prog.el (follow)
Commit message (Collapse)AuthorAgeFilesLines
* ref(lsp): switch from lsp-mode to eglotFranck Cuny2022-10-171-3/+0
| | | | | | | | | The experimentation was a disaster: lsp-mode was buggy, slow, littering the interface with a lot of things I did not understood or needed. eglot is definitely simpler and works better for me. Change-Id: Ie7e7e6f1c8e7b26aa5cb79d636f3ad53c1493617
* ref(prog): wrap comments at column 120Franck Cuny2022-10-171-2/+2
| | | | Change-Id: Idd2b673af070b7cca02b1e62d6a3c40bdad1b7ec
* ref: move flymake and eldoc configuration to their own librariesFranck Cuny2022-10-141-11/+0
| | | | Change-Id: Ida9e0f7d6ee25cf5fe8e208d1ae5c77aa28cfe0f
* fix(prog/lsp): apply a number of fixes related to LSPFranck Cuny2022-09-241-45/+0
| | | | | | | | | | | | | | | | | Now that I've switched back to LSP, there's some configuration to delete and some to update. The configuration related to rust is fixed (lsp-ui was not loading properly), and I'm also disabling the inlay hints (they are very intrusive, I'll need more tweaks to find something that works for me). For the LSP configuration itself, I'm tweaking a few more things, especially related to the UI. Finally, deleting configurations for eglot and go: eglot is replaced by lsp-mode, and go's configuration is in its own file. Change-Id: I73fb4b66c8816dd6e20efa358c81e3c306b8255f
* ref(tree-sitter): move the configuration to its own fileFranck Cuny2022-09-171-10/+0
| | | | Change-Id: I424e4d32e05d6bf21f6abb3cc68cad598af31708
* fix(prog): set python indentation to 4 spacesFranck Cuny2022-09-151-1/+1
| | | | | | That's what everyone use. Change-Id: I312dd88218e2d82be0aad8fd6377483ffb2c0390
* feat(rust): add initial configuration for rustFranck Cuny2022-08-221-1/+6
| | | | | | Use rustic and configure it to use eglot as the backend for LSP. Change-Id: I71bd51a863f1a03b0b3f2d218d970d5ea0867e9e
* ref(prog): clean up comment settingsFranck Cuny2022-06-271-2/+3
| | | | | | | Change-Id: Ic3ec94047e39618d9d1bbc5aa9afce8565b87208 Reviewed-on: https://cl.fcuny.net/c/emacs.d/+/583 Tested-by: CI Reviewed-by: Franck Cuny <franck@fcuny.net>
* fix(prog): configure eglot for buffer formattingFranck Cuny2022-06-261-12/+14
| | | | | | | | | | Load eglot when Emacs starts, and delegate buffer formatting for supported language to LSP. The LSP for nix is using `nixpkgs-fmt' already, so there's no need to configure this ourselves. Change-Id: I6ee4adcc34b30c75031974c10a47fd3f7d240bed
* feat(prog): require go-modeFranck Cuny2022-06-261-7/+5
| | | | Change-Id: I2b008d6039714475d7bbf487ac7674dd80b1bac4
* feat(prog): enable tree-sitterFranck Cuny2022-06-261-0/+9
| | | | | | | | | tree-sitter is an incremental parsing system, and this change installs the binding for Emacs. It improves significantly the code highlighting system (as it does not rely on regex when the language is supported by tree-sitter) and improves performance in general. Change-Id: Ibc3d6c7133b323d6cc63ffb248cd83f71ac5c94a
* fix(prog): simplify flymake's configuration to work with eglotFranck Cuny2022-06-261-11/+7
| | | | | | | | | | | | | | | I've been running into issues between flymake and eglot for a while now. After modifying the file, flymake would report errors on possible typo I made while editing, but once I've fixed them, the diagnostic would not be updated until I either restart eglot or kick another flymake run. After some debugging I'm pretty convinced this is due to the variable `flymake-no-changes-timeout`. I've been running for a few hours without it being set and I've not been able to reproduce the problem. The other variables are set to their default value, this is not needed. Change-Id: I770857a93b6d90193f39ee0dceaa19b0b928a3d8
* fix(prog): use gofmt to format go codeFranck Cuny2022-06-201-3/+4
| | | | | | When using LSP to format the code, I keep running into weird behavior with eglot. I think it's because the code is being formatted while flymake is running but I need more evidence.
* feat(prog): use nixpkgs-fmt to format nix codeFranck Cuny2022-06-201-0/+1
| | | | | | | This is the tool recommend by the nix community to format nix code (see [1]). [1] https://github.com/nix-community/nixpkgs-fmt
* feat(eglot): use LSP to format go buffers instead of gofmtFranck Cuny2022-06-031-1/+3
| | | | Change-Id: Iccef9b065eb4bee917b0084fb0b34e76f09223f2
* fix(prog): add support for workspace module to goplsFranck Cuny2022-05-311-3/+5
| | | | | I've multiple go modules inside a git repository. Without support for workspaces, gopls is confused and does not work correctly.
* go: define command for compilingFranck Cuny2022-04-291-1/+7
|
* don't specify where to write Emacs filesFranck Cuny2022-04-291-1/+0
| | | | | We don't care, as long as these files are ignored in `.gitignore' it does not matter where we write them.
* my-prog: enable electric-pair-modeFranck Cuny2022-04-271-0/+4
|
* my-prog: more simplificationFranck Cuny2022-04-271-126/+97
|
* emacs: where to save stateFranck Cuny2022-04-161-0/+2
|
* replace company with corfuFranck Cuny2022-04-141-11/+1
|
* prog: tweak configuration for goFranck Cuny2022-03-251-1/+14
| | | | | Add a few bindings to run the tests, and make the test output more verbose by default.
* this should be the last mass refactorFranck Cuny2022-03-231-1/+0
|
* rename a few more librariesFranck Cuny2022-03-231-0/+131