about summary refs log tree commit diff stats
Commit message (Collapse)AuthorAge
* fix(modules/nvim/plgs/neorg): Move keybind setup to separate ftpluginBenedikt Peetz6 days
|
* refactor(modules/nvim/files): Move to their respective pluginBenedikt Peetz6 days
|
* refactor(pkgs/less): Remove and use less from nixpkgsBenedikt Peetz6 days
| | | | Nixpkgs has fixed their less `withSecure` regression.
* style(tests/README.md): FormatBenedikt Peetz6 days
|
* tests(less): InitBenedikt Peetz6 days
|
* style(treewide): RefactorBenedikt Peetz6 days
|
* fix(modules/taskwarrior): Provide an enable optionBenedikt Peetz6 days
|
* tests(flake): Hook up to the flake's check and package attributesBenedikt Peetz6 days
|
* tests({tests,modules}): Couple via a co-importBenedikt Peetz6 days
|
* tests(tests): Initialize infrastructure and documentation for itBenedikt Peetz6 days
|
* build(flake): UpdateBenedikt Peetz6 days
|
* style(treewide): FormatBenedikt Peetz12 days
|
* fix(pkgs/mpp): Include the `mpc` zsh completionsBenedikt Peetz12 days
|
* refactor(legacy/conf/mpd): Move to a unified `mpd` by-name moduleBenedikt Peetz12 days
|
* fix(legacy/wms/river): Ensure that `mpc` is available to riverBenedikt Peetz12 days
| | | | | There is currently no way to tell river where the `mpc` executable is, thus requiring it to be present in the `$PATH`.
* refactor(pkgs/mpc): Rename to `mpp`Benedikt Peetz12 days
| | | | | | | Nixpkgs has renamed their `mpc-cli` package to just `mpc`, turning this `mpc` package into an override for the official nixpkgs one. I think being explicit about the fact, that `mpc` is wrapped is the best solution here.
* build(treewide): UpdateBenedikt Peetz13 days
|
* fix(flake/packages): Re-add the `nvim` packageBenedikt Peetz2024-11-09
| | | | | | This implementation is by-far better than the previous one and avoids a whole set of problems that came, with trying to re-implement the NixOS module system.
* fix(modules/nvim/options): Use shell binary, not only directoryBenedikt Peetz2024-11-09
|
* fix(modules/nvim/plgs/neorg): Reactivate the calendar moduleBenedikt Peetz2024-11-09
| | | | `nvim` has by now reached version 0.10.0 in nixpkgs.
* fix(modules/nvim/plgs/neorg): Use updated versionBenedikt Peetz2024-11-09
|
* refactor(modules/legacy/conf/nvim): Move to `by-name`Benedikt Peetz2024-11-09
|
* fix(pkgs/stamp): Fallback to `dot-license`Benedikt Peetz2024-11-04
| | | | | Otherwise, licensing files which need a dot-license becomes really bothersome.
* chore(pkgs/yt): 1.3.1 -> 1.3.2Benedikt Peetz2024-11-04
|
* refactor(pkgs/tree-sitter-yts): Move to `yt`Benedikt Peetz2024-11-04
|
* fix(modules/less/command.less): Improve keymappingsBenedikt Peetz2024-11-03
| | | | | This removes some unneeded ones and adds the ARROW UP mapping to ensure symmetry to the ARROW DOWN mapping.
* fix(modules/legacy/conf/nvim/plgs/femacro): Add additional config functionBenedikt Peetz2024-11-02
|
* style(modules/legacy/conf/nvim): FormatBenedikt Peetz2024-11-02
|
* feat(flake/treefmt.nix): Specify lua formatter settingsBenedikt Peetz2024-11-02
|
* fix(lib/maybeMerge): Actually output a warning if things will be overriddenBenedikt Peetz2024-11-02
|
* fix(flake/nixosConfigurations/common): Actually enable the new `less` moduleBenedikt Peetz2024-11-02
|
* fix(pkgs/less): Temporary override broken `less` packageBenedikt Peetz2024-11-02
| | | | | | The `less` package in `nixpkgs` is currently compiled with secure mode enabled, which prevents less from e.g. loading its configuration file.
* fix(modules/less/command): Re-enable `#stop`Benedikt Peetz2024-11-02
| | | | | | | | This feature regressed in less 661 but was fixed in 668. See: - https://github.com/gwsw/less/issues/551 - https://github.com/gwsw/less/commit/2526c72b1212b2a1f34bdda88a93b0c7f7b4a16a
* refactor(modules/legacy/less): Migrate to `by-name`Benedikt Peetz2024-11-02
|
* fix(modules/legacy/alacitty): Update config to include new subsectionBenedikt Peetz2024-11-01
|
* build(treewide): UpdateBenedikt Peetz2024-11-01
|
* build(modules/lf/secrets/update_secrets.sh): Migrate to `nix-shell` shebangBenedikt Peetz2024-11-01
| | | | | Lix does not (maybe yet) support the nix3 `nix shell` shebang. Thus, switching to a version supported by both is the obvious choice.
* chore(modules/legacy/taskserver/projects): UpdateBenedikt Peetz2024-11-01
|
* chore(modules/taskserver/secrets): Rotate certificatesBenedikt Peetz2024-10-31
|
* chore(pkgs/yt): 1.3.0 -> 1.3.1Benedikt Peetz2024-10-31
|
* feat(modules/legacy/nvim/plgs/luasnippets): Use the git config valuesBenedikt Peetz2024-10-31
| | | | Using hard-coded values could result in desyncs.
* refactor(modules/git): Migrate and parameterizeBenedikt Peetz2024-10-24
|
* feat(modules/unison): Auto-ignore all symlinks made with `home.file`Benedikt Peetz2024-10-24
| | | | | | | This allows us to actually sync symlinks, as most of the unwanted ones are already ignored. And syncing the `.direnv` gc-roots is not that much of a problem, because these are re-generated when they point to a non-existent path.
* refactor({modules,pkgs}/by-name-overlay): De-duplicate in a `nixLib`Benedikt Peetz2024-10-24
|
* refactor(modules/unison): Migrate to `by-name` and parameterizeBenedikt Peetz2024-10-24
|
* build(build.sh): Fail, when the build failsBenedikt Peetz2024-10-24
|
* feat(modules/legacy/impermanence): Merge with the system configBenedikt Peetz2024-10-24
| | | | | | | | Using the home-manager module uses a non-root fuse bindfs, which is obviously slower than simply mounting the directories via bind mounts. Besides, mounting them via the bindfs fuse mount, results in a potential DOS, when this mount processes runs out of open file descriptors (leading to the issue described in the `nix -> lix` commit.)
* refactor(modules/impermanence): Move all optional dirs to their modulesBenedikt Peetz2024-10-24
| | | | | This increases the locality of configurations and makes it easier to see, what happens, when you enable a module.
* feat(modules/nix): Switch to `lix`Benedikt Peetz2024-10-24
| | | | | | | | | `nix` is currently constantly failing to open some of the sqlite databases, which is just really obnoxious. Although, as I have later found out, this is caused by a `bindfs` fuse mount on `~/.cache`. This process is running against the `1024` open fd soft limit, and thus refusing to open another one for the database.
* fix(modules/zsh): Support completion againBenedikt Peetz2024-10-23
| | | | | These mappings were missed in the `zsh` rework, and are apparently needed for completion to work.
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-29 05:29:24 +0000