about summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* machines/sternenseemann/wolfgang: disable lorri HEAD mastersternenseemann24 hours1-2/+0
| | | | | | I don't use it actively anymore and it's kind of annoying when it starts doing stuff in the background because you cd into the wrong folder by accident.
* modules/sternenseemann/sway: make color selection more granularsternenseemann24 hours1-19/+31
| | | | | | | This is done in the same step as changing the default colors to a Plan9/acme inspired look since I can't be bothered to split the changes up. A handy list for the acme colors can be found here <https://github.com/prodhe/acme-setup/blob/master/acme-colors.md>.
* profiles/base: be more explicit with adding apptainerdevhell7 days1-1/+6
|
* machines/gunnr: remove kernelPackagesdevhell8 days1-1/+0
| | | | | Not sure why this didn't trigger earlier, but now nix will complain that it already exists of course. Removing it fixes that.
* profiles/base: fix my idiocydevhell8 days1-1/+1
|
* profiles/base: add singularitydevhell8 days1-0/+1
| | | | Need to build a container. Yay.
* machines/{eir,herja,hildr}: remove wavemondevhell8 days3-3/+0
| | | | This is already defined in the base profile.
* machines/gunnr: Remove gastly rtl88xxau driverdevhell10 days1-17/+3
| | | | | | | Blessed be my new ALFA AWUS036AXML, and TP-Link Archer with that nasty Realtek driver can go eat a big fat dorito. The only thing that I had to do was to blacklist `btusb`. Somehow there seems to be a race condition which prevents WiFi from loading properly.
* machines/gunnr: temporarily disable custom NTPdevhell11 days1-3/+3
|
* profiles/packages: add rustcatdevhell12 days1-0/+1
|
* machines/sternenseemann/wolfgang: install quasselsternenseemann2024-04-291-0/+1
|
* profiles/packages: add curliedevhell2024-04-281-0/+1
|
* machines/gunnr: forcefully disable sshddevhell2024-04-231-0/+3
| | | | | Work is going nuts with cybersecurity, so I'm preempting any weird discussions. I don't really need to ssh into gunnr anyway.
* profiles/base: add fzf optionsdevhell2024-04-151-0/+4
|
* Revert "profiles/base: fix sourcing of fzf's zsh keybindings"devhell2024-04-131-1/+1
| | | | | | | | This reverts commit 6b50ee59dcec9f8682c12516ceead31db10e63f9. Hmm, thought this had fixed it, but turns out that `fzf` is missing completions in `share/`, particularly `fish` and `fzf` directories are missing now. Might be an upstream issue.
* profiles/base: fix sourcing of fzf's zsh keybindingsdevhell2024-04-131-1/+1
|
* profiles/packages: rustdeskdevhell2024-04-121-0/+1
|
* firefox: Disable pinch-to-zoomaszlig2024-04-121-0/+3
| | | | | | | This is another very annoying misfeature (IMHO) that I accidentally trigger sometimes when scrolling via a multi-touch touchpad. Signed-off-by: aszlig <aszlig@nix.build>
* firefox: Do not download PDF attachmentsaszlig2024-04-121-0/+1
| | | | | | | | | | This is a behaviour that has been introduced at some point and it's been annoying to have all those PDFs laying around in my Downloads directory. If I still want to save the PDF, I can do so explicitly so I see no downside to always open PDF attachments inline (without saving). Signed-off-by: aszlig <aszlig@nix.build>
* profiles/packages: rename and add packagesdevhell2024-04-071-2/+3
|
* profiles/packages: reintroduce jqdevhell2024-04-071-0/+1
| | | | `jql` is nice, but there are still things `jq` does better.
* tests/flameshot: Remove icewm background imageaszlig2024-03-251-1/+9
| | | | | | | | | | | | | | | | | | | | | The icewm start command has been changed[1] to be "icewm-session" a while ago, which not only starts the main icewm program but a bunch of related services, one of them being icewmbg. We're doing an OCR of the flameshot UI, which is layered over the screen with partial transparency, which in turn makes the OCR result pretty unreliable since now we have a background image. Since we're not testing icewm but flameshot, I decided to just use the "icewm" command we've used before, since we're not interested in all the auxiliary programs that form a full "icewm-session". This fixes the test since with the background image gone, OCR is new able to correctly pick up the text from the flameshot UI. [1]: https://github.com/NixOS/nixpkgs/commit/73cb33678527ddc323c7fcef3d17ac85603e5b02 Signed-off-by: aszlig <aszlig@nix.build>
* pkgs/aszlig: Remove refs to axbo and librxtx_javaaszlig2024-03-251-2/+0
| | | | | | | I've removed the directory in 061b9452ca81f2e96373973d4e675f40e92fd7dd, but failed to remove the actual callPackage references. Signed-off-by: aszlig <aszlig@nix.build>
* vim: Remove setting g:rustc_pathaszlig2024-03-251-4/+1
| | | | | | | | | | | | Having a fallback rustc executable in my Vim closure at all times looked like a good idea in the first place, but since g:rustc_path is only used for things like :RustRun it is of very limited use (only for standalone Rust files). I had exactly *zero* uses for that since I introduced it, so there is really no need to bloat the closure size for something I don't use. Signed-off-by: aszlig <aszlig@nix.build>
* vim: Don't rustfmt without an explicit configaszlig2024-03-254-46/+4
| | | | | | | | | | | This is getting annoying when contributing to external projects that don't have a rustfmt.toml, because I constantly forget to disable it for a buffer and then get a massive diff that I need to revert. My own projects all have a rustfmt.toml anyway, so there really is no reason to do it by default nor have a custom rustfmt config in Vuizvui. Signed-off-by: aszlig <aszlig@nix.build>
* Remove aXbo stuffaszlig2024-03-254-110/+0
| | | | | | | | I'm no longer using that device and I'm not willing to fix up random Java stuff every time the build breaks for things I don't use, so let's remove it. Signed-off-by: aszlig <aszlig@nix.build>
* machines/herja: fix nix.settingsdevhell2024-03-241-1/+1
| | | | Fat fingers be damned.
* machines/herja: add nix.conf extra-platformsdevhell2024-03-241-0/+1
| | | | Seems like it's not enough to simply add binfmt.
* machines/herja: add binfmt aarch64-linuxdevhell2024-03-201-0/+1
| | | | Need this to cross-compile to RPis.
* machines/sternenseemann: reflect pinentry-gnome renamesternenseemann2024-03-151-1/+1
|
* pkgs/tvl: 2023-05-01 -> 2024-03-14sternenseemann2024-03-151-1/+1
|
* profiles/base: fix pinentryPackagedevhell2024-03-131-1/+1
|
* profiles/base: pinentryFlavor -> pinentryPackagedevhell2024-03-131-1/+1
| | | | There's been a recent upstream change.
* pkgs/profpatsch/nman: don't use non portable option --local-filesternenseemann2024-03-081-1/+1
| | | | | `man -l` is sadly not part of POSIX, but supported by both GNU's and OpenBSD's implementations of man(1) – contrary to `--local-file`.
* core/tests: Switch to new boot.zfs.package optionaszlig2024-03-071-2/+2
| | | | | | | | | | | | When fixing evaluation for supportedFilesystems in 21fdf5e0407d11524dbc0c42743358d9b973943c, I forgot about another change related to ZFS options, which leads to the following evaluation error: The option `boot.zfs.enableUnstable' can no longer be used since it's been removed. Instead set `boot.zfs.package = pkgs.zfs_unstable;` Signed-off-by: aszlig <aszlig@nix.build> Reported-by: https://github.com/devhell
* machines/gunnr: zfsUnstable -> pkgs.zfs_unstabledevhell2024-03-071-1/+1
|
* {machines,packages}/devhell: move rwpspreaddevhell2024-03-072-1/+1
|
* core/tests: Fix evaluating supportedFilesystemsaszlig2024-03-051-5/+5
| | | | | | | | | The option boot.supportedFilesystems is now[1] an attribute set instead of a list, so we need to fix all checks. [1]: https://github.com/NixOS/nixpkgs/commit/258b935d705be03e939b8205215 Signed-off-by: aszlig <aszlig@nix.build>
* pkgs/profpatsch/nman: verbose log all found manpages on failureProfpatsch2024-03-041-18/+45
|
* pkgs/profpatsch/nman: really enumerate (all) manpagesProfpatsch2024-03-041-29/+39
| | | | | | We enumerate every single manpage, then we only take the first manpage that is in our section or just the first manpage that is named correctly.
* pkgs/profpatsch/nman: rewrite directory iteration to more imperativeProfpatsch2024-03-031-19/+19
| | | | | This is gonna make it a lot easier to enumerate the manpages as well. Plus the code gets somewhat easier to read in my opinion.
* pkgs/profpatsch/nman: don’t return the full pathProfpatsch2024-03-031-15/+13
| | | | | We can construct the path of a manpage from it’s metadata alone (given a derivation output path).
* pkgs/profpatsch/nman: move checking man-prefix into enumerateProfpatsch2024-03-031-34/+43
| | | | | We can already validate these directories early, returning just a struct with the plain section name and path to section.
* pkgs/profpatsch/nman: always enumerate all manpagesProfpatsch2024-03-031-17/+19
| | | | | First step towards getting some info about which manpages are indeed available if we don’t find any.
* pkgs/profpatsch/nman: separate drv building from manpage checkProfpatsch2024-03-031-33/+47
|
* pkgs/profpatsch/nman: start returning more info from buildProfpatsch2024-03-031-6/+12
|
* pkgs/profpatsch/nman: initial debug logging & add "doc"Profpatsch2024-03-031-6/+60
| | | | | For some reason the doc output was missing from the list, but we want to sort it kinda high? At least over "bin" and such!
* pkgs/profpatsch/nman: run `rustfmt`Profpatsch2024-03-032-136/+194
|
* pkgs/profpatsch/nman: add a `--verbose` flagProfpatsch2024-03-032-25/+44
| | | | And only print debug stuff iff we enabled it.
* pkgs/profpatsch/nman: put functions into struct contextProfpatsch2024-03-031-153/+160
| | | | so that we can add a global debug flag
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-05-18 22:24:22 +0000