* nix: add options 'nixPath' and 'keepOldNixPath'
By default, the system value for $NIX_PATH is kept as a fallback.
To completely override the system value for $NIX_PATH:
nix.keepOldNixPath = false;
* nix: add more tests
* nix: add a declarative alternative to Nix channels
This adds a new option, 'nix.channels'. It's the Nix channels equivalent
of the 'nix.registry' option, and compatible with pre-Flake Nix tooling
including nix-env and nix-shell. Like 'nix.registry', this option is
useful for pinning Nix channels.
Channels defined in the new option can coexist with channels introduced
through the nix-channel command. If the same channel exists in both, the
one from Home Manager will be prioritized.
* nix: add news entry
* nix: make channels respect use-xdg-base-directories
* nix: remove 'with lib;'
---------
Co-authored-by: Michael Hoang <enzime@users.noreply.github.com>
Some prefer to handle modules-* imperatively for more flexibility, e.g.,
via waybar "includes" (https://github.com/Alexays/Waybar/wiki/Configuration).
Current HM defaults modules-* to `[]` instead of null for most other
settings, so they get written to ~/.config/waybar/config even though
no explicit configuration has been made.
And they take precedence over the imperative ones because of how
waybar's includes work:
====
Paths to additional configuration files.
Each file can contain a single object with any of the bar configuration options. In case of duplicate options, the first defined value takes precedence, i.e. including file -> first included file -> etc. Nested includes are permitted, but make sure to avoid circular imports.
For a multi-bar config, the include directive affects only current bar configuration object.
====
We currently have no way of specifying the sender's name inside the
From field, making a patch sent through `git send-email` appear as
coming from "xxx@domain.com".
In this commit we make this field follow the standard
realName <email>
format.
In the generated neomutt configuration, source the optional
`vim-keys.rc` before applying the `binds` configuration, to allow the
user to override keybindings from `vim-keys.rc`.
Adapt the test files to account for the empty space left by the unset
`vimKeys` option.
Helps avoid successful build but Sway failing to start.
To meaningfully test this, I had to actually use `pkgs.sway` in the
test rather than the stub, but left all other tests using the stub and
changed them to skipping the test.
We need to pass `--unsupported-gpu` as Sway checks for `nvidia` in
`/proc/modules`, and the Nix sandbox has `/proc/modules` available.
Adds a module to enable managing Remmina, an RDP client, with a Home
Manager module, providing a systemd service and mimetype integration
that can be disabled if so desired.
Support alternate configuration files for k9s in darwin where XDG is
not mandated and k9s expects configuration files in
`~/Library/Application Support/k9s/`.
* neovim: add extraWrapperArgs option
pass external arguments to neovim-unwrapper
this gives users more flexibility in managing neovim configuration
* neovim: add test for `extraWrapperArgs`
The current module constrains to values in enum but khal supports RGB
colors as well
khal.readthedocs.io/en/latest/configure.html#the-calendars-section !
(be careful when setting an RGB value, it has to be quoted else it is ignored, got bitten by it with a manual config )
NB: It's also not possible to set addresses khal.readthedocs.io/en/latest/configure.html#the-calendars-section
Nixpkgs has recently made a few major changes to its
xdg-desktop-portal package, which silently breaks our module here:
- The NIXOS_XDG_DESKTOP_PORTAL_CONFIG_DIR variable patch has been
removed (in favor of putting portal configurations in /etc or
XDG_CONFIG_HOME).
- A new variable, NIX_XDG_DESKTOP_PORTAL_DIR, was introduced in a
patch to avoid setting XDG_DESKTOP_PORTAL_DIR (which also affected
portal configuration reading, not only portal definitions)
I updated our module to match the changes, but this breakage also made
me revisit this module and look into some improvements.
Long story short, I think it's worth it to make it more similar to the
NixOS one, as it will make behavior more predictable and consistent.
The main change is relying on the upstream linked systemd
unit (instead of using systemd.user.services), and setting the
environment variables globally instead of scoping it to the unit, as
it's a very global thing anyway.
The documentation for the option says...
> If you want a default binding to be passed through to the website,
> bind it to null.
but if you actually try to set a key to `null`, it causes an error.
> A definition for option
> `programs.qutebrowser.keyBindings."<Ctrl+Shift+Tab>".normal' is not
> of type `strings concatenated with " ;; "'.
So this commit implements unbinding as it is documented.
Adding unmailboxes option to neomutt, which adds the `unmailboxes`
option it to every account.email with neomutt enabled.
See https://www.mutt.org/doc/manual/#mailboxes for more.
Borgmatic has support for Borg's pattern matching. It is mutually
exclusive with the existing `sourceDirectories` option, so assertions
have been added to make sure that both are not set at the same
time (but also that at least one of them is). Additionally, tests have
been added to test the following configurations: `patterns` instead of
`sourceDirectories`, both at the same time, and neither.
This commit changes the way aliases are rendered, adding the "no more
flags" flag `--`, which means that, for example, the alias `"-" = "cd
-";` will work as expected. (I was getting a syntax error before this
change.)
Additionally, now the alias key is shell escaped, which may help some
edge cases. I'm honestly not sure if this part is necessary since I
assume an alias can't contain spaces anyway, but it definitely
shouldn't break anything.
* khal: fix contact integration
- Add tests for contact+khal
- Make options `color`/`priority` available for contact accounts
* khal: add separate calendar for each contact collection
A contact account may have multiple VCARD collections, but Khal doesn't
search recursively. Collection folder names must be hardcoded, and each
has its own calendar.
- Add khal.collections option for contact accounts
- Default to previous setup for accounts with a single collection
- Add tests
* khal: specify how priority is defined by Khal
See https://khal.readthedocs.io/en/latest/configure.html
The `exec` command does not do any shell parsing and does not
understand the `&&` which is how the extraCommands are added after
dbus activation. There doesn't seem to be a reason for `exec` here
anyway so just remove it and allow shell parsing.
neomutt: Updated options and added tests
neomutt: Added test for individual mailbox type
neomutt: Formatted code
neomutt: Enable ssl_force_tls based on IMAP instead of SMTP
neomutt: Applied suggestions from @chayleaf
neomutt: fix breaking tests
* imapnotify: enable STARTTLS if enabled in email account config
Since version 2.3.10 goimapnotify supports starttls. In version 2.3.11 a
typo in the settings was fixed, using tlsOptions.starttls to enable it.
This commit enables starttls in the goimapnotify config file if it is
enabled in the email account's imap settings.
* imapnotify: test generated config file
Compare the generated service config file with a manually verified
sample.
workspace 10 is part of upstream's default config, but was missing in home-manager.
The initial "sway: add module" PR (02d6040003) went through multiple iterations and had workspace 10 included for a brief moment.
Until the author removed it in a force-push commenting
> Have removed the last change which added bound ${modifer}+0 to workspace number 10 as this messed up workspace numbering in sway.
The reason might have been, that sway used to sort the workspaces in the order they appeared in the config.
Attribute sets in nix are sorted, but not "naturally sorted", meaning `bindsym Mod1+0 workspace number 10` comes before `bindsym Mod1+0 workspace number 1`.
It's unclear if that's what really happened. A workaround would have been to use `lib.lists.naturalSort` in `keybindingsStr`.
But I cannot reproduce this anymore in any way.
I assume this has been fixed many years ago by now.
upstream config: 020a572ed6/config.in (L113-L134)
This ensures that `KITTY_SHELL_INTEGRATION` is manually set to the
user's `shellIntegration.mode`. This is necessary because sometimes
the variable is not set in subshells or splits.
`shellIntegration.mode` is also now ensured to always contain `no-rc`.
