Don't try to validate a limited set of hardcoded options, instead just
convert them as-is. Now, users can keep all their options in a single
attribute set, including arbitrary `user_{option}`s which was impossible
to express with a hard-coded submodule. As a plus, there is also less
maintainence burden.
While the locale options were declared, the weren't used in the
generation of the config file, because the locale submodule missed a
default, which failed the tests. I added an empty attribute set as a
default, which fixes the test, and works with the defaults in the
submodule options as expected.
* gh: option to enable helper for additional hosts
`gh` can also be used with github enterprise
hosts, for which there exists no easy option
to enable the credential helper except for
directly working with `programs.git.extraConfig`.
Not sure if this is a needed addition since it's
somewhat niche, at the same time it's not very
complex and makes the life of github enterprise
a little easier.
* gh: update credential-helper tests
* gh: refactor credential helper option
this moves from `enableGitCredentialHelper` to
`gitCredentialHelper.enable` and
`gitCredentialHelper.hosts`.
* gh: lib.mkIf -> mkIf
These (and the `*MD` functions apart from `literalMD`) are now no-ops
in nixpkgs and serve no purpose other than to add additional noise and
potentially mislead people into thinking unmarked DocBook documentation
will still be accepted.
Note that if backporting changes including documentation to 23.05,
the `mdDoc` calls will need to be re-added.
To reproduce this commit, run:
$ NIX_PATH=nixpkgs=flake:nixpkgs/e7e69199f0372364a6106a1e735f68604f4c5a25 \
nix shell nixpkgs#coreutils \
-c find . -name '*.nix' \
-exec nix run -- github:emilazy/nix-doc-munge/98dadf1f77351c2ba5dcb709a2a171d655f15099 \
--strip {} +
$ ./format
This process was automated by [my fork of `nix-doc-munge`]. All
conversions were automatically checked to produce the same DocBook
result when converted back, modulo minor typographical/formatting
differences on the acceptable-to-desirable spectrum.
To reproduce this commit, run:
$ NIX_PATH=nixpkgs=flake:nixpkgs/e7e69199f0372364a6106a1e735f68604f4c5a25 \
nix shell nixpkgs#coreutils \
-c find . -name '*.nix' \
-exec nix run -- github:emilazy/nix-doc-munge/98dadf1f77351c2ba5dcb709a2a171d655f15099 \
{} +
$ ./format
[my fork of `nix-doc-munge`]: https://github.com/emilazy/nix-doc-munge/tree/home-manager
The NixOS variant of Markdown doesn't make a distinction between
`<code>` and `<literal>` or `<quote>` and... quotes, and doesn't
support `<parameter>` or `<replaceable>`. These are infrequently used
(apart from `<code>`) and don't add much, so just convert them to
simpler forms to allow the options containing them to be converted
to Markdown automatically.
A few minor syntactic adjustments were also made to make
`nix-doc-munge`'s job easier.
The Markdown options processor cannot handle rendering tables
to DocBook. This could be fixed, but as we won't be using the
DocBook output for long I just removed them for now in the interest
of expediency; they were all well-suited to being description lists
showing option types anyway, apart from one awkward case in the form
of trayer, which also had ad-hoc syntax for enumerating acceptable
values in the documentation. Since the types aren't actually used for
option processing anyway, I changed them to use `enum` and similar to
give a single description of the acceptable values without a big table.
`nix-doc-munge` can't handle these, which is understandable as I can
barely handle them either. There are a few infelicities here: the
current processor can't handle multiple terms to one description in
a description list so they get comma-separated in one case, and one
case that should ideally render as a `<figure>` with a `<figcaption>`
in HTML is reduced to a paragraph with some `<strong>` text. (Which, in
fairness, is how it rendered in practice with the DocBook anyway.) The
docs generator has since been updated to handle figures, but we can't
use it until moving off DocBook output.
These files all have options that trip up the `nix-doc-munge`
conversion tool for one reason or another (syntax that clashes with
Markdown, options that were already using Markdown syntax despite not
being marked that way, output that differs slightly after conversion,
syntax too elaborate to convert with some cheap regular expressions,
...). Translate them manually and do a little copyediting to options
in the vicinity while we're at it.
Previously, lines in .zshrc were added with quotes in keys
(e.g. ZSH_HIGHLIGHT_STYLES['comment']='fg=#6c6c6c'). However, zsh
considered these quotes to be part of the key, so the "comment" key
remained unchanged.
* aerc: fix per-account extraConfig section names
The aerc configuration file `aerc.conf` can contain 10 different
sections, but only the UI section supports what the aerc manual calls
contextual configuration. This works by appending to the section heading
either `:account=name` or `:folder=bar`.
The aerc-accounts module, however, applied `mkAccountConfig` to each
section heading declared in
`config.accounts.email.accounts.<name>.aerc.extraConfig.*`. This means
home-manager will generate files with `[general:account=default]` and
the options will not be recognized by aerc.
To address this, and since it doesn't make sense for other sections to
only be under a single account's scope, an assertion has been added
to confirm that only sectons that support contextual config (i.e.,
only the UI section) is declared.
This also addresses confusions like declaring
`accounts.email.accounts.*.aerc.extraConfig.general.unsafe-accounts-conf
= true` and triggering a warning message because
`programs.aerc.extraConfig.general.unsafe-accounts-conf` was unset.
This commit also updated documentation throughout the aerc modules to
be in line with this change, and fixed minor typos/formatting therein.
Co-authored-by: Genevieve <genevieve@sunlashed.garden>
* aerc: make assertion plaintext and add test case
This commit adds a test case to check both the warning on unset
`unsafe-accounts-conf = true` when aerc accounts are configured
with Nix, and the new assertion when per-account configuration
contains unsupported subsections (i.e. general).
It also fixes minor formatting issues and typos.
Nushell has the option to source from the login.nu file in the case
that nushell is used as a login shell. This commit adds the login file
alongside the existing config and env files as another configuration
option.
Previously, IMAP was preferred over notmuch, even if notmuch was
configured, causing problems with setting account flavor (which
automatically sets IMAP settings). The new backend order is:
notmuch > IMAP > maildir
This also fixes the notmuch DB path being set to the wrong location.
The notmuch DB is located at the maildir base path, not in each
account's maildir.
The current `lla` alias, together with the `total-size` option try to get
the size of the `..` directory, and for this has to recursively open all
sibling folders. This may be super slow if some of those siblings
contain too many files, and raise a ton of useless errors if some of
those siblings contains non-readable files.
I'm suggesting to use `-A` instead, which will skip the obvious `.` and
`..` folders.
While here, I think we could also add `llt`.
Many of the terminals supported inside emacs work perfectly fine with STARSHIP.
The TERM=dumb case already handles the tramp and eterm cases, so as far as I can
tell, this is basically just a check for the benefit of OLD versions of
term-mode (see
emacswiki.org/emacs/AnsiTerm#:~:text=Historically%2C%20'M%2Dx%20ansi%2Dterm,the%20older%20'C%2Dc'%20binding.,
which indicates that it also now handles colors).
PR #3747 renamed the option wayland.windowManager.sway.systemdIntegration
to wayland.windowManager.sway.systemd.enable.
This commit simply updates documentation to reference the new format.
Starship has an advanced, experimental feature where fancy stuff in the
prompt can be replaced with something more simple after the command is
ran. This is very helpful for copy and pasting shell history somewhere
else.
docs: https://starship.rs/advanced-config/#transientprompt-and-transientrightprompt-in-fish
Fish is currently the only shell as far as I can tell that both
home-manager and starship support for this feature. Since the function
has to be called after starship is loaded, this seems like the best
place to put it.
format
* aerc: add space after definitions
* aerc: only generate files, if options were set
* aerc: improve file permission warning
* aerc: remove redundant access to builtins
* aerc: allow overwriting of derived values
the order of merging the config subsets did not allow the user to specify
outgoing, source and password command values,
if they were previously derived from the SMTP, IMAP, Maildir etc config.
The values from `account.<name>.extraAccounts` now have the highest precedence.
Appropriate tests were added as well.
* aerc: write primary account first
* Add infrastructure for contacts and calendars
This also adds the modules
- programs.vdirsyncer,
- programs.khal, and
- services.vdirsyncer
that integrate with the new infrastructure.
Co-authored-by: Andrew Scott <3648487+ayyjayess@users.noreply.github.com>
Co-authored-by: Sebastian Zivota <sebastian.zivota@mailbox.org>
wip
* vdirsyncer: allow option userName, disallow userNameCommand
1. account option `userName` is now allowed by `programs.vdirsyncer`
2. The commented out account option `userNameCommand` was required to be set
by `programs.vdirsyncer` (e.g. as `null`).
It is now disallowed (commented out) by vdirsyncer.
* khal: added options 'color' and 'priority'
* Apply nixfmt
---------
Co-authored-by: Sebastian Zivota <sebastian.zivota@mailbox.org>
Co-authored-by: Johannes Rosenberger <johannes.rosenberger@jorsn.eu>
Co-authored-by: Johannes Rosenberger <johannes@jorsn.eu>
Co-authored-by: Robert Helgesson <robert@rycee.net>
* boxxy: add module
* boxxy: added nikp123 to maintainers list
* boxxy: use mkPackageOption instead for the package
Co-authored-by: Naïm Favier <n@monade.li>
* boxxy: use yaml generator instead of json
Co-authored-by: Naïm Favier <n@monade.li>
* boxxy: various fixes
* boxxy: various fixes (part 2)
* boxxy: various fixes (part 3)
* boxxy: various fixes (part 4)
forgot to run ./format, whoops
* boxxy: use literalExpression for the rewrite example
Co-authored-by: Naïm Favier <n@monade.li>
* boxxy: add news entry
---------
Co-authored-by: Naïm Favier <n@monade.li>
* Revert "maintainers: add rasmus-kirk as a maintainer"
This reverts commit 301b364892.
* Revert "joshuto: add the joshuto file manager"
This reverts commit e7fdcb40b2.
* home-environment: add `home.sessionVariablesPackage`
Allow the `hm-session-vars.sh` derivation to be referenced from other
modules, e.g. to translate it to fish with babelfish at build time.
* fish: use babelfish for `hm-session-vars.sh`
Translate `hm-session-vars.sh` to fish at system build time,
significantly decreasing shell startup time.
Based on https://github.com/NixOS/nixpkgs/pull/108947 by @kevingriffin.
This is backwards compatible and allows for more flexibility(eg.
allows for defining custom waybar modules in separate nixos modules,
before merging them together)
Co-authored-by: mae <git@badat.dev>
Make use of the recently added nullable `mkPackageOption` feature
to disable installing an SSH client by default: most people should use
the client provided by their system.
Kitty provides integrations for bash, fish and zsh. The new
programs.kitty.shellIntegration options allow the configuration of
these integrations.
See <https://sw.kovidgoyal.net/kitty/shell-integration> for more details.
Signed-off-by: Sefa Eyeoglu <contact@scrumplex.net>
* zellij: adds options to integrate with zsh, bash and fish shells
* zellij: add tests for shell integration options
* zellij: eval setup auto start for fish integration
* zellij: use interactiveShellInit for fish integration
* zellij: fixes format issues
* zellij: enable shell integrations by default
* zellij: compresses shell integration test cases
* zellij: removes the disabled shell integration tests
* zellij: formats tests
* himalaya: add soywod to maintainers
* himalaya: make the config safer
Also added two services and more tests.
* himalaya: fix doc + typos
* himalaya: use freeform
* himalaya: run ./format
* himalaya: make use of mkPackageOption
It's pretty common to need multiple bindings to
history-substring-search, since different terminals will send different
keys for up/down.
This does not break back-compatibility, and introduces a new test
The current zplug nixpkgs puts everything under `$out/`. It pollutes the nix
profile dir.
This is a breaking change. It depends on an change of the output path in the
nixpkgs zplug package.
Virtual mailboxes (described by Notmuch queries) can now configured for each account in NeoMutt.
Plus, it is possible to disable Notmuch section for a specific account.
The added extraConfig option allowes users to insert custom text at
the end of the generated profile `user.js` file. This allows the users
to import templates as part of their configuration.
link the packpath in expected folder so that even unwrapped neovim can pick home-manager's plugins.
I sometimes need to run neovim not wrapped/configured by nix (when
developing neovim or when other projects bring their own neovim in
PATH). Currently they dont find plugins installed by home-manager in the
cases where packpath is not set to the generated nix packpath directory.
With this change, neovim can discover HM-installed plugins by itself.
Some of the email providers (like GMail and Fastmail) save Sent messages
automatically, so make the folders optional in the configuration.
Make Drafts folder optional as well, to allow it to be configured
manually in the extraConf with location outside of the maildir.
* i3status-rust: update it to handle 0.30.x releases
0.30.0 is a major release that brings many breaking changes to the
configuration file. See:
https://github.com/greshake/i3status-rust/blob/master/NEWS.md#i3status-rust-0300
The only one that actually affects the module though is the change in
how the theme/icons are defined. Other changes are mostly on how to
specify formatting/blocks, and since we just generate the TOML as-is, it
needs changes in the user side.
So most changes in this commit are documentation updates, having
up-to-date examples from things that changed, e.g.: the new `click`
attribute that now can be applied to any block.
* i3status-rust: only use new format if i3status-rust >= 0.30.0
* news: document the i3status-rust changes
* i3status-rust: add thiagokokada as maintainer
This command adds the ability to specify lists of qutebrowser
commands as values for key bindings, which avoids the need for
concatenating commands with ` ;; `.
* exa: add more options
* exa: use `escapeShellArgs`
* exa: don't hardcode executable path in aliases
Prevents aliases from going stale in open terminals when the system is updated.
* exa: use `command` for self-referential alias
Otherwise fish complains about the recursive call.
Drop the aliases from ion shell since it doesn't implement the POSIX
`command` built-in.
* exa: re-add ion aliases
* exa: drop `command`
Fish doesn't complain about recursion if `exa` isn't escaped.
---------
Co-authored-by: Naïm Favier <n@monade.li>
The `-X` prevents that screen is cleared when showing a diff that's
larger than my screen.
I.e. when running `git diff` and press `q`, the last thing I want to see
is the prompt with `git diff` and *not* the part of the diff I browsed,
to be clear
$ git diff
$ <cursor>
Considering that this is somewhat opinionated, I decided to build an
option which allows you to pass arbitrary commands to the less
invocation.
Xsession (and hence ~/.xsession) is executed in bash but does not set
SHELL to the full path to bash. In case the user's login shell is
something other than bash then SHELL is set to that shell. Keychain
inspects the SHELL variable to find out what shell it has to generate
code for, so in .xsession it generates code for the user's login shell
instead for bash.
This change forces SHELL to bash for keychain when invoked from
.xsession, the same way it's done when generating keychain's code for
bash integration.
Closes#3693
* vscode: add extensions.json file in extensions dir
This change generates an 'extensions.json` file the same way that
nixpkgs' vscode-with-extensions does, and makes sure it is placed in the
directory with the extensions.
* vscode: remove leftover trace
Co-authored-by: Naïm Favier <n@monade.li>
* vscode: fix adding extensions.json with mutable extension dir
Co-authored-by: Naïm Favier <n@monade.li>
* vscode: let vscode regenerate the mutable extensions.json
* Remove nixpkgs duplication; only apply on vscodes new enough to need it
* Use lib.versionAtLeast
Co-authored-by: Naïm Favier <n@monade.li>
* Format vscode.nix
---------
Co-authored-by: Naïm Favier <n@monade.li>
Internally we already managed them per-profile but exposed a global
option to maintain backwards compatibility. The benefit to having
per-profile extensions is quite large though, so it is time to switch.
Users of the global extensions option will get an error message that
indicates how to edit their configuration to work again.
Firefox internally only supports bool, int, and string types for
preferences, but often stores objects, arrays and floats as strings.
This change makes it nicer to specify those type of preferences in
Nix, and it also makes it possible to merge objects & arrays across
multiple modules.
Some JVMs pass through `home` as a derivation rather than as a string, as `openjdk` does. Since the module option for session variables expects a string, this is a type error. I suspect that this incorrect, and have changed the assignment here to coerce the `cfg.package.home` attribute to a string to be safe.
After discussing with @NobbZ, we have decided it is best to mitigate this problem in HM rather than to make potentially breaking changes to Nixpkgs.
Please do mention if you think we ought to propose a change to Nixpkgs instead.
Allow setting the application package and storePath used by the
config. Since the `programs.password-store` Home Manager module sets
config values via global environment variables, the default behavior
of the module should continue to behave as before for the user.
Additionally,
- Adds a few tests.
- Use "escapeShellArg" function call to the path parameter call to
ensure paths with spaces work.
- Allow not setting storePath, which will cause `pass_secret_service`
to default to using `~/.password-store`.
- If `pass-secret-service` is enabled, set its store path to default
to the one defined in our password-store environment settings.
- Add myself (houstdav000) as maintainer.
Fish shell doesn't require arguments to `eval` to be double quoted
like in a bash shell. At the moment doing so gives us the following
error:
~/.config/fish/config.fish (line 12): $(...) is not supported. In fish, please use '(/nix/store/8asq…)'.
eval "$(/nix/store/8asqgnhs89wzyjvs8p1n5hvxn7lkn9wa-opam-2.1.3/bin/opam env --shell=fish)"
^
from sourcing file ~/.config/fish/config.fish
called during startup
source: Error while reading file “/home/user/.config/fish/config.fish”
This commit fixes the above error.
The default value of `programs.ncmpcpp.mpdMusicDir` is taken from
`services.mpd.musicDirectory` if the mpd module is enabled, which has
type `either path str`. `programs.ncmpcpp.mpdMusicDir` did not accept
`str` values, though, so an error was raised when the default value was
used and `services.mpd.musicDirectory` was set to a value of type `str`.
This commit changes the type of `programs.ncmpcpp.mpdMusicDir` to also
accept `str` to reflect the type of `services.mpd.musicDirectory`.
Fixes#3560
Assigning to `programs.neovim.extraLuaPackages` a function taking a lua package set as input
and returning a list of packages, as described in the documentation,
threw an error because the rest of the code assumed that the value was always a plain list.
Using `lib.types.coercedTo`, we can accept such functions, as per the documentation,
as well as plain lists, which we then convert to a function ignoring its input argument.
We print a warning when a plain list is assigned, since the function
form is preferred, as it ensures that the right lua package set is used.
For the lua packages, we also get the lua package set from the
finalPackage, to make sure that we are always using the same package set
as the actual unwrapped neovim package being built.
For `programs.neovim.extraPythonPackages` I did the same.
I updated the test case so that we test both ways of setting these options.
This enables nushell integration by default for direnv, similar to
bash/zsh/fish. The slightly verbose way of setting this is to ensure
that peoples' existing nushell configuration isn't overwritten, only
appended to, as would be the case if we just used the integration
example from the nushell docs:
https://www.nushell.sh/cookbook/direnv.htmlCloses#3520
Previously the nushell module did not differentiate between Linux and
Darwin when deciding where to place config files, whereas nushell
does. This commit fixes that.
The previous variant used IFD to generate the `JAVA_HOME` variable and relied on internal hooks of the `java` package, this failed for a user cross compiling their configuration.
This PR changes that and uses the `home` attribute, as documented in the very last sentence of the https://nixos.org/manual/nixpkgs/stable/#sec-language-java chapter.
The previous version linked the file into home, then sourced that. Since
nothing else expects that file to be there, this is unnecessary.
Additionally, doing so made it impossible to test a built config without
switching, e.g. using `XDG_CONFIG_HOME=… nvim` or `nvim -u`. This
remedies that, at least for this particular reference.
To test this, change from asserting contents of the config file to
actually starting nvim, outputting sentinel values, and then asserting
their values are present. This way it’s tested that nvim loaded the
config, rather than that some config is in a specific place.
This is all in one commit as the test, as written now, would not have
worked before since the previously hard-coded home path was not an
actual file in the test environment.
* ssh: add generic Match support for matchBlocks
Introduce conservative support for actual `Match`
blocks in ssh config.
"Conservative" means this PR doesn'tt try to process
the `match` expression and simply uses it as a string
provided by the user.
If set, `match` has precedence over `host` meaning
if both are set, `match` is used and `host` is ignored.
* Add news entry
Rather than reject a configuration when this option is set, just
silently ignore it when the platform isn't darwin. The name makes it
obvious that it won't be applied outside of darwin, and this allows
people to use the same configuration between hosts without any special concern.
Co-authored-by: Nicholas Sielicki <git@opensource.nslick.com>
This commit allows imperative management of "urls" file. It can be
useful if "urls" file is treated as a secret.
With this change, it's possible to provision "urls" via Syncthing,
agenix, sops-nix or other means, while still managing Newsboat
declaratively.
Add a new Thunderbird module that uses the configuration in
`accounts.email.accounts` to setup SMTP and IMAP accounts.
Multiple profiles are not supported at this point.
Update notification popups are annoying when vscode/vscodium is
managed by Home Manager. However, as these settings also require the
configuration to be managed via `userSettings`, they are disabled by
default.
With this change, it's now possible to configure the default search
engine in Firefox with
programs.firefox.profiles.<name>.search.default
and add custom engines with
programs.firefox.profiles.<name>.search.engines.
It's also recommended to enable
programs.firefox.profiles.<name>.search.force = true
since Firefox will replace the symlink for the search configuration on
every launch, but note that you'll loose any existing configuration by
enabling this.
This will cache the output of `passwordCommand` per authentication
realm.
Context: the `credentials` key in `sbt` is a `TaskKey[Seq[Credentials]]`.
In `sbt`, tasks are evaluated on-demand and their output is not cached.
This particular key is referenced by all submodules in a project. When
the command is relatively expensive (e.g.: `pass show foo`), this
results in several seconds of delay when doing basic things like
`compile` or `test` which makes this unusable without some kind of
caching.
