* feat: Pijul VCS support
* Extra bits needed for new module.
* Format Markdown table.
* Fix lint.
* Don't test Pijul module so thoroughly.
Installing from source is too expensive, and compiled binaries are only
available for Windows (and unofficially as well). Perhaps once Pijul
1.0.0 comes out of beta there will be more binaries available in package
repos.
* Format!
* Bad rebase, remove Pijul install from workflow.
* Mock Pijul commands for code coverage.
* Make fake .pijul directory in fixture.
* Truly mock `pijul` command.
* Rename module from `pijul` to `pijul_channel`.
* Format!
* Fix config-schema.json.
* Missed changing module name in docs/ folder.
* feat(hg_branch): Add support for mercurial topics and find hg root dir
* Fix clippy errors
* Use crate::utils::read_file
* Update config-schema.json
* Extend PathExt to retrieve device ID of Path
* Break hg root search when switching to another device
* Fix clippy and formatting errors
* Update docs/config/README.md
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* Update src/modules/utils/path.rs
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* Update src/configs/hg_branch.rs
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* Update hg_branch description
* Revert to lazy loading, use truncate_text from utils and use fake topic
* Format code and fix clippy error
* Revert to previous test string as topic is optional in the config
* Fix doc formatting
* Stub device_id for windows
* Update config-schema.json
* Update src/modules/hg_branch.rs
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* Do not use unwrap in device_id
* Fix formatter error
* Use dev under non linux unixes
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* Support `nix shell`
* Remove unnecessary `Debug` implementation
* Add test to detect false positive
* Improve detection of `/nix/store` in $PATH
* Add docs about unknown state
* Gate under `heuristic` flag
* Regenerate config schema
Kubernetes module was previously a bit messy, with lots of
unnecessarily nested options. Clean this up using filter_map and
find_map, with two major results:
1. `ctx_components` is now a Vec<KubeCtxComponent> instead of a
Vec<Option<KubeCtxComponent>>. This greatly simplified downstream
processing of these context components.
2. Instead of storing a partial computation of the namespace in
variables `kube_ns`, etc, compute them directly in the formatter
mapping. This is made simpler (read: actually doable) by change 1.
* Add Haxe support
* avoid unwrap
* fix doc formatting
* removed extra newline
* fixed formatter and linter issues
* fixed config file
* better version of detecting contents of .haxerc
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* removed openfl related defaults from detect_files
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* fixed formatting
* reworked reading .haxerc with fallback to haxe --version
* fixed formatting
* added fallback to executable for dev paths in .haxerc
* fixed linter issue
* added support for Windows paths
* use or_else
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* use shorter version with `?`
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* simplified regex check
removed check for "null" string
* fixed format
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
Instead of the remote address of 8.8.8.8 (Google DNS) in the crate
local_ipaddress use a reserved IPv4 address, that should never be
assigned.
Also forward the underlying error on failure.
Supersedes: #4614
Disables the display of notifications from cmd_duration on Linux if
none of DISPLAY, WAYLAND_DISPLAY, or MIR_SOCKET are set.
This prevents starship from attempting to create notifications in tty
environments, which was previously causing hangs.
* add username to azure module config
* add username to azure module config
* formatting with cargo fmt
* Handle parse failure on azureProfile.json
allow program to procede if unable to parse azure profile due to missing
keys from the JSON structure.
remove unused keys from struct
Code cleanup with suggestions from PR maintainer
Cargo clippy fixes
* feat(bug-report): ask for confirmation before opening issue
* Apply suggestions from code review
Co-authored-by: Thomas O'Donnell <andytom@users.noreply.github.com>
Co-authored-by: Thomas O'Donnell <andytom@users.noreply.github.com>
* Fix #4481, config does not error when unrecognized properties are present
* cleanup: use stuct update syntax to improve readability
from review feedback
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* cleanup: renamed ValueDeserializer func w/ better name
* cleanup: added test to cover unknown key retry condition
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
This patch uses the which crate to resolve the starship path, replacing the
current_exe() mechanism.
Co-authored-by: Kevin Song <chips@ksong.dev>
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
Co-authored-by: Dario Vladović <d.vladimyr@gmail.com>
* feat: added showing gradle version based on the gradle.properties file
* fix: wouldn't return version
* fix: forgot to remove "version=" from returned version"
* fix: ran rustfmt
* fix: test now actually tests for something
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* fix: the regex actually makes sense now
* fix: complete refactor of control flow
* Delete flake.nix
* changed order in which files are processed
Co-authored-by: BattleCh1cken <BattleCh1cken@Larkov.de>
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* feat(guix_shell): Initial implementation (#3999)
* fix(guix_shell): Change guix nerd font icon to water buffalo emoji
* fix(guix_shell): Added guix_shell entries in preset files
* fix(guix_shell): Moved guix_shell config docs in to the correct place (alphabetically)
* feat(aws): add a fallback for `expiration`
* fix(aws): intermittent test failures
- extend the time range from `-2s,0s` to `-5s,+2s`
* fix: `docs/config/README.md` readability
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* feat: Open Policy Agent module (#1740)
* Format documentation
* Fix typo, `ropa` -> `rego`
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* Update presets for OPA module
* Add extra space to OPA module symbol
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* feat(module): Add a meson devenv indicator
Adds a Meson Developer Environment indicator, if the MESON_DEVENV
variable is set. Inside a `meson devenv`, the prompt will include the
current Meson project name
This also contains a new Truncate utility function, which may be adapted for other modules in the future
* docs: Add Meson to presets
* Support formatting of pipestatus separator
* Format pipestatus separator with each pipestatus
* Add third exit code to pipestatus test
* Clean up pipestatus mapping
* Add comment that was removed
* docs(config): add color palette to docs
* feat: add user-defined color palette
* fix: update config schema
* refactor: apply suggestions from code review
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* fix: update new test
* feat: add support for multiple palettes
* docs(config): update docs for multiple color palettes
* docs(config): fix formatting
* test: test overriding a predefined color with itself
* docs: mention palettes cannot reference themselves
* refactor: warn when using a nonexistent palette
* test: test retrieving a nonexistent color palette
* fix: fix issues with palette log messages
* fix: update config schema
* fix: skip serializing palette if none
* refactor: change nonexistent palette message to warning
* fix: update config schema
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* upgrade `gitoxide` to v0.21
This release comes with lenient configuration handling by default,
allowing to open repositories even their configuration values are
invalid (even for git), as long as there are viable defaults.
Furthermore this release adds the ability to open submodule repsitories.
Fixes https://github.com/starship/starship/issues/4266 and
fixes https://github.com/starship/starship/issues/4272
* Assure an object cache is set to speed up `commit.describe()`
Related to https://github.com/starship/starship/issues/4275 bringing
performance to spitting distance compared to git.
* Add starship preset command
* Use ValueEnum for preset command
* Generate ValueEnum struct in build.rs
* Use absolute paths and refactor codegen
* Use dunce to canonicalize path
* Use raw string literal in include_bytes!
* Use .cloned()
* Apply fixes
* Fix path escaping
* Removed error message if stdout is unavailable
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
* test that we can match a multi-part file extension such as in foo.tar.gz
* now we can match multi-part file extensions like on foo.tar.gz
* add a test that a !ext is a negative match and over-rides any positive match
* test that negative extensions that don't match any file have no effect
* fail the match if any negative extensions exist
* cargo fmt
I'm not happy with this, in particular it's made the structures of has_any_positive_extension and has_no_negative_extension look different, and the logic in is_match is harder to follow
* placate clippy
* documentation for multi-part extensions and negative extensions
* get rid of an unnecessary .to_string() and comment the necessary but weird-looking invocations of .to_string_lossy().to_string()
* tests for negative matching of files and folders
* fail the match is any negative files/folders match
* document file/folder negative matching; be less prolix
* suppress Nodejs if Deno files are present (#2627)
* Revert "suppress Nodejs if Deno files are present (#2627)"
This reverts commit c1394fd7b37bb0bf06b1449e074020a2e16bfa04.
This was a terrible way of doing this, there's got to be a better way!
Have added configuration options to the k8s module to allow activating
the module only in directories that contains certain files or folders.
To ensure this is backwards compatible and because there are not really
any standard files or folders for Kubernetes I have set the defaults to
empty and will activate the module for all directories.
Have switched all vi/vim symbols to have the same prefix 'vim'. To
preserve backwards compatibility with existing configs I have added an
alias for the previous config name.
* add proper vi mode detection for fish shell
* update tests
* fix test
* update config-schema.json
* update docs
* add warning about symbols only supported in fish
* check for go.work file to display go version
* add test to check for go.work file
* update docs to include go.work file
* chore(dprint): fmt & upgrade plugins (#3969)
Co-authored-by: David Knaack <davidkna@users.noreply.github.com>
This is an actualization of PR #559 as originally envisioned by qryxip.
Adds the ability to display toolchain versions, either as extracted from
environment/settings files or by getting the host triple. As part of
this, several other major changes were needed:
- Many of the smaller functions within the code have been fused, moved,
or dropped.
- The Rustup environmental info is now initialized lazily using
OnceCells. This will hopefully lead to a performance increase.
- New configuration variables (`toolchain` and `numver`) have been added
to allow finer-grained configuration.
- Override information is no longer read from `rustup` output. Instead,
it is parsed from the same files that rustup would use to determine
this info.
Co-authored-by: qryxip <qryxip@gmail.com>
Co-authored-by: qryxip <qryxip@gmail.com>
* feat(package): Extract package version from PEP621 pyproject.toml
* Update docs explaining PEP 621 package version
* Only read pyproject.toml once
* Simplify get_pep621_version
* Handle version formatting in get_pyproject_version
* fix: Do not panic in config if editor not found
* Add tests for edit_configuration
Adds tests for no-panic condition on editor by adding an override to
edit_configuration.
* Sorry clippy :(
* perf(package): only try to read files that exist
Have refactored the package module to improve performance. Before this
change the module would try to open every single file that could contain
some package information until it found a valid version. This resulted
in a lot of unneeded disk IO. Have added a new fn, `read_file_from_pwd`
that uses the current context to check if that file already exists and
fast failing if it doesn't. From my local testing this speeds up the
package module from taking ~1ms to ~50µs in an empty directory.
* refactor: move read_file_from_pwd to context
* refactor(haskell): use read_files_from_pwd
* refactor(nodejs): use read_files_from_pwd
