toolkit.nu in nushell repo root problems #1
Comments
|
Hmmm, this could be because it's not finding the config file or something like that? Seems like it's ignoring my $env.TOPIARY_blah settings. 🤔 |
|
I think it's finding my config now, but I have no idea what's going on here and why it things the languages.ncl is invalid. It looks good to me. |
|
I think your topiary version is out of date, maybe try to install the github version with cargo? |
That was definitely the case. Now it just doesn't compile on Windows. |
|
Just found out that latest topiary supports referring grammar to local path like this: grammar = {
# source.git = {
# git = "https://github.com/nushell/tree-sitter-nu.git",
# rev = "7e0f16f608a9e804fae61430ade734f9f849fb80",
# },
source.path = "<local-directory>/libtree-sitter-nu.dylib",
symbol = "tree_sitter_nu",
},Maybe you can try manually compiling the parser with Remember to checkout the commit above since the latest grammar is not well-supported by this project (still struggling with annoying whitespaces issues). |
|
Interesting. I was able to get it to do something at least.
I did all that on the hash you specified above. On the latest main I get this: |
|
You need to run tree-sitter generate before the build, otherwise the grammar is still the latest. Make doesn't require that because it directly use the generated c files, which is git tracked. Actually I'm not sure how tree-sitter build works, but the error message definitely suggests that the built parser is newer than that commit point, maybe even need to cleanup something before tree-sitter generate. |
ifeq ($(OS),Windows_NT)
$(error Windows is not supported)
endifAha, so the makefile is designed to not work on windows, interesting. |
Ya, I noticed that too.
I tried that and it didn't make a difference. ❯ tree-sitter generate
❯ rm nu.dll
❯ tree-sitter build
❯ z nush
toolkit module detected...
activating toolkit module with `use toolkit.nu`
❯ open toolkit.nu | topiary format --language nushell --verbose --tolerate-parsing-errors
[2024-12-03T18:07:44Z ERROR topiary] Error parsing query file
[2024-12-03T18:07:44Z ERROR topiary] Cause: QueryError { row: 4, column: 7, offset: 65, message: "long_flag_equals_value", kind: NodeType } |
|
Yeah I realized my guess was wrong, it doesn't make any sense to me, if tree-sitter-nu is at the correct head, it should contain the rule for long-flag-equals-value (removed recently by me). |
|
@fdncred I think this is now compatible with the latest parser, if you have the I've tested all Lines 5 to 34 in 68572e8 |
|
@blindFS I wonder what I'm doing wrong.
❯ $env.TOPIARY_CONFIG_FILE
C:/Users/fdncred/source/repos/topiary-nushell/languages.ncl
❯ $env.TOPIARY_LANGUAGE_DIR
C:/Users/fdncred/source/repos/topiary-nushell/languages❯ open toolkit.nu | topiary format --language nushell --verbose --tolerate-parsing-errors
[2024-12-20T14:43:28Z ERROR topiary] Error parsing query file
[2024-12-20T14:43:28Z ERROR topiary] Cause: QueryError { row: 4, column: 7, offset: 65, message: "long_flag_equals_value", kind: NodeType }
❯ topiary format toolkit.nu
[2024-12-20T14:44:22Z ERROR topiary] Error parsing query file
[2024-12-20T14:44:22Z ERROR topiary] Cause: QueryError { row: 4, column: 7, offset: 65, message: "long_flag_equals_value", kind: NodeType } |
|
@fdncred This topiary-nushell on main branch with latest commit? {
languages = {
nu = {
extensions = ["nu"],
grammar.source.path = ".../tree-sitter-nu/nu.dylib",
},
},
}
|
|
I think the language name on the main branch has changed to |
|
I didn't have the topiary-nushell latest but I do now. This is what I get This is my languages.ncl {
languages = {
nu = {
extensions = ["nu"],
grammar.source.path = "C:/Users/fdncred/source/repos/tree-sitter-nu/nu.dll",
symbol = "tree_sitter_nu",
},
},
}It tried to do something and |
|
@fdncred I think it's working, except for the write access. I'll ask the owners of topiary to see what's going on with the write accessibility. |
|
yup, you're right. yay! it's interesting to see what it changed. 99.9% is fine but the section with diff --git a/toolkit.nu b/toolkit.nu
index d848dce18..6096879a5 100644
--- a/toolkit.nu
+++ b/toolkit.nu
@@ -8,108 +8,108 @@
# check standard code formatting and apply the changes
export def fmt [
- --check # do not apply the format changes, only check the syntax
- --verbose # print extra information about the command's progress
+ --check # do not apply the format changes, only check the syntax
+ --verbose # print extra information about the command's progress
] {
- if $verbose {
- print $"running ('toolkit fmt' | pretty-format-command)"
- }
+ if $verbose {
+ print $"running ('toolkit fmt' | pretty-format-command)"
+ }
- if $check {
- try {
- ^cargo fmt --all -- --check
- } catch {
- error make --unspanned {
- msg: $"\nplease run ('toolkit fmt' | pretty-format-command) to fix formatting!"
- }
- }
- } else {
- ^cargo fmt --all
+ if $check {
+ try {
+ ^cargo fmt --all -- --check
+ } catch {
+ error make --unspanned {
+ msg: $"\nplease run ('toolkit fmt' | pretty-format-command) to fix formatting!"
+ }
}
+ } else {
+ ^cargo fmt --all
+ }
}
# check that you're using the standard code style
#
# > it is important to make `clippy` happy :relieved:
export def clippy [
- --verbose # print extra information about the command's progress
- --features: list<string> # the list of features to run *Clippy* on
+ --verbose # print extra information about the command's progress
+ --features: list<string> # the list of features to run *Clippy* on
] {
- if $verbose {
- print $"running ('toolkit clippy' | pretty-format-command)"
- }
+ if $verbose {
+ print $"running ('toolkit clippy' | pretty-format-command)"
+ }
- # If changing these settings also change CI settings in .github/workflows/ci.yml
- try {(
- ^cargo clippy
- --workspace
- --exclude nu_plugin_*
- --features ($features | str join ",")
- --
- -D warnings
- -D clippy::unwrap_used
- -D clippy::unchecked_duration_subtraction
+ # If changing these settings also change CI settings in .github/workflows/ci.yml
+ try {
+ (
+ ^cargo clippy
+ --workspace
+ --exclude nu_plugin_*
+ --features ($features | str join ",")
+ --
+ -D warnings
+ -D clippy::unwrap_used
+ -D clippy::unchecked_duration_subtraction
)
if $verbose {
- print $"running ('toolkit clippy' | pretty-format-command) on tests"
+ print $"running ('toolkit clippy' | pretty-format-command) on tests"
}
# In tests we don't have to deny unwrap
(
- ^cargo clippy
- --tests
- --workspace
- --exclude nu_plugin_*
- --features ($features | str join ",")
- --
- -D warnings
+ ^cargo clippy
+ --tests
+ --workspace
+ --exclude nu_plugin_*
+ --features ($features | str join ",")
+ --
+ -D warnings
)
if $verbose {
- print $"running ('toolkit clippy' | pretty-format-command) on plugins"
+ print $"running ('toolkit clippy' | pretty-format-command) on plugins"
}
(
- ^cargo clippy
- --package nu_plugin_*
- --
- -D warnings
- -D clippy::unwrap_used
- -D clippy::unchecked_duration_subtraction
+ ^cargo clippy
+ --package nu_plugin_*
+ --
+ -D warnings
+ -D clippy::unwrap_used
+ -D clippy::unchecked_duration_subtraction
)
-
- } catch {
- error make --unspanned {
- msg: $"\nplease fix the above ('clippy' | pretty-format-command) errors before continuing!"
- }
+ } catch {
+ error make --unspanned {
+ msg: $"\nplease fix the above ('clippy' | pretty-format-command) errors before continuing!"
}
+ }
}
# check that all the tests pass
export def test [
- --fast # use the "nextext" `cargo` subcommand to speed up the tests (see [`cargo-nextest`](https://nexte.st/) and [`nextest-rs/nextest`](https://github.com/nextest-rs/nextest))
- --features: list<string> # the list of features to run the tests on
- --workspace # run the *Clippy* command on the whole workspace (overrides `--features`)
+ --fast # use the "nextext" `cargo` subcommand to speed up the tests (see [`cargo-nextest`](https://nexte.st/) and [`nextest-rs/nextest`](https://github.com/nextest-rs/nextest))
+ --features: list<string> # the list of features to run the tests on
+ --workspace # run the *Clippy* command on the whole workspace (overrides `--features`)
] {
- if $fast {
- if $workspace {
- ^cargo nextest run --all
- } else {
- ^cargo nextest run --features ($features | str join ",")
- }
+ if $fast {
+ if $workspace {
+ ^cargo nextest run --all
+ } else {
+ ^cargo nextest run --features ($features | str join ",")
+ }
+ } else {
+ if $workspace {
+ ^cargo test --workspace
} else {
- if $workspace {
- ^cargo test --workspace
- } else {
- ^cargo test --features ($features | str join ",")
- }
+ ^cargo test --features ($features | str join ",")
}
+ }
}
# run the tests for the standard library
export def "test stdlib" [
- --extra-args: string = ''
+ --extra-args: string= ''
] {
- ^cargo run -- --no-config-file -c $"
+ ^cargo run -- --no-config-file -c $"
use crates/nu-std/testing.nu
testing run-tests --path crates/nu-std ($extra_args)
"
@@ -117,7 +117,7 @@ export def "test stdlib" [
# formats the pipe input inside backticks, dimmed and italic, as a pretty command
def pretty-format-command [] {
- $"`(ansi default_dimmed)(ansi default_italic)($in)(ansi reset)`"
+ $"`(ansi default_dimmed)(ansi default_italic)($in)(ansi reset)`"
}
# return a report about the check stage
@@ -130,36 +130,31 @@ def pretty-format-command [] {
# otherwise, the truth values will be incremental, following
# the order above.
def report [
- --fail-fmt
- --fail-clippy
- --fail-test
- --fail-test-stdlib
- --no-fail
+ --fail-fmt
+ --fail-clippy
+ --fail-test
+ --fail-test-stdlib
+ --no-fail
] {
- [fmt clippy test "test stdlib"]
- | wrap stage
- | merge (
- if $no_fail { [true true true true] }
- else if $fail_fmt { [false null null null] }
- else if $fail_clippy { [true false null null] }
- else if $fail_test { [true true false null] }
- else if $fail_test_stdlib { [true true true false] }
- else { [null null null null] }
- | wrap success
- )
- | upsert emoji {|it|
- if ($it.success == null) {
- ":black_circle:"
- } else if $it.success {
- ":green_circle:"
- } else {
- ":red_circle:"
- }
- }
- | each {|it|
- $"- ($it.emoji) `toolkit ($it.stage)`"
+ [fmt clippy test "test stdlib"]
+ | wrap stage
+ | merge (
+ if $no_fail {[true true true true]} else if $fail_fmt {[false null null null]} else if $fail_clippy {[true false null null]} else if $fail_test {[true true false null]} else if $fail_test_stdlib {[true true true false]} else {[null null null null]}
+ | wrap success
+ )
+ | upsert emoji {|it|
+ if ($it.success == null) {
+ ":black_circle:"
+ } else if $it.success {
+ ":green_circle:"
+ } else {
+ ":red_circle:"
}
- | to text
+ }
+ | each {|it|
+ $"- ($it.emoji) `toolkit ($it.stage)`"
+ }
+ | to text
}
# run all the necessary checks and tests to submit a perfect PR
@@ -253,242 +248,243 @@ def report [
#
# now the whole `toolkit check pr` passes! :tada:
export def "check pr" [
- --fast # use the "nextext" `cargo` subcommand to speed up the tests (see [`cargo-nextest`](https://nexte.st/) and [`nextest-rs/nextest`](https://github.com/nextest-rs/nextest))
- --features: list<string> # the list of features to check the current PR on
+ --fast # use the "nextext" `cargo` subcommand to speed up the tests (see [`cargo-nextest`](https://nexte.st/) and [`nextest-rs/nextest`](https://github.com/nextest-rs/nextest))
+ --features: list<string> # the list of features to check the current PR on
] {
- $env.NU_TEST_LOCALE_OVERRIDE = 'en_US.utf8'
- $env.LANG = 'en_US.UTF-8'
- $env.LANGUAGE = 'en'
-
- try {
- fmt --check --verbose
- } catch {
- return (report --fail-fmt)
- }
-
- try {
- clippy --features $features --verbose
- } catch {
- return (report --fail-clippy)
- }
-
- print $"running ('toolkit test' | pretty-format-command)"
- try {
- if $fast {
- if ($features | is-empty) {
- test --workspace --fast
- } else {
- test --features $features --fast
- }
- } else {
- if ($features | is-empty) {
- test --workspace
- } else {
- test --features $features
- }
- }
- } catch {
- return (report --fail-test)
- }
-
- print $"running ('toolkit test stdlib' | pretty-format-command)"
- try {
- test stdlib
- } catch {
- return (report --fail-test-stdlib)
+ $env.NU_TEST_LOCALE_OVERRIDE = 'en_US.utf8'
+ $env.LANG = 'en_US.UTF-8'
+ $env.LANGUAGE = 'en'
+
+ try {
+ fmt --check --verbose
+ } catch {
+ return (report --fail-fmt)
+ }
+
+ try {
+ clippy --features $features --verbose
+ } catch {
+ return (report --fail-clippy)
+ }
+
+ print $"running ('toolkit test' | pretty-format-command)"
+ try {
+ if $fast {
+ if ($features | is-empty) {
+ test --workspace --fast
+ } else {
+ test --features $features --fast
+ }
+ } else {
+ if ($features | is-empty) {
+ test --workspace
+ } else {
+ test --features $features
+ }
}
-
- report --no-fail
+ } catch {
+ return (report --fail-test)
+ }
+
+ print $"running ('toolkit test stdlib' | pretty-format-command)"
+ try {
+ test stdlib
+ } catch {
+ return (report --fail-test-stdlib)
+ }
+
+ report --no-fail
}
# run Nushell from source with a right indicator
export def run [] {
- ^cargo run -- ...[
- -e "$env.PROMPT_COMMAND_RIGHT = $'(ansi magenta_reverse)trying Nushell inside Cargo(ansi reset)'"
- ]
+ ^cargo run -- ...[
+ -e
+ "$env.PROMPT_COMMAND_RIGHT = $'(ansi magenta_reverse)trying Nushell inside Cargo(ansi reset)'"
+ ]
}
# set up git hooks to run:
# - `toolkit fmt --check --verbose` on `git commit`
# - `toolkit fmt --check --verbose` and `toolkit clippy --verbose` on `git push`
export def setup-git-hooks [] {
- print "This command will change your local git configuration and hence modify your development workflow. Are you sure you want to continue? [y]"
- if (input) == "y" {
- print $"running ('toolkit setup-git-hooks' | pretty-format-command)"
- git config --local core.hooksPath .githooks
- } else {
- print $"aborting ('toolkit setup-git-hooks' | pretty-format-command)"
- }
+ print "This command will change your local git configuration and hence modify your development workflow. Are you sure you want to continue? [y]"
+ if (input) == "y" {
+ print $"running ('toolkit setup-git-hooks' | pretty-format-command)"
+ git config --local core.hooksPath .githooks
+ } else {
+ print $"aborting ('toolkit setup-git-hooks' | pretty-format-command)"
+ }
}
def build-nushell [features: string] {
- print $'(char nl)Building nushell'
- print '----------------------------'
+ print $'(char nl)Building nushell'
+ print '----------------------------'
- ^cargo build --features $features --locked
+ ^cargo build --features $features --locked
}
def build-plugin [] {
- let plugin = $in
+ let plugin = $in
- print $'(char nl)Building ($plugin)'
- print '----------------------------'
+ print $'(char nl)Building ($plugin)'
+ print '----------------------------'
- cd $"crates/($plugin)"
- ^cargo build
+ cd $"crates/($plugin)"
+ ^cargo build
}
# build Nushell and plugins with some features
export def build [
- ...features: string@"nu-complete list features" # a space-separated list of feature to install with Nushell
- --all # build all plugins with Nushell
+ ...features: string@"nu-complete list features" # a space-separated list of feature to install with Nushell
+ --all # build all plugins with Nushell
] {
- build-nushell ($features | str join ",")
-
- if not $all {
- return
- }
-
- let plugins = [
- nu_plugin_inc,
- nu_plugin_gstat,
- nu_plugin_query,
- nu_plugin_example,
- nu_plugin_custom_values,
- nu_plugin_formats,
- ]
-
- for plugin in $plugins {
- $plugin | build-plugin
- }
+ build-nushell ($features | str join ",")
+
+ if not $all {
+ return
+ }
+
+ let plugins = [
+ nu_plugin_inc
+ nu_plugin_gstat
+ nu_plugin_query
+ nu_plugin_example
+ nu_plugin_custom_values
+ nu_plugin_formats
+ ]
+
+ for plugin in $plugins {
+ $plugin | build-plugin
+ }
}
# build crates for wasm
export def "build wasm" [] {
- ^rustup target add wasm32-unknown-unknown
-
- # these crates should compile for wasm
- let compatible_crates = [
- "nu-cmd-base",
- "nu-cmd-extra",
- "nu-cmd-lang",
- "nu-color-config",
- "nu-command",
- "nu-derive-value",
- "nu-engine",
- "nu-glob",
- "nu-json",
- "nu-parser",
- "nu-path",
- "nu-pretty-hex",
- "nu-protocol",
- "nu-std",
- "nu-system",
- "nu-table",
- "nu-term-grid",
- "nu-utils",
- "nuon"
- ]
-
- for crate in $compatible_crates {
- print $'(char nl)Building ($crate) for wasm'
- print '----------------------------'
- ^cargo build -p $crate --target wasm32-unknown-unknown --no-default-features
- }
+ ^rustup target add wasm32-unknown-unknown
+
+ # these crates should compile for wasm
+ let compatible_crates = [
+ "nu-cmd-base"
+ "nu-cmd-extra"
+ "nu-cmd-lang"
+ "nu-color-config"
+ "nu-command"
+ "nu-derive-value"
+ "nu-engine"
+ "nu-glob"
+ "nu-json"
+ "nu-parser"
+ "nu-path"
+ "nu-pretty-hex"
+ "nu-protocol"
+ "nu-std"
+ "nu-system"
+ "nu-table"
+ "nu-term-grid"
+ "nu-utils"
+ "nuon"
+ ]
+
+ for crate in $compatible_crates {
+ print $'(char nl)Building ($crate) for wasm'
+ print '----------------------------'
+ ^cargo build -p $crate --target wasm32-unknown-unknown --no-default-features
+ }
}
def "nu-complete list features" [] {
- open Cargo.toml | get features | transpose feature dependencies | get feature
+ open Cargo.toml | get features | transpose feature dependencies | get feature
}
def install-plugin [] {
- let plugin = $in
+ let plugin = $in
- print $'(char nl)Installing ($plugin)'
- print '----------------------------'
+ print $'(char nl)Installing ($plugin)'
+ print '----------------------------'
- ^cargo install --path $"crates/($plugin)"
+ ^cargo install --path $"crates/($plugin)"
}
# install Nushell and features you want
export def install [
- ...features: string@"nu-complete list features" # a space-separated list of feature to install with Nushell
- --all # install all plugins with Nushell
+ ...features: string@"nu-complete list features" # a space-separated list of feature to install with Nushell
+ --all # install all plugins with Nushell
] {
- touch crates/nu-cmd-lang/build.rs # needed to make sure `version` has the correct `commit_hash`
- ^cargo install --path . --features ($features | str join ",") --locked --force
- if not $all {
- return
- }
-
- let plugins = [
- nu_plugin_inc,
- nu_plugin_gstat,
- nu_plugin_query,
- nu_plugin_example,
- nu_plugin_custom_values,
- nu_plugin_formats,
- ]
-
- for plugin in $plugins {
- $plugin | install-plugin
- }
+ touch crates/nu-cmd-lang/build.rs # needed to make sure `version` has the correct `commit_hash`
+ ^cargo install --path . --features ($features | str join ",") --locked --force
+ if not $all {
+ return
+ }
+
+ let plugins = [
+ nu_plugin_inc
+ nu_plugin_gstat
+ nu_plugin_query
+ nu_plugin_example
+ nu_plugin_custom_values
+ nu_plugin_formats
+ ]
+
+ for plugin in $plugins {
+ $plugin | install-plugin
+ }
}
def windows? [] {
- $nu.os-info.name == windows
+ $nu.os-info.name == windows
}
# filter out files that end in .d
def keep-plugin-executables [] {
- if (windows?) { where name ends-with '.exe' } else { where name !~ '\.d' }
+ if (windows?) { where name ends-with '.exe'} else { where name !~ '\.d'}
}
# add all installed plugins
export def "add plugins" [] {
- let plugin_path = (which nu | get path.0 | path dirname)
- let plugins = (ls $plugin_path | where name =~ nu_plugin | keep-plugin-executables | get name)
+ let plugin_path = (which nu | get path.0 | path dirname)
+ let plugins = (ls $plugin_path | where name =~ nu_plugin | keep-plugin-executables | get name)
- if ($plugins | is-empty) {
- print $"no plugins found in ($plugin_path)..."
- return
- }
+ if ($plugins | is-empty) {
+ print $"no plugins found in ($plugin_path)..."
+ return
+ }
- for plugin in $plugins {
- try {
- print $"> plugin add ($plugin)"
- plugin add $plugin
- } catch { |err|
- print -e $"(ansi rb)Failed to add ($plugin):\n($err.msg)(ansi reset)"
- }
+ for plugin in $plugins {
+ try {
+ print $"> plugin add ($plugin)"
+ plugin add $plugin
+ } catch {|err|
+ print -e $"(ansi rb)Failed to add ($plugin):\n($err.msg)(ansi reset)"
}
+ }
- print $"\n(ansi gb)plugins registered, please restart nushell(ansi reset)"
+ print $"\n(ansi gb)plugins registered, please restart nushell(ansi reset)"
}
def compute-coverage [] {
- print "Setting up environment variables for coverage"
- # Enable LLVM coverage tracking through environment variables
- # show env outputs .ini/.toml style description of the variables
- # In order to use from toml, we need to make sure our string literals are single quoted
- # This is especially important when running on Windows since "C:\blah" is treated as an escape
- ^cargo llvm-cov show-env | str replace (char dq) (char sq) -a | from toml | load-env
-
- print "Cleaning up coverage data"
- ^cargo llvm-cov clean --workspace
-
- print "Building with workspace and profile=ci"
- # Apparently we need to explicitly build the necessary parts
- # using the `--profile=ci` is basically `debug` build with unnecessary symbols stripped
- # leads to smaller binaries and potential savings when compiling and running
- ^cargo build --workspace --profile=ci
-
- print "Running tests with --workspace and profile=ci"
- ^cargo test --workspace --profile=ci
-
- # You need to provide the used profile to find the raw data
- print "Generating coverage report as lcov.info"
- ^cargo llvm-cov report --lcov --output-path lcov.info --profile=ci
+ print "Setting up environment variables for coverage"
+ # Enable LLVM coverage tracking through environment variables
+ # show env outputs .ini/.toml style description of the variables
+ # In order to use from toml, we need to make sure our string literals are single quoted
+ # This is especially important when running on Windows since "C:\blah" is treated as an escape
+ ^cargo llvm-cov show-env | str replace (char dq) (char sq) -a | from toml | load-env
+
+ print "Cleaning up coverage data"
+ ^cargo llvm-cov clean --workspace
+
+ print "Building with workspace and profile=ci"
+ # Apparently we need to explicitly build the necessary parts
+ # using the `--profile=ci` is basically `debug` build with unnecessary symbols stripped
+ # leads to smaller binaries and potential savings when compiling and running
+ ^cargo build --workspace --profile=ci
+
+ print "Running tests with --workspace and profile=ci"
+ ^cargo test --workspace --profile=ci
+
+ # You need to provide the used profile to find the raw data
+ print "Generating coverage report as lcov.info"
+ ^cargo llvm-cov report --lcov --output-path lcov.info --profile=ci
}
# Script to generate coverage locally
@@ -516,13 +512,13 @@ def compute-coverage [] {
# - https://github.com/umaumax/vim-lcov
# - https://github.com/andythigpen/nvim-coverage (probably needs some additional config)
export def cov [] {
- let start = (date now)
- $env.NUSHELL_CARGO_PROFILE = "ci"
+ let start = (date now)
+ $env.NUSHELL_CARGO_PROFILE = "ci"
- compute-coverage
+ compute-coverage
- let end = (date now)
- print $"Coverage generation took ($end - $start)."
+ let end = (date now)
+ print $"Coverage generation took ($end - $start)."
}
# Benchmark a target revision (default: current branch) against a reference revision (default: main branch)
@@ -530,37 +526,37 @@ export def cov [] {
# Results are saved in a `./tango` directory
# Ensure you have `cargo-export` installed to generate separate artifacts for each branch.
export def benchmark-compare [
- target?: string # which branch to compare (default: current branch)
- reference?: string # the reference to compare against (default: main branch)
+ target?: string # which branch to compare (default: current branch)
+ reference?: string # the reference to compare against (default: main branch)
] {
- let reference = $reference | default "main"
- let current = git branch --show-current
- let target = $target | default $current
-
- print $'-- Benchmarking ($target) against ($reference)'
-
- let export_dir = $env.PWD | path join "tango"
- let ref_bin_dir = $export_dir | path join bin $reference
- let tgt_bin_dir = $export_dir | path join bin $target
-
- # benchmark the target revision
- print $'-- Running benchmarks for ($target)'
- git checkout $target
- ^cargo export $tgt_bin_dir -- bench
-
- # benchmark the comparison reference revision
- print $'-- Running benchmarks for ($reference)'
- git checkout $reference
- ^cargo export $ref_bin_dir -- bench
-
- # return back to the whatever revision before benchmarking
- print '-- Done'
- git checkout $current
-
- # report results
- let reference_bin = $ref_bin_dir | path join benchmarks
- let target_bin = $tgt_bin_dir | path join benchmarks
- ^$target_bin compare $reference_bin -o -s 50 --dump ($export_dir | path join samples)
+ let reference = $reference | default "main"
+ let current = git branch --show-current
+ let target = $target | default $current
+
+ print $'-- Benchmarking ($target) against ($reference)'
+
+ let export_dir = $env.PWD | path join "tango"
+ let ref_bin_dir = $export_dir | path join bin $reference
+ let tgt_bin_dir = $export_dir | path join bin $target
+
+ # benchmark the target revision
+ print $'-- Running benchmarks for ($target)'
+ git checkout $target
+ ^cargo export $tgt_bin_dir -- bench
+
+ # benchmark the comparison reference revision
+ print $'-- Running benchmarks for ($reference)'
+ git checkout $reference
+ ^cargo export $ref_bin_dir -- bench
+
+ # return back to the whatever revision before benchmarking
+ print '-- Done'
+ git checkout $current
+
+ # report results
+ let reference_bin = $ref_bin_dir | path join benchmarks
+ let target_bin = $tgt_bin_dir | path join benchmarks
+ ^$target_bin compare $reference_bin -o -s 50 --dump ($export_dir | path join samples)
}
# Benchmark the current branch and logs the result in `./tango/samples`
@@ -568,30 +564,30 @@ export def benchmark-compare [
# Results are saved in a `./tango` directory
# Ensure you have `cargo-export` installed to generate separate artifacts for each branch.
export def benchmark-log [
- target?: string # which branch to compare (default: current branch)
+ target?: string # which branch to compare (default: current branch)
] {
- let current = git branch --show-current
- let target = $target | default $current
- print $'-- Benchmarking ($target)'
+ let current = git branch --show-current
+ let target = $target | default $current
+ print $'-- Benchmarking ($target)'
- let export_dir = $env.PWD | path join "tango"
- let bin_dir = ($export_dir | path join bin $target)
+ let export_dir = $env.PWD | path join "tango"
+ let bin_dir = ($export_dir | path join bin $target)
- # benchmark the target revision
- if $target != $current {
- git checkout $target
- }
- ^cargo export $bin_dir -- bench
+ # benchmark the target revision
+ if $target != $current {
+ git checkout $target
+ }
+ ^cargo export $bin_dir -- bench
- # return back to the whatever revision before benchmarking
- print '-- Done'
- if $target != $current {
- git checkout $current
- }
+ # return back to the whatever revision before benchmarking
+ print '-- Done'
+ if $target != $current {
+ git checkout $current
+ }
- # report results
- let bench_bin = ($bin_dir | path join benchmarks)
- ^$bench_bin compare -o -s 50 --dump ($export_dir | path join samples)
+ # report results
+ let bench_bin = ($bin_dir | path join benchmarks)
+ ^$bench_bin compare -o -s 50 --dump ($export_dir | path join samples)
}
# Build all Windows archives and MSIs for release manually
@@ -601,28 +597,28 @@ export def benchmark-log [
# You need to have the cross-compilers for MSVC installed (see Visual Studio).
# If compiling on x86_64, you need ARM64 compilers and libs too, and vice versa.
export def 'release-pkg windows' [
- --artifacts-dir="artifacts" # Where to copy the final msi and zip files to
+ --artifacts-dir= "artifacts" # Where to copy the final msi and zip files to
] {
- $env.RUSTFLAGS = ""
- $env.CARGO_TARGET_DIR = ""
- hide-env RUSTFLAGS
- hide-env CARGO_TARGET_DIR
- $env.OS = "windows-latest"
- $env.GITHUB_WORKSPACE = ("." | path expand)
- $env.GITHUB_OUTPUT = ("./output/out.txt" | path expand)
- let version = (open Cargo.toml | get package.version)
- mkdir $artifacts_dir
- for target in ["aarch64" "x86_64"] {
- $env.TARGET = $target ++ "-pc-windows-msvc"
-
- rm -rf output
- _EXTRA_=bin nu .github/workflows/release-pkg.nu
- cp $"output/nu-($version)-($target)-pc-windows-msvc.zip" $artifacts_dir
-
- rm -rf output
- _EXTRA_=msi nu .github/workflows/release-pkg.nu
- cp $"target/wix/nu-($version)-($target)-pc-windows-msvc.msi" $artifacts_dir
- }
+ $env.RUSTFLAGS = ""
+ $env.CARGO_TARGET_DIR = ""
+ hide-env RUSTFLAGS
+ hide-env CARGO_TARGET_DIR
+ $env.OS = "windows-latest"
+ $env.GITHUB_WORKSPACE = ("." | path expand)
+ $env.GITHUB_OUTPUT = ("./output/out.txt" | path expand)
+ let version = (open Cargo.toml | get package.version)
+ mkdir $artifacts_dir
+ for target in ["aarch64" "x86_64"] {
+ $env.TARGET = $target ++ "-pc-windows-msvc"
+
+ rm -rf output
+ _EXTRA_=bin nu .github/workflows/release-pkg.nu
+ cp $"output/nu-($version)-($target)-pc-windows-msvc.zip" $artifacts_dir
+
+ rm -rf output
+ _EXTRA_=msi nu .github/workflows/release-pkg.nu
+ cp $"target/wix/nu-($version)-($target)-pc-windows-msvc.msi" $artifacts_dir
+ }
}
-export def main [] { help toolkit }
+export def main [] {help toolkit} |
|
to be more clear, the funny part is that it formatted this if $no_fail { [true true true true] }
else if $fail_fmt { [false null null null] }
else if $fail_clippy { [true false null null] }
else if $fail_test { [true true false null] }
else if $fail_test_stdlib { [true true true false] }
else { [null null null null] }
like this if $no_fail {[true true true true]} else if $fail_fmt {[false null null null]} else if $fail_clippy {[true false null null]} else if $fail_test {[true true false null]} else if $fail_test_stdlib {[true true true false]} else {[null null null null]}i didn't expect it to align things, of course, but i also didn't expect it to put all that on one line. lol |
|
@fdncred, yes, it's definitely a style issue, I'm trying to fix it. Seems a lot rules about newlines in parentheses are still missing. |
|
@fdncred The style issue of |
|
I guess I've done something else wrong. This is on MacOS. I pulled the latest from topiary-nushell. I pulled the latest from tree-sitter-nu, did tree-sitter generate and tree-sitter build. Updated the languages.ncl file. Ran it from the command line and from nvim with the same result. |
|
@fdncred I think it's caused by my latest PR to tree-sitter-nu (the renaming). Since it is merged now, I'm going to update this repo. |
|
@fdncred Done. |
|
Seems like it did something. Nice work! That's better than before. |
|
@fdncred I'm closing this issue for now, since it's getting a little bit too long to follow. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
First of all, I'm super excited for this. I realize it's not 100% yet because tree-sitter-nu isn't 100% yet but I'm still excited nonetheless.
When I tried to run
topiary fmt toolkit.nufrom the nushell repo root. It produces this error.I haven't dug in to see what's going on yet, but I thought I'd post it.
Thanks for working on this and supporting nushell! ❤️
The text was updated successfully, but these errors were encountered: