Quick fixes have stopped working for me.

Now I just get stuck on “Checking for quick fixes...” and the following warning in the Extension Host console:

[warning] trunk.io - Code actions of kind 'quickfix 'requested but returned code action is of kind 'source.fixAll'. 
Code action will be dropped. Please check 'CodeActionContext.only' to only return requested code actions.

cli:
  version: 1.21.0
plugins:
  sources:
    - id: trunk
      ref: v1.4.5

I need some guidance in solving this issue.

I have an unexpected (for me) new issue after working with this extension for a couple of months now within .trunk/trunk.yaml:

So I tested out the command (found on https://github.com/trunk-io/docs/tree/main/check/advanced-setup/plugins) in the root of my project to force a configuration update and ultimately the creation of missing plugin.yaml.

trunk plugins add https://github.com/trunk-io/plugins v1.2.6 --id=trunk

but the command also returns an error:

> trunk plugins add https://github.com/trunk-io/plugins v1.5.0 --id=trunk1
✖ plugin operations require a valid trunk config

Details

• Force updated to latest trunk --version: 1.21.0
• Force updated VSCode extension: Trunk Check v3.18.0
• VSCode: v1.90.0 (on Windows remote SSH into Ubuntu)

I added a stylua linter to my .trunk.yaml and thought it worth inclusion.

I don't have the time to massage it into shape for a PR but it mostly works.

Here are the relavant parts:

version: 0.1
cli:
  version: 1.6.1
lint:
  files:
    - name: lua
      extensions:
        - lua
      shebangs:
        - lua
  downloads:
    - name: stylua
      version: 0.17.0
      downloads:
        - os:
            linux: linux
            macos: macos
          cpu:
            x86_64: x86_64
            arm_64: aarch64
          url: https://github.com/JohnnyMorganz/StyLua/releases/download/v${version}/stylua-${os}-${cpu}.zip
  enabled:
    - [email protected]
  definitions:
    - name: stylua
      files: [lua]
      download: stylua
      commands:
        - formatter: true
          output: rewrite
          run: stylua --verify --search-parent-directories --stdin-filepath ${target} -
          stdin: true
          success_codes: [0, 1]
      run_linter_from: workspace
      good_without_config: true
      direct_configs:
        - .stylua.toml
      environment:
        - name: PATH
          list: ["${linter}"]
      version_command:
        parse_regex: ${semver}
        run: stylua --version

It seems like trunk upgrade -v is very very slow here.

I've tested it much longer as well.

Anyone can help with that?

Infos about my system:

OS: Manjaro Linux Sway Edition
CPU: AMD Ryzen 9 5900X 12-Core Processor
RAM: 48093 MiB

When using perltidy, it fails if it isn't installed on the host system.

This is problematic in CI:

Example:

https://github.com/docwhat/bashables/actions/runs/9483902974/job/26132129649

# .trunk/out/ppqBs.yaml
  trunk_cli_version: 1.22.1
  title: "Error while executing: tests/test_helper/pseudo-tty.pl"
  report:
    - message: Unable to find binary in PATH
      binary: perltidy
      PATH:
        - /home/runner/.local/share/bats/bin
        - /snap/bin
        - /home/runner/.local/bin
        - /opt/pipx_bin
        - /home/runner/.cargo/bin
        - /home/runner/.config/composer/vendor/bin
        - /usr/local/.ghcup/bin
        - /home/runner/.dotnet/tools
        - /usr/local/sbin
        - /usr/local/bin
        - /usr/sbin
        - /usr/bin
        - /sbin
        - /bin
        - /usr/games
        - /usr/local/games
        - /snap/bin
    - linter:
        command: |
  
        stdin_path: (none)
        run_from: ""
        timeout: 0
        rerun: ""
        exit_status: exited
        exit_code: 0
        stdout: (none)
        stderr: (none)
      parser: |
        (none)

The commit message for 097a7d3 says this was done deliberately.

I can be convinced that we don't need to install a local perl runtime, but the command should be installed at the very least with a custom, trunk-specific PERL5LIB.

It'd be great to have a mechanism which allows extending a set of dependencies for a given tool rather than replacing them. E.g.

Suppose you have

      extra_packages:
        - [email protected]
        - [email protected]

for one of your configs. Then

      extra_packages:
        - [email protected]

will replace all of them.

Use case:

1. Your organization uses ESLint and has a global config.
2. A local project has specific-requirements requiring another ESLint plugin not defined in the global config.

It's likely that extra_packages will always be additive, so it may be worth just making extra_packages additive without another mechanism.

trunk, especially trunk action commands, could really use shell completion.

Currently, I am not able to configure Trunk to use eslint and prettier to check *.svelte files. These linters have Svelte support via plugins.

        Traceback (most recent call last):
          File "/home/ryang/.cache/trunk/plugins/https---github-com-trunk-io-plugins/v0.0.11-7784b17d1d8ff658/linters/ruff/ruff_to_sarif.py", line 59, in <module>
            "deletedRegion": get_region(fix),
          File "/home/ryang/.cache/trunk/plugins/https---github-com-trunk-io-plugins/v0.0.11-7784b17d1d8ff658/linters/ruff/ruff_to_sarif.py", line 10, in get_region
            location = entry["location"]
        KeyError: 'location'

When running ruff on 0.0.272 with .ruff.toml

# Enable flake8-pyinterface, pylint error, pylint warning
select = ["PYI", "PLE", "PLW"]
ignore = ["B905", "PLW0603", "PLW2901", "B018", "B028"]
# Full list of error codes https://beta.ruff.rs/docs/rules/
# Disable Bugbear zip
# Pylint warning of global
# Overwritten variables
# pointless setement
# stacklevel not added to log.warning

# Allow autofix for all enabled rules (when `--fix`) is provided.
fixable = [
  "A",
  "B",
  "C",
  "D",
  "E",
  "F",
  "G",
  "I",
  "N",
  "Q",
  "S",
  "T",
  "W",
  "ANN",
  "ARG",
  "BLE",
  "COM",
  "DJ",
  "DTZ",
  "EM",
  "ERA",
  "EXE",
  "FBT",
  "ICN",
  "INP",
  "ISC",
  "NPY",
  "PD",
  "PGH",
  "PIE",
  "PL",
  "PT",
  "PTH",
  "PYI",
  "RET",
  "RSE",
  "RUF",
  "SIM",
  "SLF",
  "TCH",
  "TID",
  "TRY",
  "UP",
  "YTT",
]
unfixable = []

# Exclude a variety of commonly ignored directories.
exclude = [
  ".keys",
  ".cache",
  ".vscode",
  ".git",
  ".git-rewrite",
  ".hg",
  ".mypy_cache",
  ".nox",
  ".pants.d",
  ".pytype",
  ".ruff_cache",
  ".svn",
  ".tox",
]
per-file-ignores = {}

line-length = 120

target-version = "py310"

This is what I get after running the linter upgrade:

And then after that, the Trunk Side Panel continuously shows there is an upgrade available:

title: "Error while executing: Downloading hermetic hadolint v2.12.0"
report:

• HTTP 404 'https://github.com/hadolint/hadolint/releases/download/v2.12.0/hadolint-Darwin-arm64'

trunk check --all --filter cspell checks 187 files.

cspell --gitignore . checks 163 files.
cspell --gitignore "**" checks 163 files.

Can't figure out how to get 187 files when running cspell directly, so curious how trunk is running it...

e.g. sveltekit uses *.postcss files to trigger automatic postcss processing when using the default vite preprocessor. Prettier supports formatting these files.

I have custom tflint config in the root of my git repository. What I've noticed, is that tflint doesn't use it, if i try to specify it via direct_configs like this

lint:
  definitions:
    - name: tflint
      direct_configs: 
       - .tflint.hcl
       - ${workspace}/.tflint.hcl # doesn't work

tbh, I haven't find any clear description for direct_configs attribute, so still not sure that this is proper place to specify custom config location

Current workaround for me is to specify config dir via env variable

lint:
  definitions:
    - name: tflint
      environment:
        - name: TFLINT_CONFIG_FILE
          value: ${workspace}/.tflint.hcl

or via --config argument

lint:
  definitions:
    - name: tflint
      commands:
        - name: lint
          run: tflint --format=sarif --chdir=${target} --force --config=${workspace}/.tflint.hcl

Info

paths:

• ${workspace}/.trunk/trunk.yaml
• ${workspace}/.tflint.hcl

versions

• [email protected]
• [email protected]

Trunk doesn't seem to honor the markdownlint config changes I've made.

link to example of mapping.

I've tried a lot of variations but I can't get trunk to honor .markdownlint.yaml
I run the markdownlint vscode plugin and it honors the file, but in repo when I tweak it seems to ignore the file.

For example changing indentation level to 12, it works with markdownlint tool itself, but then using trunk as my formatter it seems to go back to standard config and not honor it.
I've tried both .markdownlint.yaml and .markdownlint.jsonc and neither seems to be picked up.
I've also tried placing in root of repo and copying to .trunk/configs and no change in behavior.

Trunk config

version: '0.1'
cli:
  version: 1.22.1
plugins:
  sources:
    - id: trunk
      ref: v1.5.0
      uri: https://github.com/trunk-io/plugins
lint:
  enabled:
    - [email protected]

.markdownlint.yaml

$schema: https://raw.githubusercontent.com/DavidAnson/markdownlint/main/schema/markdownlint-config-schema.json
MD007:
  indent: 12
  start_indented: false
  start_indent: 4

- [ ] one
  - _In use by:_
    - this
- [ ] more detail
  - _the result_
    - more
    - foo

- [ ] one
            - _In use by:_
                        - this
- [ ] more detail
            - _the result_
                        - more
                        - foo

Issue

When running trivy through trunks plugin, I'm getting a failure on all files due to trivy hitting problems surrounding being invoked. Here is the copy / pasta of the stderr:

        2023-06-21T14:06:56.666-0600	�[34mINFO�[0m	Need to update DB
        2023-06-21T14:06:56.666-0600	�[34mINFO�[0m	DB Repository: ghcr.io/aquasecurity/trivy-db
        2023-06-21T14:06:56.666-0600	�[34mINFO�[0m	Downloading DB...
        2023-06-21T14:06:56.666-0600	�[31mFATAL�[0m	init error: DB error: failed to download vulnerability DB: database download error: OCI repository error: 1 error occurred:
        	* error getting credentials - err: exec: "docker-credential-osxkeychain": executable file not found in $PATH, out: ``

This is due to trunk setting the PATH to only the linter in the environment as seen here: https://github.com/trunk-io/plugins/blob/main/linters/trivy/plugin.yaml#L56-L58

This seems to be an artifact of trivy having an implicit dependency on docker-credential-osxkeychain.

I have the tool on my system, but it is not PATH set by trivy.

See screenshot of the full error report:

Expected

I would expect trunk to either download ALL dependencies for their plugins so they run in an isolated fashion. This would be preferred and would get around this issue.

The other option is to provide a workaround so users like myself can better override the environment / PATH set for the plugin. That would allow me to specify the correct PATH so that trivy can execute as part of trunk. This has major issues as it is likely that setting the PATH in something like trunk.yaml would not be portable to other team members machines and would be fragile.

Notes

If we have a suggested plan of action for this issue, then I'd be happy to implement it as I don't believe the fix should be a major one. Maybe adding an additional downloads entry for the docker-credential-osxkeychain tool so that is installed alongside trivy?

Problem

When attempting to create a codespace from this repository's main branch, the codespace fails to complete building.

Steps To Reproduce

1. Open Visual Studio Code (attempted through GitHub UI as well)
2. Run command Codespaces: Create New Codespace...
3. Select trunk-io/pluginsusing the official GitHub Repositories extension
4. Wait for Setting up remote connection: Building codespace... to enter into a recovery container

Expected Result

A codespace to be built and usable.

Actual Result

Setting up remote connection: Building codespace... spins indefinitely.

Suggested Solution

The .devcontainer.json file contains the configuration for any built codespaces, so the issue is most likely in this file. The image key seems to be using an outdated image of "mcr.microsoft.com/devcontainers/typescript-node:0-18" based on its documentation, so replacing it with a newer image might fix the issue. It is important that semantic versioning is still used along with the same Node and Debian versions so as to not introduce more issues.

The issue might also be related to the keys git-lfs or onCreateCommand.

The help-URLs of clang-tidy need a / instead a - at the first occasion in the replacement here:

E.g. for cppcoreguidelines-pro-type-vararg:

• current (returns 404 Not Found): https://clang.llvm.org/extra/clang-tidy/checks/cppcoreguidelines-pro-type-vararg.html
• fix: https://clang.llvm.org/extra/clang-tidy/checks/cppcoreguidelines/pro-type-vararg.html

(I am using yarn 3)

If I add commitlint as an action to my project it fails silently. If I use the commitlint cli, the output is helpful.

Hello,

Was trying trunk out on a rust library and per best practices for libraries, Cargo.lock is not part of the repository. So while trunk runs fine locally after I have built the library once, it fails in CI. It also fails locally if I delete Cargo.lock first. This happens because the command for the clippy linter has "--locked" as part of the command which doesn't permit changing (or creating by extension?) the Cargo.lock file. And because of this, clippy errors.

I looked through your documentation but what I saw about configuring plugins appeared to be only what options are defined in the plugin.yaml file. However, the locked flag is hard-coded and not affected by any values that I saw when looking in the plugins repo.

If I'm missing something, I apologize but I did investigate as best I could.

trunk output:

trunk_cli_version: 1.18.1
title: "Error while executing: lib"
report:
  - Unable to find lib for Cargo.lock in any parent directory of /Volumes/LaCie/Repos/dfraw_json_parser/lib
  - linter:
      command: |
        /Users/nwesterhausen/.cache/trunk/tools/rust/1.65.0-e995c0a94cbb3cb0410a49183f3ef2b7/bin/cargo clippy --message-format json --locked -- --cap-lints=warn --no-deps
      stdin_path: (none)
      run_from: /Volumes/LaCie/Repos/dfraw_json_parser/lib
      timeout: 10m
      rerun: (cd /Volumes/LaCie/Repos/dfraw_json_parser/lib; env -i PATH=/Users/nwesterhausen/.cache/trunk/tools/rust/1.65.0-e995c0a94cbb3cb0410a49183f3ef2b7/bin:/Users/nwesterhausen/.opam/default/bin:/Users/nwesterhausen/.bin:/opt/homebrew/bin:/opt/homebrew/sbin:/usr/local/bin:/System/Cryptexes/App/usr/bin:/usr/bin:/bin:/usr/sbin:/sbin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/local/bin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/bin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/appleinternal/bin:/Library/Apple/usr/bin:/Users/nwesterhausen/.opam/default/bin:/Users/nwesterhausen/Library/pnpm:/Users/nwesterhausen/.bin:/opt/homebrew/bin:/opt/homebrew/sbin:/Users/nwesterhausen/.cargo/bin CARGO_TARGET_DIR=/Users/nwesterhausen/.cache/trunk/repos/a3e74e59ed19111a76af0d59e8fcaafe/results/clippy-current/cargo_target_dir CARGO_HOME=/Users/nwesterhausen/.cache/trunk/repos/a3e74e59ed19111a76af0d59e8fcaafe/results/clippy-current/cargo_home HOME=/Users/nwesterhausen /Users/nwesterhausen/.cache/trunk/tools/rust/1.65.0-e995c0a94cbb3cb0410a49183f3ef2b7/bin/cargo clippy --message-format json --locked -- --cap-lints=warn --no-deps)
      affects_cache:
        - Cargo.toml
      direct_configs:
        []
      exit_status: exited
      exit_code: 101
      stdout: (none)
      stderr: |
            Updating crates.io index
        error: the lock file /Volumes/LaCie/Repos/dfraw_json_parser/Cargo.lock needs to be updated but --locked was passed to prevent this
        If you want to try to generate the lock file without accessing the network, remove the --locked flag and use --offline instead.

    parser: |
      (none)

linters/plugin.yaml has a clear alphabetical ordering for array values but there are some inconsistencies.

I'm not sure if this is the right repository to report this issue, but when trunk-check-pre-push is enabled, I cannot use git push --delete <remote> <revspec> syntax.

🕙 11:05:00 quickprice ❯ git push --delete origin uno
✖ Git Error: revspec '(delete)' not found
✖ Push blocked by failing hook 'trunk-check-pre-push'
error: failed to push some refs to '...'

Reference

Hello, I'm using vscode trunk extension and I find that the binary ~/.cache/trunk/cli/1.19.0-darwin-arm64/crashpad_handler will be run in the background from time to time, which triggered my firewall. How to disable this behavior? I can't find any settings in vscode.

The markdown-table-prettify tool formats markdown tables.

I don't know if this type of formatting can be done with prettier or any other plugin already listed, but I would really appreciate you guys adding this since it's pretty (pun intended) useful.

I can't find anything explaining what the trunk-share-with-everyone action does.

👋 hullo!

trunk doesn't appear to surface biome format errors, not sure why, haven't dug into it. still on WSL.

here's a malformed biome config that complains correctly when run with pnpm biome format .:

❯ cat biome.json
{
        "$schema": "https://biomejs.dev/schemas/1.7.3/schema.json",
        "organizeImports": {
                "enabled": true
        },
        "linter": {
                "enabled": true,
                "rules": {
                        "recommended": true
                }
        },
        "formatter": {
                "enabled": true,
                "rules": {
                        "recommended": true,
                        "formatWithErrors": true,
                        "indentStyle": "spaces",
                        "indentWidth": 2,
                        "lineEnding": "lf",
                        "lineWidth": 120
                }
        },
        "javascript": {
                "formatter": {
                        "quoteStyle": "single",
                        "jsxQuoteStyle": "single",
                        "semicolons": "asNeeded",
                        "trailingComma": "all"
                }
        }
}

I'm seeing terrascan failing, due to lack of support for a terraform feature I'm using. This causes trunk to report failure to run as well. What would be the suggest appraoch for dealing with the scenario where upstream has a known bug that is causing one of the plugins to fail?

I'd prefer trunk to understand that for 3rd praty plugin version X, it will fail on this fail and thus should be set to skip it up until the next version is released (after which resume scanning..)

Reference tenable/terrascan#1642 (comment)

Info

Aqua is a great tool for managing other tools and adding new tools to that registry looks easy overall. It would be great to see trunk in their primary registry. Here is an example of doing so: https://github.com/aquaproj/aqua-registry/pull/13657/files

There is a special ref, 0000000000000000000000000000000000000000, that is the NULL commit. It is the first parent commit in any new repository.

push.sh should detect and handle this gracefully, either with --all or by just quitting with exit code 0.

Example output from https://github.com/docwhat/gamelist-xml-util/actions/runs/3897696813/jobs/6655627435

Run # Run trunk check on push
  # Run trunk check on push
  ${GITHUB_ACTION_PATH}/push.sh
  shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}
  env:
    TRUNK_TMPDIR: /tmp/tmp.mPY6YzZiZU
    TRUNK_PATH: /tmp/tmp.mPY6YzZiZU/trunk
    TRUNK_CHECK_MODE: push
    MSYS: winsymlinks:nativestrict
    GITHUB_EVENT_AFTER: 2e5cacb1cd3e93e5c1553[40](https://github.com/docwhat/gamelist-xml-util/actions/runs/3897696813/jobs/6655627435#step:3:44)3359772213eb98c83
    GITHUB_EVENT_BEFORE: 0000000000000000000000000000000000000000
    GITHUB_REF_NAME: remove-renovate
    GITHUB_REPOSITORY: docwhat/gamelist-xml-util
    GITHUB_TOKEN: ***
    INPUT_ARGUMENTS: 
    INPUT_LABEL: 
fatal: remote error: upload-pack: not our ref 0000000000000000000000000000000000000000
Error: Process completed with exit code 128.

N.B. The NULL commit is also a blob as well. You can see that with renames. However, that doesn't apply here.

trunk_cli_version: 1.15.0
title: "Error while executing: Querying compile commands"
report:
  - "[json.exception.type_error.302] type must be string, but is number"

I tried to place compile_commands.json in:

1. root workspace
2. folder build in root workspace
3. folder with current cpp file

Anything does not work

I'd like to disable B101 for tests per:

https://bandit.readthedocs.io/en/latest/plugins/b101_assert_used.html#module-bandit.plugins.asserts

You can configure files that skip this check. This is often useful when you use assert statements in test cases.

assert_used:
  skips: ['*_test.py', '*test_*.py']

Right now, if I understand correctly, we can only use a .bandit ini file. As far as I can tell only has the ability to skip files or disable checks across for the entire repo.

https://github.com/trunk-io/plugins/blob/main/linters/bandit/plugin.yaml#L20

https://github.com/trunk-io/plugins/blob/bc24c45b5e2502bad62de4e30dc9825c863efa66/linters/shellcheck/plugin.yaml#L41C1-L44C1

This can be set in user home file to override preferences and make them more global.
This doesn't seem to be mapped or respected by trunk.

Let me open the feature request in the repository, since trunk-io/trunk-action doesn't allow me to create issues.

Currently, trunk-io/trunk-action/upgrade doesn't support signed commit. If we turn on the setting that commit is required in a GitHub repository, the check doesn't pass. So, it would be great to support signed commit in the action.

I'm running this command for all 3 scenarios:

trunk check --all --filter cspell

Scenario 1

cspell.yaml:

version: "0.2"
suggestionsTimeout: 5000

Result: 94 issues

Scenario 2

cspell.yaml:

version: "0.2"
suggestionsTimeout: 5000
dictionaries:
  - python

Result: 69 issues

Scenario 3

cspell.yaml:

version: "0.2"
suggestionsTimeout: 5000
dictionaryDefinitions:
  - name: cspell-custom-dict
    path: ./cspell-custom-dict.txt
dictionaries:
  - cspell-custom-dict

Result: No issues¹

Other Info

Reference Config:

• https://github.com/streetsidesoftware/cspell/blob/main/examples/cspell.config.yaml
• https://cspell.org/docs/dictionaries-custom/

trunk.yaml:

---
# This file controls the behavior of Trunk: https://docs.trunk.io/cli
# To learn more about the format of this file,
# see https://docs.trunk.io/reference/trunk-yaml
version: 0.1
cli:
  version: 1.17.1
plugins:
  sources:
    - id: trunk
      ref: v1.2.6
      uri: https://github.com/trunk-io/plugins
runtimes:
  definitions:
    - type: python
      system_version: allowed
lint:
  disabled:
    - autopep8
    - checkov
    - pyright
    - sourcery
    - yapf
  enabled:
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - git-diff-check
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
    - [email protected]
actions:
  disabled:
    - commitlint
    - trunk-announce
    - trunk-check-pre-push
    - trunk-fmt-pre-commit
  enabled:
    - trufflehog-pre-commit
    - trunk-upgrade-available

Motivation

It would be awesome to support linters and formatters for Open Policy Agent and Rego. For example, regal is a linter for Rego.

https://github.com/StyraInc/regal

Hello,

Thanks for the awesome "master-linter" - the UI/UX is great, it's very useful.

Just a little feedback, share by other dev friends : it's heavy on the processor, and we had to uninstall it (at least from VSCode) to solve the issue. After 5 minutes of code, compute usage would shoot to 100%, and my MacBook Air 2021 would get very, very warm.

I think the VSCode plugin is at fault, however the behavior persisted after I removed it, and stopped when I did a clean uninstallation. I reinstalled only the CLI tool after, which solved the problem.

Thanks for the good work, I hope this feedback helps :-)

Hi,

I am using trunk in a Bazel project. It works nicely so far, the only issue that I have is that the Bazel calls by trunk constantly invalidate the cache of the Bazel server. That means that I have to recompile every dependency, when I want to run a file (takes about 1 minute, instead of an incremental build taking 10 secs).

Maybe this issue is caused by the env being different for my zsh shell and trunk. Are there any advises on how to fix this?

Best,

Henrik

Can we include some integrity checks for the linters configured in this repo?

The schema for trunk.yaml allows a shasum attribute, but the problem is that the URLs are parametrized by version for most linters configured in this repo, and so there is no one fixed shasum. Instead, we could have a central map/registry of all known download URLs to their shasums, and then if a user is using a download URL with an unavailable shasum, warn them and show them how to extend the shasum registry for their repo.

Filing it here for consideration. Apologies if this already happens in some way.

Motivation

Config files of linters are located at the path specialized to trunk.io under .trunk/configs. Some extensions or plugins of IDEs as VS code and IntelliJ don't recognize the config files, because those expect the default location of each linter. We can use a function to format code on IDE with config files using extensions and plugins. If we can seamlessly share or synchronize config files between Trunk and IDE, that would be awesome.

Ideas

Plan 1: Use symbolic links between paths for Trunk and default paths

I usually create some symbolic links to make Trunk and the IDE recognize. If I already have a config file of a linter before introducing Trunk, I create a symbolic link from the existing file in .trunk/configs. On the contrary, if I first introduce a new linter by Trunk, I create the symbolic link to the default path which the linter expects. We can also implement a sub command to manage symbolic links between Trunk and default paths of linters.

Plan 2: Trunk tris to load a config file of a linter from the default path

As far as I know, we define paths of config files so that Trunk can load. For instance, the default config path of autopep8 in Trunk is .pep8. We can also make Trunk load config files from default paths of linters.

Hey!

This issue is a feature request to add support for terraform-docs, a tool that automatically generates documentation from Terraform modules.

Currently, we use pre-commit to ensure that our Terraform documentation stays in sync with the code, but it would be great to have just one tool that contains all the typical Terraform plugins.

I believe this feature would be a valuable addition to the plugins library and would greatly benefit Terraform users.

Thank you for considering this! 🌟

This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.

Rate-Limited

These updates are currently rate-limited. Click on a checkbox below to force their creation now.

• chore(deps): update actions/setup-node action to v4
• chore(deps): update dependency @types/node to v20
• chore(deps): update dependency eslint-config-prettier to v9
• fix(deps): update commitlint monorepo to v18 (major) (@commitlint/cli, @commitlint/config-conventional)
• 🔐 Create all rate-limited PRs at once 🔐

Open

These updates have all been created already. Click a checkbox below to force a retry/rebase of any.

• chore(deps): update trunk-io/trunk-action digest to c4b725d
• chore(deps): update all non-major dependencies (@types/caller, @types/debug, @types/jest, @types/jest-specific-snapshot, @types/node, @typescript-eslint/eslint-plugin, @typescript-eslint/parser, eslint, github/codeql-action, simple-git, typescript)
• chore(deps): update actions/checkout action to v4
• Click on this checkbox to rebase all open PRs at once

Detected dependencies

• Check this box to trigger a request for Renovate to run again on this repository

Replaced since August.

It also supports JSON.

Hello,

I'm trying to use Trunk+VSCode in remote mode on a headless Pi 4 4GB and I get a mmap crash.

uname -m
aarch64

command -v trunk
/home/etu/.cache/trunk/launcher/trunk

trunk init
external/tcmalloc/tcmalloc/system-alloc.cc:662] MmapAligned() failed - unable to allocate with tag (hint, size, alignment) - is something limiting address placement? 0x241a40000000 1073741824 1073741824 @ 0x556cf3caa8 0x556cf3a740 0x556cf3a324 0x556cf29a78 0x556cf37638 0x556cf25ba0 0x556ca8230c 0x556bd3e618 0x556bd0b848 0x7fb427783c
external/tcmalloc/tcmalloc/arena.cc:64] FATAL ERROR: Out of memory trying to allocate internal tcmalloc data (bytes, object-size); is something preventing mmap from succeeding (sandbox, VSS limitations)? 131072 640 @ 0x556cf3d328 0x556cf29af4 0x556cf37638 0x556cf25ba0 0x556ca8230c 0x556bd3e618 0x556bd0b848 0x7fb427783c

Has anyone met the same issue ?

TIA,

I propose that this repository moves to a continuous delivery model where every change/commit accepted into main and passes all the CI tests will trigger a release automatically.

This should be possible now that there is test coverage and good linting (trunk!).

The use case for this is PR #237, which breaks hadolint on arm64 macOS systems. Not that big a deal if updating, but on a new project is a pain and is confusing. A CD setup would have this fixed within minutes of #237 being approved and merged.

A quick estimate of the work required:

1. Ensure the branch protections are up-to-date.
2. Ensure the test coverage is sufficient for confidence.
3. Write a CI workflow to tag and release a version based on comments. Something like Release Please could be used with conventional commits.
4. Write documentation.
5. Profit!

Hi,

I downloaded "trufflehog" with the latest update, however it seems to fail when processing the trunk.yaml files.

trunk version : 1.11.1

This seems like something that should be relatively easy to catch with a test: #505