Project

miniexec

0.0
No commit activity in last 3 years
No release in over 3 years
There's a lot of open issues
A minimal interpretor/executor for .gitlab-ci.yml
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Runtime

~> 1.8
 Project Readme

miniexec logo

miniexec

A minimal interpretor/executor for .gitlab-ci.yml files

Usage: miniexec.rb [options]

specific options:
    -p, --path PATH                  Path to the repository containing a valid .gitlab-ci.yml
    -j, --job JOBNAME                Specify the gitlab job to run
    -b, --bind BIND                  Specify a bind mapping
                                     Example: /some/local/dir:/mapping/in/container
    -e, --environment VARIABLE       Specify an environment variable to be passed to the container
                                     Example: SOMEVAR=thing

Motivation

Writing jobs for Gitlab CI can be a frustrating experience. Having to write a job, push your change, wait for N steps in the pipeline to complete only to attempt to run your job and tell you that you forgot a particular environment variable... we've all been there.

Gitlab's own CI runner 'gitlab-runner' offers a function called 'gitlab-runner exec', which will attempt to run a particular CI job locally. It works ok, but doesn't support advanced YAML features like anchors correctly, and exec generally isn't actively developed by Gitlab.

In comes miniexec - a simple parser and executor for .gitlab-ci.yml files. Using miniexec, you can write Gitlab CI jobs and quickly test them on your local machine.

Installation

miniexec is available on RubyGems and can be installed like so:

gem install miniexec

Alternatively, clone this repository and build it yourself:

git clone https://github.com/s3krit/miniexec
gem build miniexec.gemspec
gem install miniexec-*.gem

Basic usage

In the directory of a project that has a .gitlab-ci.yml file, simply call miniexec like so (the following examples can all be run directly from this repository):

miniexec -j miniexec-example-1

miniexec will then attempt to spawn a Docker container using the image specified for that job (or the global image), and execute the steps in that job.

By default, miniexec will attempt to detect the WorkDir of the container - if this is not set, the WorkDir will be set to /gitlab. The contents of the working directory will also automatically be mounted inside the container to the WorkDir.

You can mount additional directories using the -b flag. The following example will mount the current working directory to /build and echo README.md (the file you're reading right now!)

miniexec -j miniexec-example-2 -b $PWD:/build

You'll also likely want to pass certain environment variables to miniexec, such as API keys or configuration arguments. With miniexec, just specify each one with a -e flag, just like Docker!

miniexec -j miniexec-example-3 -e MY_USER=$USER -e MY_KEY=c3r34l-k1ll4h

What works

  • Specifying and running a job from a .gitlab-ci.yml file
  • YAML anchors and other fancy features
  • Podman! If you run podman with a user socket, simply specify the location of the socket with -d or the DOCKER_HOST environment variable
  • Mocking of Gitlab's predefined environment variables as best I can, so some of the Git stuff. This needs to be expanded.

What doesn't

  • MiniExec currently makes the assumption that the job is a shell-script and that /bin/bash is present - This is a gitlab restriction more than MiniExec's fault
  • If you uncleanly exit miniexec, you'll probably have to kill the container manually.
  • Sometimes the first couple lines of output in a job will be skipped
  • Advanced Gitlab CI features like services
  • Running full pipelines
  • Probably some other stuff. If you find one, make an issue, I'd love to fix it.
  • Specifying your environment variables with a file instead of one-by-one (soon™)
  • 'File' style variables