Giter Club home page Giter Club logo

cqfd's Introduction

What is cqfd ?

cqfd provides a quick and convenient way to run commands in the current directory, but within a Docker container defined in a per-project config file.

This becomes useful when building an application designed for another Linux system, e.g. building an old embedded firmware that only works in an older Linux distribution.

Using cqfd

Getting started

Just follow these steps:

  • Install cqfd (see below)
  • Make sure your user is a member of the docker group
  • Go into your project's directory
  • Create a .cqfdrc file
  • Put a Dockerfile and save it as .cqfd/docker/Dockerfile
  • Run cqfd init

Examples are available in the samples/ directory.

cqfd will use the provided Dockerfile to create a normalized runtime build environment for your project.

Using cqfd on a daily basis

Regular builds

To build your project from the configured build environment with the default build command as configured in .cqfdrc, use:

$ cqfd

Alternatively, you may want to specify a custom build command to be executed from inside the build container.

$ cqfd run make clean
$ cqfd run "make linux-dirclean && make foobar-dirclean"

When cqfd is running, the current directory is mounted by Docker as a volume. As a result, all the build artefacts generated inside the container are still accessible in this directory after the container has been stopped and removed.

Release

The release command behaves exactly like run, but creates a release tarball for your project additionally. The release files (as specified in your .cqfdrc) will be included inside the release archive.

$ cqfd release

The resulting release file is then called according to the archive template, which defaults to %Po-%Pn.tar.xz.

Flavors

Flavors are used to create alternate build scenarios. For example, to use another container or another build command.

The .cqfdrc file

The .cqfdrc file at the root of your project contains the information required to support project tooling. samples/dot-cqfdrc is an example.

Here is a sample .cqfdrc file:

[project]
org='fooinc'
name='buildroot'
build_context='.'

[build]
command='make foobar_defconfig && make && asciidoc README.FOOINC'
files='README.FOOINC output/images/sdcard.img'
archive='cqfd-%Gh.tar.xz'

The [project] section

org: a short, lowercase name for the project’s parent organization.

name: a short, lowercase name for the project.

build_context (optional): a directory to pass as the build context to Docker. This should be specified relatively to where cqfd is invoked. In the example above, the whole current working directory of cqfd is passed to Docker as its build context, so most people will also add a .dockerignore file to avoid that.

Generated Docker images for your project will be named $org_$name.

The [build] section

command: the command (or list of commands) to be executed when cqfd is invoked. This string will be passed as an argument to a classical bash -c "commands", within the build container, to generate the build artefacts.

files: an optional space-separated list of files generated by the build process that we want to include inside a standard release archive.

archive: the optional name of the release archive generated by cqfd. You can include environment variable names, as well as the following template marks:

  • %Gh - git short hash of last commit
  • %GH - git long hash of last commit
  • %D3 - RFC3339 date (YYYY-MM-DD)
  • %Cf - current cqfd flavor name (if any)
  • %Po - value of the project.org configuration key
  • %Pn - value of the project.name configuration key
  • %% - a litteral '%' sign

By default, cqfd will generate a release archive named org-name.tar.xz, where 'org' and 'name' come from the project's configuration keys. The .tar.xz, .tar.gz and .zip archive formats are supported.

For tar archives, setting tar_transform=yes will cause all files specified for the archive to be stored at the root of the archive, which is desired in some scenarios. This feature is not supported with .zip archives.

distro: the name of the directory containing the Dockerfile. By default, cqfd uses "docker", and ``.cqfd/docker/Dockerfile` is used.

flavors: the list of build flavors (see below). Each flavor has its own command just like build.command.

Using build flavors

In some cases, it may be desirable to build the project using variations of the build and release methods (for example a debug build). This is made possible in cqfd with the build flavors feature.

In the .cqfdrc file, one or more flavors may be listed in the [build] section, referencing other sections named following flavor's name.

[centos7]
command='make CENTOS=1'
distro='centos7'

[debug]
command='make DEBUG=1'
files='myprogram Symbols.map'

[build]
command='make'
files='myprogram'
flavors='centos7 debug'

A flavor will typically redefine some keys of the build section: command, files, archive, distro.

Flavors from a .cqfdrc file can be listed using the flavors argument.

Environment variables

The following environment variables are supported by cqfd to provide the user with extra flexibility during his day-to-day development tasks:

CQFD_EXTRA_RUN_ARGS: A space-separated list of additional docker-run options to be append to the starting container. Format is the same as (and passed to) docker-run’s options. See 'docker run --help'.

Other command-line options

In some conditions you may want to use an alternate config file with cqfd. This is what the -f option is for:

$ cqfd -f .cqfdrc.test

Build Container Environment

When cqfd runs, a docker container is launched as the environment in which to run the command. Within this environment, commands are run as the same user as the one invoking cqfd (with a fallback to the 'builder' user in case it cannot be determined). So that this user has access to local files, the current working directory is mapped to the same location inside the container.

SSH Handling

The local ~/.ssh directory is also mapped to the corresponding directory in the build container. This effectively enables SSH agent forwarding so a build can, for example, pull authenticated git repos.

Terminal job control

When cqfd runs a command as the unprivileged user that called it in the first place, su(1) is used to run the command. This brings a limitation for processes that require a controlling terminal (such as an interactive shell), as su will prevent the command executed from having one.

$ cqfd run bash
bash: cannot set terminal process group (-1): Inappropriate ioctl for device
bash: no job control in this shell

To workaround this limitation, cqfd will use sudo(8) when it is available in the container instead. The user is responsible for including it in the related Dockerfile.

Requirements

To use cqfd, ensure the following requirements are satisfied on your workstation:

  • Bash 4.x

  • Docker

  • A docker group in your /etc/group

  • Your username is a member of the docker group

  • Restart your docker service if you needed to create the group.

Installing/removing cqfd

The cqfd script can be installed system-wide.

Install or remove the script and its resources:

$ make install
$ make uninstall

Makefile honors both PREFIX (/usr/local) and DESTDIR ([empty]) variables:

$ make install PREFIX=/opt
$ make install PREFIX=/usr DESTDIR=package

Testing cqfd (for developers)

The codebase contains tests which can be invoked using the following command, if the above requirements are met on the system:

$ make tests

Trivia

CQFD stands for "Ce qu'il fallait Dockeriser", french for "what needed to be dockerized".

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.