Skip to content

Latest commit

 

History

History
608 lines (455 loc) · 22.3 KB

CONTRIBUTING.md

File metadata and controls

608 lines (455 loc) · 22.3 KB

Contributing to Boost.GIL

Boost.GIL is a member of Boost libraries.

If you wish to contribute a new feature or a bug fix, please follow the workflow explained in this document.

Table of Content

Prerequisites

  • C++11 compiler
  • Build and run-time dependencies for tests and examples:
    • Boost.Filesystem
    • Boost.Test
    • Headers and libraries of libjpeg, libpng, libtiff, libraw for the I/O extension.
  • Experience with git command line basics.
  • Familiarity with build toolset and development environment of your choice.
  • Although this document tries to present all commands with necessary options, it may be a good idea to skim through the Boost Getting Started chapters, especially if you are going to use B2 for the first time.

Pull Requests

  • DO all your work in fork this repository.
  • DO base your work against the develop branch, not the master.
  • DO submit all major changes to code via pull requests (PRs) rather than through a direct commit. PRs will be CI-checked first, then reviewed and potentially merged by the repo maintainers after a peer review that includes at least one maintainer. Contributors with commit access may submit trivial patches or changes to the project infrastructure configuration via direct commits (CAUTION!)
  • DO NOT mix independent, unrelated changes in one PR. Separate unrelated fixes into separate PRs, especially if they are in different components (e.g. core headers versus extensions). Separate real product/test code changes from larger code formatting/dead code removal changes, unless the former are extensive enough to justify such refactoring, then also mention it.
  • DO start PR subject with "WIP:" tag if you submit it as "work in progress". A PR should preferably be submitted when it is considered ready for review and subsequent merging by the contributor. Otherwise, clearly indicate it is not yet ready. The "WIP:" tag will also help maintainers to label your PR with status/work-in-progress.
  • DO give PRs short-but-descriptive names (e.g. "Add test for algorithm XXX", not "Fix #1234").
  • DO refer to any relevant issues, and include the keywords that automatically close issues when the PR is merged.
  • DO mention any users that should know about and/or review the change.
  • DO ensure each commit successfully builds. The entire PR must pass all tests in the Continuous Integration (CI) system before it'll be merged.
  • DO ensure any new features or changes to existing behaviours are covered with test cases.
  • DO address PR feedback in an additional commit(s) rather than amending the existing commits. This makes it easier for reviewers to track changes.
  • DO sync your PR branch with the upstream develop branch frequently resolving any conflicts if necessary. You can either git merge upstream/develop or git rebase upstream/develop with git push --force for the latter. The merge may make it easier for reviewers to track changes though.
  • DO assume that the Squash and Merge will be used to merge your commit unless you request otherwise in the PR.
  • DO NOT submit changes to the original legacy tests, see test/legacy/README.md.

Merging Pull Requests (for maintainers with write access)

  • DO use Squash and Merge by default for individual contributions unless requested by the PR author. Do so, even if the PR contains only one commit. It creates a simpler history than Create a Merge Commit. Reasons that PR authors may request the true merge recording a merge commit may include (but are not limited to):
    • The change is easier to understand as a series of focused commits. Each commit in the series must be buildable so as not to break git bisect.
    • Contributor is using an e-mail address other than the primary GitHub address and wants that preserved in the history. Contributor must be willing to squash the commits manually before acceptance.

Getting started with Git workflow

First, you need learn some minimal basics of the modular Boost super-project workflow.

The following steps are based on the official Boost Getting Started.

NOTE: For brevity, commands below use notation for POSIX-like operating systems and you may need to tweak them for Windows systems.

1. Clone Boost super-project

The preparation involves the following steps:

  1. Clone the Boost super-project

    git clone --recurse-submodules --jobs 8 https://github.com/boostorg/boost.git
  2. Switch the Boost super-project to desired branch, master or develop

    cd boost
    git checkout master

    TIP: Modular Boost Library Maintenance guide, for more realistic test environment, recommends to develop and test individual Boost library against other Boost libraries as defined by the Boost super-project master branch:

    cd boost
    git checkout master
    git pull
    git submodule update --init --recursive --jobs 8
  3. Build the b2 driver program for B2 engine.

    ./bootstrap.sh
    ./b2 --version

    TIP: For more convenient path-less invocation, you can copy the b2 program to a location in your PATH.

  4. Create full content of /boost virtual directory with all Boost headers linked from the individual modular Boost libraries. If you skip this step, executing b2 to run tests will automatically create the directory with all headers required by Boost.GIL and tests.

    ./b2 headers

TIP: If something goes wrong, you end up with incomplete or accidentally modified files in your clone of the super-project repository, or you simply wish to start fresh, then you can clean and reset the whole repository and its modules:

git clean -xfd
git submodule foreach --recursive git clean -xfd
git reset --hard
git submodule foreach --recursive git reset --hard
git submodule update --init --recursive --jobs 8

2. Checkout Boost.GIL development branch

Regardless if you decide to develop again master (recommended) or develop branch of the Boost super-project, you should always base your contributions (i.e. topic branches) on Boost.GIL develop branch.

  1. Go to the Boost.GIL library submodule.

    cd libs/gil
  2. Checkout the develop branch and bring it up to date

    git checkout develop
    git branch -vv
    git pull origin develop

3. Fork Boost.GIL repository on GitHub

Follow Forking Projects guide to get personal copy of boostorg/gil repository from where you will be able to submit new contributions as pull requests.

Add your fork as git remote to the Boost.GIL submodule:

cd libs/gil
git remote add <username> https://github.com/<username>/gil.git

or, if you cloned from your fork already, add the upstream as origin remote:

git remote add upstream https://github.com/boostorg/gil.git
# or
git remote rename origin <username>
git remote add origin https://github.com/boostorg/gil.git

4. Submit a pull request

All Boost.GIL contributions should be developed inside a topic branch created by branching off the develop branch of boostorg/gil.

IMPORTANT: Pull Requests must come from a branch based on develop, and never on master.

NOTE: The branching workflow model Boost recommends is called Git Flow.

For example:

cd libs/gil
git checkout develop
git checkout -b feature/foo

Now, you are set to to develop a new feature for Boost.GIL, then git add and git commit your changes.

Once it's finished, you can submit it as pull request for review:

cd libs/gil
git checkout feature/foo
git push <username> feature/foo

Finally, sign in to your GitHub account and create a pull request.

Your pull request will be automatically built and tests will run on GitHub ACtions, Azure Pipelines, AppVeyor and Circle CI (see README for builds status). Please, keep an eye on those CI builds and correct any problems detected in your contribution by updating your pull request.

5. Update your pull request

Depending on actual purpose of the update, you can follow a different strategy to update your pull request:

  • Use git commit --amend, git rebase and git push --force when your pull request is still work-in-progress and not ready for review yet.
  • Use git commit, git merge and git push to update your pull request during review, in response to requests from reviewers.

NOTE: Once review of your work has started, you should not rebase your work. You should create new commits and update your topic branch. This helps with traceability in the pull request and prevents the accidental history breakage. Those who review your work may be fetching it into their fork for local review.

Synchronise pull request branch

Keep your topic branch up to date and synchronized with the upstream develop branch:

cd libs/gil
git checkout develop
git pull origin develop
git checkout feature/foo

If review of your work has not started, prefer to merge:

git merge develop
git push <username> feature/foo

If your PR is still work-in-progress, you may rebase if you like:

git rebase develop
git push --force <username> feature/foo

Amend last commit of pull request

If your pull request is a work-in-progress and has not been reviewed yet, you may amend your commit or rebase onto the develop branch:

cd libs/gil
git checkout feature/foo
git add -A
git commit --amend
git push --force <username> feature/foo

Add new commits to pull request

In order to update your pull request, for example in response to a change request from reviewer, just add new commits:

cd libs/gil
git checkout feature/foo
git add -A
git commit -m "Fix CI build failure"
git push <username> feature/foo

Development

Boost.GIL is a header-only library which does not require sources compilation. Only test runners and example programs have to be compiled.

By default, Boost.GIL uses B2 to build all the executables.

We also provide configuration for two alternative build systems:

NOTE: The CMake is optional and the corresponding build configurations for Boost.GIL do not offer equivalents for all B2 features. Most important difference to recognise is that B2 will automatically build any other Boost libraries required by Boost.GIL as dependencies.

Install dependencies

Boost.GIL tests and examples use the GIL I/O extension which depends on third-party libraries for read and write support of specific image formats:

sudo apt-get install libjpeg-dev libpng-dev libtiff5-dev libraw-dev

TIP: On Windows, use vcpkg with user-config.jam configuration provided in example/b2/user-config-windows-vcpkg.jam.

Using B2

The b2 invocation explains available options like toolset, variant and others.

Simply, just execute b2 to run all tests built using default variant=debug and default toolset determined for your development environment.

TIP: Pass b2 option -d 2 to output complete action text and commands, as they are executed. It is useful to inspect compilation flags.

If no target or directory is specified, everything in the current directory is built. For example, all Boost.GIL tests can be built and run using:

cd libs/gil
b2

Run core tests only specifying location of directory with tests:

cd libs/gil
b2 cxxstd=11 test/core

Run all tests for selected extension (from Boost root directory, as alternative):

b2 cxxstd=11 libs/gil/test/extension/dynamic_image
b2 cxxstd=11 libs/gil/test/extension/io
b2 cxxstd=11 libs/gil/test/extension/numeric
b2 cxxstd=11 libs/gil/test/extension/toolbox

Run I/O extension tests bundled in target called simple:

b2 cxxstd=11 libs/gil/test/io//simple

TIP: Pass b2 feature cxxstd=11,14,17,2a to request compilation for multiple C++ standard versions in single build run.

Using CMake

Maintainer: @mloskot

WARNING: The CMake configuration is only provided for convenience of contributors. It does not export or install any targets, deploy config files or support subproject workflow.

NOTE: CMake configuration does not build any dependencies required by Boost.GIL like Boost.Test and Boost.Filesystem libraries or any third-party image format libraries used by the I/O extension.

The provided CMake configuration allows a couple of ways to develop Boost.GIL:

  1. Using Boost installed from binary packages in default system-wide location.
  2. Using Boost installed from sources in arbitrary location (CMake may need -DBOOST_ROOT=/path/to/boost/root, see FindBoost documentation for details).
  3. Using cloned Boost super-project, inside modular libs/gil. This mode requires prior deployment of boost virtual directory with headers and stage build of required libraries, for example:

For CMake, you only need to build Boost libraries required by Boost.Test library which is used to run GIL tests. Since the CMAKE_CXX_STANDARD option in the current CMakeLists.txt defaults to C++11, pass the default cxxstd=11 to b2:

./b2 headers
./b2 variant=debug,release cxxstd=11 --with-filesystem stage

or, depending on specific requirements, more complete build:

./b2 variant=debug,release address-model=32,64 cxxstd=11 --layout=versioned --with-filesystem stage

If you wish to build tests using different C++ standard version, then adjust the cxxstd accordingly.

Using the installed Boost enables a lightweight mode for the library development, inside a stand-alone clone Boost.GIL repository and without any need to clone the whole Boost super-project.

TIP: For the lightweight setup, prefer latest release of Boost.

For available custom CMake options, open the top-level CMakeLists.txt and search for option.

Here is an example of such lightweight workflow in Linux environment (Debian-based):

  • Install required Boost libraries

    sudo apt-get update
    sudo apt-get install libboost-dev libboost-test-dev libboost-filesystem-dev
  • Clone Boost.GIL repository

    git clone https://github.com/boostorg/gil.git
    cd gil
  • Configure build with CMake

    mkdir _build
    cd _build/
    cmake ..

    TIP: By default, tests and examples are compiled using the minimum required C++11. Specify -DCMAKE_CXX_STANDARD=14|17|20 to use newer version. For more CMake options available for GIL, check option-s defined in the top-level CMakeLists.txt.

    TIP: If CMake is failing to find Boost libraries, especially built with --layout=versioned, you can try a few hacks: - option -DBoost_ARCHITECTURE=-x64 to help CMake find Boost 1.66 and above add an architecture tag to the library file names in versioned build The option added in CMake 3.13.0. - option -DBoost_COMPILER=-gcc5 or -DBoost_COMPILER=-vc141 to help CMake earlier than 3.13 match your compiler with toolset used in the Boost library file names (i.e. libboost_filesystem-gcc5-mt-x64-1_69 and not -gcc55-). Fixed in CMake 3.13.0. - if CMake is still failing to find Boost, you may try -DBoost_DEBUG=ON to get detailed diagnostics output from FindBoost.cmake module.

  • List available CMake targets

    cmake --build . --target help
  • Build selected target with CMake

    cmake --build . --target gil_test_pixel
  • List available CTest targets

    ctest --show-only | grep Test
  • Run selected test with CTest

    ctest -R gil.tests.core.pixel

CMake configuration for Visual Studio

We provide example/cmake/CMakeSettings.json with reasonable default settings for the CMake support in Visual Studio. See example/cmake/README.md for more details.

CMake configuration for Visual Studio Code

We provide example/cmake/cmake-variants.yaml with reasonable default settings for the CMake Tools extension. See example/cmake/README.md for more details.

Running clang-tidy

clang-tidy can be run on demand to diagnose or diagnose and fix or refactor source code issues.

Since the CMake configuration is provided for building tests and examples, it is easy to run clang-tidy using either the integration built-in CMake 3.6+ as target property CXX_CLANG_TIDY or the compile command database which can be easily generated.

Linting

This mode uses the CMake built-in integration and runs clang-tidy checks configured in .clang-tidy. All custom compilation warning levels (e.g. -Wall) are disabled and compiler defaults are used.

cd libs/gil
cmake -S . -B _build -DGIL_USE_CLANG_TIDY=ON

# all targets
cmake --build _build

# selected target
cmake --build _build --target test_headers_all_in_one

Refactoring

WARNING: This is advanced processing and depending on checks, it may fail to deliver expected results, especially if run against all configured translation units at ones.

  1. Generate compile_commands.json database

    cd libs/gil
    cmake -S . -B _build -DCMAKE_EXPORT_COMPILE_COMMANDS=ON
  2. Edit compile_commands.json and remove entries of commands for all but the .cpp files you wish to refactor. For example, keep test_headers_all_in_one.cpp only to refactor all headers.

  3. Run the parallel clang-tidy runner script to apply the desired checks (and fixes) across the library source code:

    run-clang-tidy.py -p=_build -header-filter='boost\/gil\/.*' -checks='-*,modernize-use-using' -fix > cl.log 2>&1

Guidelines

Boost.GIL is a more than a decade old mature library maintained by several developers with help from a couple of dozens contributors. It is important to maintain consistent design, look and feel. Thus, below a few basic guidelines are listed.

First and foremost, make sure you are familiar with the official Boost Library Requirements and Guidelines.

Second, strive for writing idiomatic C++11, clean and elegant code.

NOTE: The Boost.GIL source code does not necessary represent clean and elegant code to look up to. The library has recently entered the transition to C++11. Major refactoring overhaul is ongoing.

Maintain structure your source code files according to the following guidelines:

  • Name files in meaningful way.

  • Put copyright and license information in every file

  • If your changes meet a certain threshold of originality, add yourself to the copyright notice. Do not put any additional authorship or file comments (eg. no \file for Doxygen), revision information, etc.

  • In header, put #include guard based on header path and file name

    #ifndef BOOST_GIL_<DIR1>_<DIR2>_<FILE>_HPP
    #define BOOST_GIL_<DIR1>_<DIR2>_<FILE>_HPP
    ...
    #endif
  • Make sure each header is self-contained, i.e. that they include all headers they need.

  • All public headers should be placed in boost/gil/ or boost/gil/<component>/.

  • All non-public headers should be placed boost/gil/detail or boost/gil/<component>/detail.

  • All public definitions should reside in scope of namespace boost { namespace gil {...}}.

  • All non-public definitions should reside in scope of namespace boost { namespace gil { namespace detail {...}}}.

  • Write your code to fit within 100 columns of text.

  • Use EditorConfig for your editor and enable .editorconfig to: - Indent with 4 spaces and no tabs. - Trim any trailing whitespaces.

  • Do not increases the indentation level within namespace.