CONTRIBUTING.md

Contributing to pmemkv

• Opening New Issues
• Code Style
• Submitting Pull Requests
• Creating New Engines
• Creating Experimental Engines
• Extending Public API
• Configuring GitHub fork

Opening New Issues

Please log bugs or suggestions as GitHub issues. Details such as OS and PMDK version are always appreciated.

Code Style

• See .clang-format file in the repository for details
• Indent with tabs (width: 8)
• Max 90 chars per line
• Space before '*' and '&' (rather than after)

If you want to check and format your source code properly you can use CMake's DEVELOPER_MODE and CHECK_CPP_STYLE options. When enabled additional checks are switched on (cppstyle, whitespaces and headers).

cmake .. -DDEVELOPER_MODE=ON -DCHECK_CPP_STYLE=ON

If you just want to format your code you can make adequate target:

make cppformat

NOTE: We're using specific clang-format - version exactly 9.0 is required.

Submitting Pull Requests

We take outside code contributions to PMEMKV through GitHub pull requests.

If you add a new feature, implement a new engine or fix a critical bug please append appropriate entry to ChangeLog under newest release.

NOTE: If you do decide to implement code changes and contribute them, please make sure you agree your contribution can be made available under the BSD-style License used for PMEMKV.

NOTE: Submitting your changes also means that you certify the following:

Developer's Certificate of Origin 1.1

By making a contribution to this project, I certify that:

(a) The contribution was created in whole or in part by me and I
    have the right to submit it under the open source license
    indicated in the file; or

(b) The contribution is based upon previous work that, to the best
    of my knowledge, is covered under an appropriate open source
    license and I have the right under that license to submit that
    work with modifications, whether created in whole or in part
    by me, under the same open source license (unless I am
    permitted to submit under a different license), as indicated
    in the file; or

(c) The contribution was provided directly to me by some other
    person who certified (a), (b) or (c) and I have not modified
    it.

(d) I understand and agree that this project and the contribution
    are public and that a record of the contribution (including all
    personal information I submit with it, including my sign-off) is
    maintained indefinitely and may be redistributed consistent with
    this project or the open source license(s) involved.

In case of any doubt, the gatekeeper may ask you to certify the above in writing, i.e. via email or by including a Signed-off-by: line at the bottom of your commit comments.

To improve tracking of who is the author of the contribution, we kindly ask you to use your real name (not an alias) when committing your changes to PMEMKV:

Author: Random J Developer <random@developer.example.org>

Adding new dependency

Adding each new dependency (including new docker image and package) should be done in a separate commit. The commit message should be:

New dependency: dependency_name

license: SPDX license tag
origin: https://dependency_origin.com

Creating New Engines

There are several motivations to create a pmemkv storage engine:

• Using a new/different implementation strategy
• Trying out a significant change to an existing engine
• Creating a new version of an existing engine with some tweaks

Next we'll walk you through the steps of creating a new engine.

Picking Engine Name

• Relatively short (users will have to type this!)
• Formatted in all lower-case
• No whitespace or special characters
• For this example: mytree

Creating Engine Header

• Create src/engines/mytree.h header file
• For new engines, use blackhole.h as a template
• Define engine class in pmem::kv namespace
• Use snake_case for implementation class name (class my_tree)

Creating Engine Implementation

• Create src/engines/mytree.cc implementation file
• For new engines, use blackhole.cc as a template
• Use pmem::kv namespace defined by the header
• implement engine interface with its factory
• register engine factory (e.g. look at the blackhole)

Providing Unit Test

• Select testcases (based on your engine's capabilities) you wish to use from tests/engine_scenarios (or, optionally, create your own)
• Write .cmake scripts for running selected testcases and put them in tests/engines directory. If engine is based on pmemobj or memkind you can just use scripts from tests/engines/pmemobj_based or tests/engines/memkind_based
• If extraordinary tests are required, consider adding them in tests/engines
• Update tests/wrong_engine_name_test.cc test with the new engine
• In tests/CMakeLists.txt:
  • Check if all new files are covered by patterns for cppstyle and whitespace checks
  • Add selected testcases to tests/CMakeLists.txt using add_engine_test function (see for comparison, current sections for testcases of existing engines)
  • If engine-specific tests were written, build and add them separately (e.g. using build_test and add_test_generic functions defined in tests/ctest_helpers.cmake)

Updating Build System

• In CMakeLists.txt:
  • Add a build option for a new engine with a name like ENGINE_MYTREE and use it to ifdef all includes, dependencies and linking you may add
  • Add definition of the new option, like -DENGINE_MYTREE, so it can be used to ifdef engine-specific code, like:

  #ifdef ENGINE_MYTREE
  ...
  #endif
  

  • Add src/engines/mytree.h and src/engines/mytree.cc to SOURCE_FILES
  • Use pkg_check_modules and/or find_package for upstream libraries
• CMake build and make test should complete without any errors now

Updating Common Source

• In script(s) executed in CIs (at least in utils/docker/run-test-building.sh and utils/docker/images/install-bindings-dependencies.sh) add a check/build for new engine
• Build & verify engine now works with high-level bindings (see README for information on current bindings)

Documentation

• In README.md, link mytree in the table of supported engines
• Update manpages in doc directory

Creating Experimental Engines

The instructions above describe creating an engine that is considered stable. If you want, you can mark an engine as experimental and not include it in a build by default.

Experimental engines are unsupported prototypes. These engines are still being actively developed, optimized, reviewed or documented, but they are still expected to have proper automated and passing tests.

Experimental Code

There are subdirectories engines-experimental (in src and tests directories) where all experimental source files should be placed.

Experimental Build Assets

Extend existing section in CMakeLists.txt if your experimental engine requires libraries that are not yet included in the default build.

if(ENGINE_MYTREE)
    include(foo-experimental)
endif()

As noted in the example above, the experimental CMake module should use -experimental suffix in the file name.

Documentation

• In doc/ENGINES-experimental.md, add mytree section

Extending Public API

When adding a new public function, you have to make sure to update:

• manpages libpmemkv.3 or libpmemkv_config.3 in doc dir
• doc/CMakeLists.txt with new manpage link
• map file with debug symbols (src/libpmemkv.map)
• source and header files (incl. libpmemkv.h, libpmemkv.cc, libpmemkv.hpp)
• engines' sources if needed (in src/engines)
• appropriate examples, to show usage
• tests (incl. compatibility, c_api and more...)
• ChangeLog with new entry for next release

Configuring GitHub fork

To build and submit documentation as an automatically generated pull request, the repository has to be properly configured.

• Personal access token for GitHub account has to be generated.

  • Such personal access token has to be set in pmemkv repository's secrets as DOC_UPDATE_GITHUB_TOKEN variable.

• DOC_UPDATE_BOT_NAME secret variable has to be set. In most cases it will be the same as GitHub account name.

• DOC_REPO_OWNER secret variable has to be set. Name of the GitHub account, which will be target to make an automatic pull request with documentation. In most cases it will be the same as GitHub account name.

To enable automatic images pushing to GitHub Container Registry, following variables:

• CONTAINER_REG existing environment variable (defined in workflow files, in .github/ directory) has to be updated to contain proper GitHub Container Registry address (to forking user's container registry),

• GH_CR_USER secret variable has to be set up - an account (with proper permissions) to publish images to the Container Registry (tab Packages in your GH profile/organization).

• GH_CR_PAT secret variable also has to be set up - Personal Access Token (with only read & write packages permissions), to be generated as described here for selected account (user defined in above variable).