# Contributing Feedback and contributions are very welcome! Here's help on how to make contributions, divided into the following sections: * general information, * vulnerability reporting, * code changes, * documentation, * how to check proposed changes before submitting them, * reuse (supply chain for third-party components, including updating them), and * keeping up with external changes. ## General information For specific proposals, please provide them as [pull requests](https://github.com/QMatrixClient/libqmatrixclient/pulls) or [issues](https://github.com/QMatrixClient/libqmatrxclient/issues) For general discussion, feel free to use our Matrix room: [#quaternion:matrix.org](https://matrix.to/#/#quaternion:matrix.org). If you're new to the project (or FLOSS in general), [issues tagged as simple](https://github.com/QMatrixClient/libqmatrixclient/labels/simple) are smaller tasks that may typically take 1-3 days. You are welcome aboard! ### Pull requests and different branches recommended Pull requests are preferred, since they are specific. See the GitHub Help [articles about pull requests](https://help.github.com/articles/using-pull-requests/) to learn how to deal with them. We recommend creating different branches for different (logical) changes, and creating a pull request when you're done into the master branch. See the GitHub documentation on [creating branches](https://help.github.com/articles/creating-and-deleting-branches-within-your-repository/) and [using pull requests](https://help.github.com/articles/using-pull-requests/). ### How we handle proposals We use GitHub to track all changes via its [issue tracker](https://github.com/QMatrixClient/libqmatrixclient/issues) and [pull requests](https://github.com/QMatrixClient/libqmatrixclient/pulls). Specific changes are proposed using those mechanisms. Issues are assigned to an individual who works on it and then marks it complete. If there are questions or objections, the conversation area of that issue or pull request is used to resolve it. ### License Unless a contributor explicitly specifies otherwise, we assume that all contributed code is released under [the same license as libqmatrixclient itself](./COPYING), which is LGPL v2.1 as of the time of this writing. Any components proposed for reuse should have a license that permits releasing a derivative work under LGPL v2.1. Moreover, the license of a proposed component should be approved by OSI, no exceptions. ## Vulnerability reporting (security issues) If you find a significant vulnerability, or evidence of one, use either of the following contacts: * send an email to Kitsune Ral [Kitsune-Ral@users.sf.net](mailto:Kitsune-Ral@users.sf.net) * reach out in Matrix to #kitsune:matrix.org (if you can, switch encryption **on**) In any of these two options, _indicate that you have such information_ (do not share the information yet), and we'll tell you the next steps. We will gladly give credit to anyone who reports a vulnerability so that we can fix it. If you want to remain anonymous or pseudonymous instead, please let us know that; we will gladly respect your wishes. ## Code changes The code should strive to be DRY (don't repeat yourself), clear, and obviously correct. Some technical debt is inevitable, just don't bankrupt us with it. Refactoring is welcome. ### Library API and doc-comments Whenever you add a new call to the library API that you expect to be used from client code, you must supply a proper doc-comment along with the call. Doxygen (with backslashes) style is preferred. You can find that some parts of the code still use JavaDoc (with @'s) style; feel free to replace it with Doxygen backslashes if that bothers you. Calls, data structures and other symbols not intended for use by clients should _not_ be exposed in (public) .h files, unless they are necessary to declare other public symbols. In particular, this involves private members (functions, typedefs, or variables) in public classes; use pimpl idiom to hide implementation details as much as possible. Note: As of now, all header files of libqmatrixclient are considered public; this may change eventually. ### Qt-flavoured C++ This is our primary language. We don't have a particular code style _as of yet_ but some rules-of-thumb are below: * 4-space indents, no tabs, no trailing spaces, no last empty lines. If you spot the code abusing these - we'll thank you for fixing it. * Lines within 80 characters are preferred. * Braces after if's, while's, do's, function signatures etc. take a separate line. * A historical deviation from the usual Qt code format conventions is an extra indent inside _classes_ (access specifiers go at +4 spaces to the base, members at +8 spaces) but not _structs_ (members at +4 spaces). This may change in the future for something more conventional. * Please don't make hypocritical structs with protected or private members. Just make them classes instead. * Qt containers are generally preferred to STL containers; however, there are notable exceptions, and libqmatrixclient already uses them: * `std::array` and `std::deque` have no direct counterparts in Qt. * Because of COW semantics, Qt containers cannot hold uncopyable classes. Classes without a default constructor are a problem too. Examples of that are `SyncRoomData` and `EventsBatch<>`. Use STL containers for those but see the next point and also consider if you can supply a reasonable copy/default constructor. * STL containers can be freely used in code internal to a translation unit (i.e., in a certain .cpp file) _as long as that is not exposed in the API_. It's ok to use, e.g., `std::vector` instead of `QVector` to tighten up code where you don't need COW, or when dealing with uncopyable data structures (see the previous point). However, exposing STL containers in the API is not encouraged (except where absolutely necessary, e.g. we use `std::deque` for a timeline). Exposing STL containers or iterators in API intended for usage by QML code (e.g. in `Q_PROPERTY`) will not work at all and will never be accepted. * Use `QVector` instead of `QList` where possible - see a [great article by Marc Mutz on Qt containers](https://marcmutz.wordpress.com/effective-qt/containers/) for details. ### Automated tests There's no testing framework as of now; either Catch or Qt Test or both will be used eventually. However, as a stopgap measure, qmc-example is used for automated end-to-end testing. Any significant addition to the library API should be accompanied by a respective test in qmc-example. To add a test you should write a new private slot with the test logic to the `QMCTest` class and call it from `QMCTest::doTests()`. `QMCTest` sets up some basic test fixture to help you with testing; notably by the moment `doTests()` is invoked you can rely on having a working connection in `c` member variable and a test room in `targetRoom` member variable. PRs to introduce a proper testing framework are very welcome (make sure to migrate tests from qmc-example though); shifting qmc-example to use Qt Test seems to be a particularly low-hanging fruit. ### Security, privacy, and performance Pay attention to security, and work *with* (not against) the usual security hardening mechanisms (however few in C++). `char *` and similar unchecked C-style read/write arrays are forbidden - use Qt containers or at the very least `std::array<>` instead. Where you see fit (usually with data structures), try to use smart pointers, especially `std::unique_ptr<>` or `QScopedPointer` instead of bare pointers. When dealing with `QObject`s, use the parent-child ownership semantics exercised by Qt. Shared pointers are not used in the code so far; but if you find a particular use case where strict semantics of unique pointers doesn't help and a shared pointer is necessary, feel free to step up with the working code and it will be considered for inclusion. Exercise the [principle of least privilege](https://en.wikipedia.org/wiki/Principle_of_least_privilege) where reasonable and appropriate. Prefer less-coupled cohesive code. Protect private information, in particular passwords and email addresses. Do not forget about local access to data (in particular, be very careful when storing something in temporary files, let alone permanent configuration or state). Avoid mechanisms that could be used for tracking where possible (we do need to verify people are logged in but that's pretty much it), and ensure that third parties can't use interactions for tracking. We want the software to have decent performance for typical users. At the same time we keep libqmatrixclient single-threaded as much as possible, to keep the code simple. That means being cautious about operation complexity (read about big-O notation if you need a kickstart on the topic). This especially refers to operations on the whole timeline and the list of users - each of these can have tens of thousands of elements so even operations with linear complexity, if heavy enough, can produce noticeable GUI freezing. A solution to such cases is to embed `processEvents()` invocations in heavy loops (see `Connection::saveState()` to get the idea). Having said that, there's always a trade-off between various attributes; in particular, readability and maintainability of the code is more important than squeezing every bit out of that clumsy algorithm. Beware of premature optimization and have profiling information around if you are about to go into some hardcore optimization. ## Documentation changes Most of the documentation is in "markdown" format. All markdown files use the .md filename extension. Where reasonable, limit yourself to Markdown that will be accepted by different markdown processors (e.g., what is specified by CommonMark or the original Markdown). In practice we use the version of Markdown implemented by GitHub when it renders .md files, and you can use its extensions (in particular, mark code snippets with the programming language used). This version of markdown is sometimes called [GitHub-flavored markdown](https://help.github.com/articles/github-flavored-markdown/). In particular, blank lines separate paragraphs; newlines inside a paragraph do *not* force a line break. Beware - this is *not* the same markdown algorithm used by GitHub when it renders issue or pull comments; in those cases [newlines in paragraph-like content are considered as real line breaks](https://help.github.com/articles/writing-on-github/); unfortunately this other algorithm is *also* called GitHub rendered markdown. (Yes, it'd be better if there were standard different names for different things.) In your markdown, please don't use tab characters and avoid "bare" URLs (in a hypertext link, the link text and URL should be on the same line). We do not care about the line length in markdown texts (and more often than not put the whole paragraph into one line). This is actually negotiable, and absolutely not enforceable. If you want to fit in a 70-character limit, go ahead, just don't reformat the whole text along the way. Take care to not break URLs when breaking lines. Do not use trailing two spaces for line breaks, since these cannot be seen and may be silently removed by some tools. Instead, use <br /> (an HTML break). ## How to check proposed changes before submitting them Checking the code on at least one configuration is essential; if you only have a hasty fix that doesn't even compile, better make an issue and put a link to your commit into it (with an explanation what it is about and why). The best check is a continuous integration pipeline automatically triggered when you submit a PR. ### Standard checks The following warnings configuration is applied with GCC and Clang when using CMake: `-W -Wall -Wextra -pedantic -Werror=return-type -Wno-unused-parameter -Wno-gnu-zero-variadic-macro-arguments` (the last one is to mute a warning triggered by Qt code for debug logging). We don't turn most of the warnings to errors but please treat them as such. In Qt Creator, the following line can be used with the Clang code model (you should explicitly enable the code model plugin): `-Weverything -Werror=return-type -Wno-c++98-compat -Wno-c++98-compat-pedantic -Wno-unused-macros -Wno-newline-eof -Wno-exit-time-destructors -Wno-global-constructors -Wno-gnu-zero-variadic-macro-arguments -Wno-documentation -Wno-missing-prototypes -Wno-shadow-field-in-constructor -Wno-padded -Wno-weak-vtables` ### Continuous Integration We use Travis CI to check buildability on Linux (GCC, Clang) and MacOS (Clang), and AppVeyor CI to build on Windows (MSVC). Every PR will go through these, and you'll see the traffic lights from them on the PR page. Failure on any platform will most likely entail a request to you for a fix before merging a PR. ### Other tools If you know how to use clang-tidy, here's a list of checks we do and do not use (a leading hyphen means a disabled check, an asterisk is a wildcard): `*,cert-env33-c,-cppcoreguidelines-pro-bounds-array-to-pointer-decay,-cppcoreguidelines-pro-bounds-constant-array-index,-cppcoreguidelines-pro-bounds-pointer-arithmetic,-cppcoreguidelines-pro-type-const-cast,-cppcoreguidelines-pro-type-union-access,-cppcoreguidelines-special-member-functions,-google-build-using-namespace,-google-readability-braces-around-statements,-hicpp-*,-llvm-*,-misc-unused-parameters,-misc-noexcept-moveconstructor,-modernize-use-using,-readability-braces-around-statements,readability-identifier-naming,-readability-implicit-bool-cast,-clang-diagnostic-*,-clang-analyzer-*`. If you're on CLion, you can simply copy-paste the above list into the Clang-Tidy inspection configuration. In Qt Creator 4.6, one can enable clang-tidy and clazy (clazy level 1 eats away CPU but produces some very relevant and unobvious notices, such as possible unintended copying of a Qt container, or unguarded null pointers). ## Git commit messages When writing git commit messages, try to follow the guidelines in [How to Write a Git Commit Message](https://chris.beams.io/posts/git-commit/): 1. Separate subject from body with a blank line 2. Be reasonable on the subject line length, because this is what we see in commit logs. Try to fit in 50 characters whenever possible. 3. Capitalize the subject line 4. Do not end the subject line with a period 5. Use the imperative mood in the subject line (*command* form) (we don't always practice this ourselves but let's try). 6. Use the body to explain what and why vs. how (git tracks how it was changed in detail, don't repeat that). Sometimes a quick overview of "how" is acceptable if a commit is huge - but maybe split a commit into smaller ones, to begin with? ## Reuse (libraries, frameworks, etc.) C++ is unfortunately not very coherent about SDK/package management, and we try to keep building the library as easy as possible. Because of that we are very (and it means _very_) conservative about adding dependencies to libqmatrixclient. That relates to both additional Qt components and even more to other libraries. Fortunately, even the Qt components now in use (Qt Core and Network) are very feature-rich and provide plenty of ready-made stuff. Regardless of the above paragraph (and as mentioned earlier in the text), we're now looking at possible options for automated testing, so PRs onboarding a test framework will be considered with much gratitude. Some cases need additional explanation: * Before rolling out your own super-optimised container or algorithm written from scratch, take a good long look through documentation on Qt and C++ standard library (including the experimental/future sections). Please try to reuse the existing facilities as much as possible. * You should have a good reason (or better several ones) to add a component from KDE Frameworks. We don't rule this out and there's no prejudice against KDE; it just so happened that KDE Frameworks is one of most obvious reuse candidates but so far none of these components survived as libqmatrixclient deps. So we are cautious. * Never forget that libqmatrixclient is aimed to be a non-visual library; QtGui in dependencies is only driven by (entirely offscreen) dealing with QPixmaps. While there's a bunch of visual code (in C++ and QML) shared between libqmatrixclient-enabled _applications_, this is likely to end up in a separate (libqmatrixclient-enabled) library, rather than libqmatrixclient. ## Attribution This text is largely based on CONTRIBUTING.md from CII Best Practices Badge project, which is a collective work of its contributors (many thanks!). The text itself is licensed under CC-BY-4.0.