Software Technology Lab (STLab) Library Source Code Repository

ASL libraries will be migrated here in the stlab namespace, new libraries will be created here.

Branch states

• main:

Content

Concurrency

This library provides futures and channels, high level abstractions for implementing algorithms that eases the use of multiple CPU cores while minimizing contention. This library solves several problems of the C++11 and C++17 TS futures.

Documentation

The complete documentation is available on the STLab home page.

Release changelogs are listed in CHANGES.md.

Tested on

• Linux with GCC 11
• Linux with Clang 14
• MacOS 11 with Apple-clang 13.0.0
• Windows with Visual Studio 16

Requirements

• A standards-compliant C++17, C++20, or C++23 compiler
• Building the library requires CMake 3.23 or later
• Testing or developing the library requires Boost.Test >= 1.74.0

Building

STLab is a standard CMake project. See the running CMake tutorial for an introduction to this tool.

Preparation

1. Create a build directory outside this library's source tree. In this guide, we'll use a sibling directory called BUILD.

2. Install a version of CMake >= 3.23. If you are on Debian or Ubuntu Linux you may need to use snap to find one that's new enough.

3. If you are using MSVC, you may need to set environment variables appropriately for your target architecture by invoking VCVARSALL.BAT with an appropriate option.

Configure

Run CMake in the root directory of this project, setting ../BUILD as your build directory. The basis of your command will be

cmake -S . -B ../BUILD -DCMAKE_BUILD_TYPE=# SEE BELOW

but there are other options you may need to append in order to be successful. Among them:

• -DCMAKE_BUILD_TYPE=[Release|Debug] to build the given configuration (required unless you're using visual studio or another multi-config generator).
• -DCMAKE_CXX_STANDARD=[17|20|23] to build with compliance to the given C++ standard.
• -DBUILD_TESTING=OFF if you only intend to build, but not test, this library.
• -DBoost_USE_STATIC_LIBS=TRUE if you will be testing on Windows.

We also suggest the installation of Ninja and its use by adding -GNinja to your cmake command line… but ninja is not required.

A typical invocation might look like this:

cmake -S . -B ../BUILD -GNinja -DCMAKE_CXX_STANDARD=17 -DCMAKE_BUILD_TYPE=Release -DBUILD_TESTING=OFF

If you organize the build directory into subdirectories you can support multiple configurations.

rm -rf ../builds/portable
cmake -S . -B ../builds/portable -GXcode -DCMAKE_CXX_STANDARD=17 -DBUILD_TESTING=ON -DSTLAB_TASK_SYSTEM=portable -DCMAKE_OSX_DEPLOYMENT_TARGET=macosx14.4

Build

If your configuration command was successful, go to your build directory (cd ../BUILD) and invoke:

cmake --build .

Testing

Running the tests in the BUILD directory is as simple as invoking

ctest -C Debug

or

ctest -C Release

depending on which configuration (CMAKE_BUILD_TYPE) you chose to build.

Generating Documentation

STLab uses hyde to generate documentation boilerplate by scanning its public headers. You will need to follow Hyde's installation instructions and have hyde on your $PATH (detectable by which hyde).

When adding a new public API, you should invoke

docker run --platform linux/x86_64 --mount type=bind,source="$(pwd)/..",target=/mnt/host --tty --interactive hyde bash

cd /mnt/host/libraries
./validate_docs.sh --update

(or, simply -u) to generate the boilerplate for it. Then, fill in any fields marked as __MISSING__. Fields marked as __OPTIONAL__ may be omitted.