# Development#

Written by Allan Leal (ETH Zurich) on Nov 18th, 2022

Here we provide tips in case you want to help with Reaktoro development. Reaktoro is primarily developed using Visual Studio Code on Linux. If you are on Windows, consider using the Windows Subsystem for Linux (WSL) with Ubuntu by following these installation instructions provided by Canonical®.

## Launching Visual Studio Code#

If you are developing Reaktoro, you should use conda as a package and dependency manager. The entire process of installing conda and creating a conda environment called reaktoro containing all Reaktoro dependencies can be found in the Installation using CMake instructions.

To ensure that Visual Studio Code works correctly with the reaktoro conda environment, it is recommended to launch it from the terminal with the reaktoro environment activated. In a terminal, go to the directory where the Reaktoro source code was git cloned, activate the reaktoro conda environment and launch Visual Studio Code using code .. See the complete sequence of commands below:

cd path/to/reaktoro
conda activate reaktoro
code .


Note

By launching Visual Studio Code in this way, it will be aware of the environment variables present when the conda environemt reaktoro is activated, and this will help with both IntelliSense, debugging, and executing Python scripts using Reaktoro.

## Ensuring IntelliSense works correctly#

Unfortunately, Visual Studio Code doesn’t automatically detect where all of Reaktoro’s C++ dependencies are located, and you might get error squiggles all over the place because it can’t find header files for those dependencies. To fix this, you should press Ctrl+Shift+P and select C/C++: Edit Configurations (JSON). In the created C/C++ configuration file, consider:

{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/", "${env:CONDA_PREFIX}/**"
],
"defines": [
"AUTODIFF_ENABLE_IMPLICIT_CONVERSION_REAL=1",
"REAKTORO_EMBEDDED_DIR=${workspaceFolder}/embedded", "REAKTORO_DATABASES_DIR=${workspaceFolder}/embedded/databases",
"REAKTORO_PARAMS_DIR=${workspaceFolder}/embedded/params" ], "compilerPath": "${env:CXX}",
"cStandard": "c17",
"cppStandard": "c++17",
"intelliSenseMode": "linux-gcc-x64",
"compileCommands": "${workspaceFolder}/compile_commands.json" } ], "version": 4 }  This configuration is telling Visual Studio Code to find header files under ${env:CONDA_PREFIX}, which resolves to /home/username/miniconda3/envs/reaktoro. It also tells Visual Studio Code to use the C++ compiler available in the conda environment, ${env:CXX}, instead of the default compiler available on the system. Note This is one of the reasons why it’s important to launch Visual Studio Code from the terminal with the reaktoro environment enabled. By doing so, it knows the CONDA_PREFIX and CXX environment variables used in the above configuration. ## Creating compilation tasks for Reaktoro# If you are looking for Ctrl+Shift+B to trigger a Reaktoro build from within Visual Studio Code, then you need a build task. Press Ctrl+Shift+P and select Tasks: Configure Default Build Tasks, then select Create tasks.json file from template and select Others. Replace the contents of the created tasks.json file with: { "version": "2.0.0", "tasks": [ { "label": "Build", "type": "shell", "command": "clear && make -j", "options": { "cwd": "${workspaceFolder}/build/debug/gcc"
},
"group": {
"kind": "build",
"isDefault": true
},
"problemMatcher": [
"$gcc" ] }, { "label": "Build & Test", "type": "shell", "command": "clear && make -j tests", "options": { "cwd": "${workspaceFolder}/build/debug/gcc"
},
"problemMatcher": [
"$gcc" ] } ] }  The configuration above assumes that you have created a  build/debug/gcc directory and executed cmake ../../.. -DCMAKE_BUILD_TYPE=Debug from inside of it: mkdir -p build/debug/gcc cmake ../../.. -DCMAKE_BUILD_TYPE=Debug  Tip Consider using the LLVM linker lld, which is faster than the default GNU linker ld, when configuring the build of Reaktoro (for either debug or release modes): cmake ../../.. -DCMAKE_BUILD_TYPE=Debug -DCMAKE_CXX_FLAGS="-fuse-ld=lld"  ## Creating run and debug tasks for Reaktoro’s tests and examples# When doing Reaktoro development, you may want to debug: Pressing Ctrl+Shift+D in Visual Studio Code allows you to execute and debug an application. You’ll need first to click on create a launch.json and then select any shown option. Once the launch.json file is created, replace its content by: { "version": "0.2.0", "configurations": [ { "name": "Debug C++ Tests", "type": "cppdbg", "request": "launch", "program": "${workspaceFolder}/build/debug/gcc/Reaktoro/reaktoro-cpptests",
"args": ["[ChemicalSystem]"],
"stopAtEntry": false,
"cwd": "${workspaceFolder}", "environment": [], "MIMode": "gdb", "setupCommands": [ { "description": "Enable pretty-printing for gdb", "text": "-enable-pretty-printing", "ignoreFailures": true } ] }, { "name": "Debug C++ Example", "type": "cppdbg", "request": "launch", "program": "${workspaceFolder}/build/debug/gcc/examples/cpp/${fileBasenameNoExtension}", "preLaunchTask": "Build", "stopAtEntry": false, "cwd": "${workspaceFolder}",
"environment": [],
"MIMode": "gdb",
"setupCommands": [
{
"description": "Enable pretty-printing for gdb",
"text": "-enable-pretty-printing",
"ignoreFailures": true
}
]
},
{
"name": "Debug Python Tests",
"type": "cppdbg",
"request": "launch",
"program": "${env:CONDA_PREFIX}/bin/python", "args": ["-m", "pytest", "Reaktoro/Core/ChemicalSystem.py"], "preLaunchTask": "Build", "stopAtEntry": false, "cwd": "${workspaceFolder}",
"environment": [
{
"name": "PYTHONPATH",
"value": "${workspaceFolder}/build/debug/gcc/python/package/build/lib:${env:PYTHONPATH}"
}
],
"MIMode": "gdb",
"setupCommands": [
{
"description": "Enable pretty-printing for gdb",
"text": "-enable-pretty-printing",
"ignoreFailures": true
}
]
},
{
"name": "Debug Python/C++ Script",
"type": "cppdbg",
"request": "launch",
"program": "${env:CONDA_PREFIX}/bin/python", "args": ["${file}"],
"stopAtEntry": false,
"cwd": "${workspaceFolder}", "environment": [ { "name": "PYTHONPATH", "value": "${workspaceFolder}/build/debug/gcc/python/package/build/lib:${env:PYTHONPATH}" } ], "MIMode": "gdb", "setupCommands": [ { "description": "Enable pretty-printing for gdb", "text": "-enable-pretty-printing", "ignoreFailures": true } ] } ] }  When you press Ctrl+Shift+D, you are taken to the Run and Debug pane. On the top, you should now have a checkbox showing the options: Now you can select one of them and start debugging. You can adjust the above configuration file as needed. Debugging different C++ tests In the Debug C++ Tests configuration, replace [ChemicalSystem] by [EquilibriumSolver] or any other C++ test. Debugging different C++ examples Make sure you have the C++ example open in the editor and press F5. Alternatively, replace ${fileBasenameNoExtension} in the Debug C++ Example configuration above with the name of the C++ example you want to debug so that you don’t need to have the C++ file open at the time F5 is pressed.

Debugging different Python tests

In the Debug Python Tests configuration, replace Reaktoro/Core/ChemicalSystem.py by Reaktoro/Equilibrium/EquilibriumSolver.py or any other Python test file.

Debugging different Python examples

Make sure you have the Python example open in the editor and press F5. Alternatively, replace \${file} in the Debug Python/C++ Script configuration above with the path to the Python file you want to debug so that you don’t need to have the Python file open at the time F5 is pressed.

Attention

Note that this actually debugs the Reaktoro C++ extension for Python. In other words, this configuration allows you to debug Reaktoro C++ code that is running from a Python script. This is not for debugging a Python script!

And that’s it for now. More development recommendations will be added in the futur as needed. If you are encountering issues with the tips above, please get in touch.