Native function interface.
To use your own native functions within LocalSolver, you have to first define your native function in one of two ways:
- Implement the LSNativeFunction interface with the call method. The call method must take the native context associated to the function and must return a double value. This native context contains the values of the expressions passed to the function. A distinction is made between integer arguments (bool, int) and floating point arguments (double)
- If your program is compiled with C++11, you can use a function, a functor or a lambda function instead. This function may take any number of lsint and lsdouble arguments, or alternatively a single native context. It must return a double value.
- Instanciate the function as an LSExpression with LSModel#createNativeFunction or the dedicated shortcut LSModel#nativeFunction.
- Pass arguments to your function and call it using LSModel#call. The first operand must the LSExpression returned by createNativeFunction. The other operands must be LSExpressions, whose values will be made accessible to your native function when it is called.
Note 1: Most of the time your native function will be called when the solver is in state S_Running. Do not attempt to call any method of the solver (to retrieve statistics, values of LSExpressions or whatever) in that state or an exception will be thrown. The only accessible function is LocalSolver#stop().
Note 2: Your functions must be thread-safe. According to the “nbThreads” parameter, LocalSolver can be multi-threaded. In that case, your native functions must be thread safe. If you cannot guarantee the thread-safety of your code, we strongly recommend you to limit the search of LocalSolver to one thread with LSParam#setNbThreads.
Note 3: LocalSolver does not manage memory of objects created outside of its environment. You are responsible for the lifetime of your LSNativeFunction, which must last as long as the search is active.