High-Performance Flat Finite State Machine

Header-only flat FSM framework in C++11, with fully statically-defined structure (no dynamic allocations), built with variadic templates.

Compiler Support

Visual Studio: 2015, 2017, 2019, 2022
GCC: 5, 6, 7, 8, 9, 10, 11, 12
ARM GCC: 9
Clang: 3.9, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
AppleClang: 12, 13, 13

(Currently CI-tested toolchains in bold)

Quick Start

Minimal Example
Extended Tutorial

Documentation

Wiki

Feature Highlights

Permissive MIT License
Written in widely-supported modern(ish) C++11
Header-only
Convenient, minimal boilerplate
Fully static, no dynamic allocations
Uses inline-friendly compile-time polymorphism, no virtual methods are used
Type-safe transitions: FSM.changeTo<TargetState>() with optional payloads
Flexible configuration using FFSM2_ENABLE_* macros
Scalable, supports robust state re-use via state injections
Gamedev-friendly, supports explicit State::update()
Also supports traditional event-based workflow with State::react()
Inspect anything: previous and current transitions, state activation status, and more!
Game AI-friendly with dynamic planning support
Debug-assisted, includes automatic structure and activity visualization API
Built-in logging support

3rd Party Libraries

doctest unit testing framework

Get In Touch

Twitter: @andrew_gresyk
Discord: HFSM.dev
Gitter: andrew-gresyk/FFSM2

Special Thanks

Alexis Paques

16-bit architectures support

FFSM2 single-header library does not compile on 16-bits architectures.

Description

FFSM2 cannot be compiled on a 16-bits architecture because the following code generates a compile-time error:

#if INTPTR_MAX == INT64_MAX
    #define FFSM2_ARCHITECTURE_64()    true
    #define FFSM2_ARCHITECTURE_32()    false

    #define FFSM2_64BIT_OR_32BIT(p64, p32)    p64
#elif INTPTR_MAX == INT32_MAX
    #define FFSM2_ARCHITECTURE_64()    false
    #define FFSM2_ARCHITECTURE_32()    true

    #define FFSM2_64BIT_OR_32BIT(p64, p32)    p32
#else
    #error
#endif

The case for INTPTR_MAX == INT16_MAX is missing.

Possible Fix

Doing a simple find in the code, neither FFSM2_64BIT_OR_32BIT nor FFSM2_ARCHITECTURE_32/64 defines seems to be used.
Commenting out those defines the code seems to compile just fine. Are they really needed?

Steps to Reproduce

Unfortunately, the compiler I'm using is not publicly available right now. You could reproduce the error by redefining INTPTR_MAX as INT16_MAX.

Environment

Version used (clang --version): clang version 14.0.4
Operating System and version: not relevant

andrew-gresyk / ffsm2 Goto Github PK

ffsm2's Introduction

High-Performance Flat Finite State Machine

Compiler Support

Quick Start

See Also

Documentation

Feature Highlights

3rd Party Libraries

Get In Touch

Special Thanks

ffsm2's People

Contributors

Stargazers

Watchers

Forkers

ffsm2's Issues

Description

Possible Fix

Steps to Reproduce

Environment

Recommend Projects

Recommend Topics

Recommend Org