Add boiler point for state machine #31
Conversation
Mchan2003
force-pushed
the
Michael-CLI-FSM
branch
from
3e30222 to
536a194
Compare
PhazonicRidley
left a comment
PhazonicRidley
left a comment
There was a problem hiding this comment.
Overall good! One small design question, is there a reason you are using raw pointers for your states? Do they need to stay alive after exiting? If so, is there a reason you are using raw pointers over smart pointers
| @@ -0,0 +1,52 @@ | |||
| #pragma once | |||
| #include <print> | |||
| #include "./state.hpp" | |||
There was a problem hiding this comment.
Nit: include/ should be in the include path (this functionally doesn't matter but imo its cleaner without relative paths, also if we ever change our file structure then it doesn't need to be updated)
| * SAFE will be stored here and shared between the states. | ||
| * | ||
| */ | ||
| class StateContext { |
There was a problem hiding this comment.
Please add doxygen style documentation (our good friend chaudexityilot is great at this)
|
|
||
| class State; | ||
|
|
||
| class StateMachine { |
| void StateMachine::run_state() { | ||
| current_state->enter(context); | ||
| current_state->handle(context); | ||
| transition_state (current_state->exit(context)); |
There was a problem hiding this comment.
Prev state is dangling and is never freed
|
|
||
| class StateMachine { | ||
| public: | ||
| StateMachine(); |
There was a problem hiding this comment.
Nit: No need to declare a default ctor or dtor
Fixes #33