Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Designing state machines

drivy.engineering

Post reply on HN
Print view View original

31–40 of 58 posts

Re: Designing state machines

#31

Post by bri3d »

activatedgeek wrote: post #22

While I understand the importance of state machines and its direct usage in complex Business Logic Workflows (and indirectly almost everywhere), there is nothing special in this post. It is a generic comment on State Machines. Am I missing something here?

A surprising number of entrepreneurs and developers are unfamiliar with the application of state machines to business logic workflows, so it's useful to see the topic presented in a succinct way. Good learning content tends to get upvoted on HN, especially when the discussion around the concept (for example, the discussion of State Machines and state machine libraries in this thread) is interesting as well.

Re: Designing state machines

#32

Post by ge96 »

activatedgeek wrote: post #22

While I understand the importance of state machines and its direct usage in complex Business Logic Workflows (and indirectly almost everywhere), there is nothing special in this post. It is a generic comment on State Machines. Am I missing something here?

Is "state machines" a global term that includes finite state machines like how vending machines deal with varying coin amounts that go in to trigger the price and dispense proper change?

Googled looks like it is. That was a cool homework/project.

Re: Designing state machines

#33

Post by carapace »

siddhant wrote: post #14

In case you're using Python, I can't recommend this library enough - https://github.com/pytransitions/transitions . An absolute pleasure to work with.

For goodness' sake, no! I just ripped that package out of our production code. It's terrible! It adds methods to your object at runtime, crazy stuff.

I replaced it with just this:

        class StateMachine(object):
	    '''
	    A very simple FSM class.

	    Params:
	        initial: Initial state
	        table: A dict (current, event) -> target
	    '''

	    def __init__(self, initial, table):
	        self.current_state = initial
	        self.state_table = table

	    def __call__(self, event):
	        '''Trigger one state transition.'''
	        self.current_state = self.state_table[self.current_state, event]


	class Foo(object):

	    STATE_TABLE = {
		(current_state, event): next_state,
		...
	    }

	    def __init__(self, xid, token, config):
		self.fsm = StateMachine('start', self.STATE_TABLE)

	    def begin(self):
	        while self.fsm.current_state not in {'success', 'error'}:
	            method = getattr(self, self.fsm.current_state)
	            self.fsm(method())
	        if self.fsm.current_state == 'error':
	            self.report_error()
If you're using Python anything more involved than a dict mapping (current, transition) -> next is just overkill. Use it, don't abuse it. ;-)

Re: Designing state machines

#34

Post by nerdponx »

karboosx wrote: post #10
nerdponx wrote: post #4

Earlier quoted context omitted.

I wanted to like Mermaid, then I realized it was just a neutered version of GraphViz. I had wanted to write a Mermaid-to-Graphviz compiler but never got around to it. It produces pretty ugly graphs that can't be exported to any other format. Meanwhile just about ever graph manipulation program (e.g. Gephi) can handle GraphViz DOT format.

I didn't know GraphViz. I have to try it. DOT language looks very similar to Mermaid. It was too hard to make that compiler or you have encountered another problem?

I didn't even know what BNF was when I had the idea ;)

It's still on my list of short projects. I have a lot of learning to do, knowing nothing about writing parsers and such. Yes, I know I could probably hack it together with a bunch of garbage Python. But that just does not sound fun to write, let alone maintain.

Re: Designing state machines

#35

Post by arambhashura »

siddhant wrote: post #14

In case you're using Python, I can't recommend this library enough - https://github.com/pytransitions/transitions . An absolute pleasure to work with.

SMC (State Machine Compiler - http://smc.sourceforge.net/) supports multiple languages and is fairly simple to learn and use.

Re: Designing state machines

#36

Post by idbfs »

A paper that changed my approach to designing state machines is "Statecharts: A Visual Formalism for Complex Systems" by Harel (http://www.wisdom.weizmann.ac.il/~harel/papers/Statecharts.p...).

Statecharts (also called hierarchical state machines) are essentially generalized state machines which allow for nesting and parallel composition of states. The 'nesting' part is my favourite, since it allows one to delegate event handling logic shared by multiple states to a 'parent' state, reducing code duplication.

The great thing about this paper is that you can glean most of its key ideas by just looking at the diagrams.

Re: Designing state machines

#37

Post by nuclx »

zcdziura wrote: post #20

There was an article posted here a while back on how to implement a state machine in Rust, purely using its type system to facilitate state transitions. It's a neat concept, even if it's a bit more verbose than what you'll find in other languages. https://hoverbear.org/2016/10/12/rust-state-machine-pattern/

How does this compare to the expressivity of Idris' dependent types? I'm intrigued by the idea of type-safe state-machines as e.g. file i/o and network socket code is almost always error-prone. So it feels crucial to rely on the support of a strong type system here.

http://docs.idris-lang.org/en/latest/st/machines.html

Re: Designing state machines

#38

Post by activatedgeek »

ge96 wrote: post #32
activatedgeek wrote: post #22

While I understand the importance of state machines and its direct usage in complex Business Logic Workflows (and indirectly almost everywhere), there is nothing special in this post. It is a generic comment on State Machines. Am I missing something here?

Is "state machines" a global term that includes finite state machines like how vending machines deal with varying coin amounts that go in to trigger the price and dispense proper change? Googled looks like it is. That was a cool homework/project.

Yes, the vending machines examples is one of the most directly relatable ones. In fact, pretty much every piece of code you write has gone through a "Finite State Automata" (during the lexical phase of compilation to be exact) which in very simple terms is just a huge set of rules saying - "if this then that".

Re: Designing state machines

#39

Post by nickpsecurity »

idbfs wrote: post #36

A paper that changed my approach to designing state machines is "Statecharts: A Visual Formalism for Complex Systems" by Harel ( http://www.wisdom.weizmann.ac.il/~harel/papers/Statecharts.p... ). Statecharts (also called hierarchical state machines) are essentially generalized state machines which allow for nesting and parallel composition of states. The 'nesting' part is my favourite, since it allows one to delegate…

Exactly. UML looked like over-complicated crap when I saw it after reading on Statecharts and Yourdon. The latter are still in use on occasion in high-assurance with Tenix's Dats Diode being an example.

Re: Designing state machines

#40

Post by clairity »

bri3d wrote: post #31
activatedgeek wrote: post #22

While I understand the importance of state machines and its direct usage in complex Business Logic Workflows (and indirectly almost everywhere), there is nothing special in this post. It is a generic comment on State Machines. Am I missing something here?

A surprising number of entrepreneurs and developers are unfamiliar with the application of state machines to business logic workflows, so it's useful to see the topic presented in a succinct way. Good learning content tends to get upvoted on HN, especially when the discussion around the concept (for example, the discussion of State Machines and state machine libraries in this thread) is interesting as well.

you don't even have to be unfamiliar with state machines to have found some usefulness here. sometimes it's just nice to be reminded of a concept as you're wrangling the 20 different problems you face as an entrepreneur on any given day.

the blog post wasn't particularly deep, but the (short) discussion here was worth perusing, as i'm considering employing a state machine for my rails app and have been generally looking for options and best practices (in a mental background thread). but it looks like most (all?) existing gems assume a fixed set of states and transitions, and i was hoping to find something designed primarily for user-defined states and transitions.

Post reply on HN