demo phx 1.7 live view + finitomata + ecto?

[niccolox]

hello

may I be so bold to ask if there is an example of phx 1.7 live view + finitomata + ecto?

if not, why? or why not?

if I made a demo, any pointers to using the example ecto_integration and the generator-driven scaffolds in phx 1.7?

is finitomata designed to be used only on custom phoenix apps or can it work with phx 1.7 mainline apps?

I see https://github.com/am-kantox/eblox but its a static site generator and was hoping for a more basic pattern I could draw from

I would like some guidance on the persistence of finitomata fsm used with say the latest live view generated code

its all a bit new, and there are zero examples afaik

apologies if this is the incorrect forum

thank you

[am-kantox]

Hi @niccolox,

the library itself is not tied to any phx/ecto version. It might be used as a business process handler in contexts, but it also might be used in any other way. It does not depend on phx/ecto, neither phx/ecto is connected to it in any way.

I am lazily working on the more concise example, involving ecto, but in general, I want it to stay as disconnected from web/persistence as possible.

I will leave this issue open to link it when more examples are to come.

[niccolox]

fair enough @am-kantox

I saw the defimpl Finitomata.Persistency.Persistable, for: EctoIntegration.Data.Post do and was wondering if there was an integration pattern I was missing

finitomata/examples/ecto_integration/lib/ecto_integration/data/post/persistable.ex

Line 1 in f9284a6

defimpl Finitomata.Persistency.Persistable, for: EctoIntegration.Data.Post do

I can do some discovery around finitomata +phx 1.7 live view and ecto

thank you for interesting work

[am-kantox]

Well, technically one might use Finitomata.Persistency.Persistable implementation in combination with Finitomata.Persistency to simplify the loading/storing FSM with a state into/from any storage, including, but not limited to DB.

This is not a well-thought concept yet (although it should work,) as in general, I am not concerned about how distributed FSM should have been handled. There would be many pitfalls and the architecture is still under heavy rethinking.