Straffern
Code reuse with Nerves and gen_statem/gen_state_machine
I am currently tinkering with a poc, which is a rewrite of an existing product originally implemented in Python.
I recently saw a presentation on :gen_statem, and wanted to play around with it. I quickly realised that it might fit nicely my case.
The problem/case:
The existing product functions both as a SaaS product and a “firmware” for an IoT product. Given that I hope to utilise Nerves, for the IoT product, I suddenly see a case for adding additional states, that only is a concern for the Nerves variant of the product.
What I’m asking
Is it possible to develop a standalone software, that can run in a container, but also be imported and extended in the Nerves software?
Such that I can extend the “Sensor” gen_state_machine, adding additional states, and override a selection of callbacks already implemented in “Sensor” module?
Do I need a macro for this?
Is this bad practice?
Thanks! 
Trending in Questions
Other Trending Topics
Latest on Elixir Forum
Sponsor Spotlight
Build Elixir applications with speed and confidence.
Categories:
Sub Categories:
Forums
Popular Tags
- #ecto
- #liveview
- #troubleshooting
- #learning-elixir
- #deployment
- #library
- #erlang
- #testing
- #genserver
- #mix
- #absinthe
- #remote-other
- #otp
- #plug
- #how-to-question
- #macros
- #postgres
- #channels
- #elixirconf
- #exunit
- #discussion
- #code-sync
- #javascript
- #podcasts
- #onsite
- #dialyzer
- #docker
- #authentication
- #umbrella
- #full-time-contract
- #podcasts-by-brainlid
- #ecto-query
- #elixir-ls
- #phoenix_html
- #iex
- #blog-post
- #graphql
- #genstage
- #ai
- #elixirconf-us
- #websockets
- #supervisor
- #advent-of-code
- #distillery
- #processes
- #api
- #forms
- #metaprogramming
- #performance
- #security
Our Sponsors
Build Elixir applications with speed and confidence.
Supporting innovation across the BEAM ecosystem.
We build reliable cloud platforms for business-critical systems.
Catch errors, track performance, monitor hosts and more.
Error tracking, logs, uptime, and Just Enough APM. Built for developers.
Practical resources that improve the lives of professional developers.
Develop your skills with books, videos, and courses.
Enabling companies to succeed by building software people love.
The team behind Membrane, Popcorn, LiveDebugger. Available for hire.
Courses that'll move you from confusion to "Aha, now I get it!"
First 5 of 5 Posts!
LostKobrakai
You don’t. If you think about it you’re doing the same with gen_statem already. It provides the underlying logic on how to run a state machine, while the module you implement provides the logic of which states act how.
You could provide your own behaviour and extension points within to further allow another behaviour implementing module to extend what you built on top of gen_statem.
Straffern
Its possible that I’ve misunderstood behaviours.
So I basically just add a bunch of typespecs for callbacks & types, similar to the implementation of gen_state_machine, and then make some state_function overridable?
LostKobrakai
That would be one way to do that, yes.
You can see similar “stacked implementations” with e.g. LiveView.
use Phoenix.LiveViewwith all it’s callbacks: Phoenix.LiveView — Phoenix LiveView v0.19.5Phoenix.LiveView.Channel, which is aGenServer: phoenix_live_view/lib/phoenix_live_view/channel.ex at v0.19.5 · phoenixframework/phoenix_live_view · GitHubGenServer.Similarly for plain channels
use Phoenix.Channelmodules are also started asGenServerprocesses.Straffern
I don’t think I understand.
Looking at the second link with Phoenix.Liveview.Channel, I see no use of Phoenix.Liveview
LostKobrakai
It’s the other way round. The channel isn’t a LV. But the channel is what “runs” a LV.
E.g. here you can see it forwarding non-internally-handled
handle_infomessages to the liveview:https://github.com/phoenixframework/phoenix_live_view/blob/v0.19.5/lib/phoenix_live_view/channel.ex#L421