What cannot be runtime config?

The first thing I can think off are modules names that macros will be used from, or will be modules generated by macros.

Basically, anything that changes the compilation result.

How to know what modules have macros?

Well if we are talking about your application then you already know it. And if we are talking about a library then it is either documented in the configuration section of the documentation, if it exists, or you have to dive in the code:

Module that define macros call defmacro. Module that use macros call use SomeModule or require SomeModule.

I am not sure if import SomeModule also “requires” the module to be able to call its macros.

Unfortunately this only gets you so far…

That is why, when they say " if it can be set at runtime, it should be set at runtime", then I would say “yes, if it is environment dependent (a hostname, a port, a database connection string, anything that you would put in your .env files)”. Otherwise, if it is something that doesn’t change, the compile-time config is fine.

I am wondering how to move most configs into runtime.exs. This works simple for local running or on production server when real values are required. runtime.exs looks something like

import Config

config :myapp,
  x: System.get_env("X"),
  y: System.get_env("Y"),
  z: System.get_env("Z"),
  # ... etc...

But how to override this for testing?

You can use Config.config_env/0 to differentiate between envs

import Config
... 
if config_env() == :test do
  config :myapp, foo: "bar"
end

You can use Application.put_env in your tests directly.

Forgive me for my struggle to understand a good solution. I think my options are to do maybe these:

Option 1: fill runtime.exs:

if config_env() == :test do
  config :myapp, 
      x: "test-x",
      y: "test-y",
      z: System.get_env("TEST_Z", "test-z"),
     # ... etc...
else
  config :myapp, 
      x: System.get_env("X"),
      y: System.get_env("Y"),
      z: System.get_env("Z"),
     # ... etc...
end

Pro: declarative, clear
Con: double long runtime.exs. Also if-statements in configs are always for me a smell.

Option 2: use test_helper.exs

Use simple runtime.exs and add test values to test_helper.exs like maybe this:

  config :myapp, 
      x: System.get_env("X"),
      y: System.get_env("Y"),
      z: System.get_env("Z"),
     # ... etc...

and then in test_helper.exs I can done Application.put_env like

Application.put_env(:myapp, :x, "test-x")
Application.put_env(:myapp, :y, "test-y")
Application.put_env(:myapp, :z, "test-z")

Pro: simple runtime.exs (no double, no if-statement)
Con: cannot enforce System ENV vars using System.fetch_env! – only get_env would work so that execution does not stop before test_helper.exs runs and a value is supplied.

Option 3: Vapor

I looked also at Vapor. I like this: it could replace runtime.exs 100% I think. But it is still a problem when you must manually take a value and use Application.put_env to put it into the correct place so your app and deps can find it. For example, imagine a dependency with config like this:

config :some_dep, 
   deep: %{
      thing1: "a",
      thing2: System.get_env("THING2")
   }

With vapor, I can get THING2 ENV var from .env but I must be careful with any deep config like this example dependancy. Where to put it once I have it? I think I must merge map and then put inside carefully into Application process dictionary.

Pro: support for .env and .env.{ENV} files for each override. Errors when config is not correct.
Con: difficult for complex config that rely on Application.get_env.

I think there must be other ways. Sorry for my confusion. I admit I thought this was a easy topic but I am understanding more that config has subtle!

What about having in config/test.exs

  config :myapp, 
      x: "test-x",
      y: "test-y",
      z: System.get_env("TEST_Z", "test-z"),

and in config/runtime.exs

if config_env() == :prod do
  config :myapp, 
      x: System.get_env("X"),
      y: System.get_env("Y"),
      z: System.get_env("Z"),
     # ... etc...
end

[/quote]

if-statements in configs are always for me a smell.

For me if-statement is smell almost everywhere but configs

Take a look at the example how configs are organized in hexpm/config at main · hexpm/hexpm · GitHub

This suffers from being long and containing ifs, but I’m posting it anyway for reference. I went 100% on runtime.exs in my project and only have a couple of things that need to be set at compile time in the config.exs.

• Runtime config
• Compile time config – for compile time settings or things I don’t ever expect to change
• Configuration helpers – utility functions for casting environment variables to the correct data type
• DotenvParser – tiny package I wrote so I can read the configurations from .env when developing locally

This has worked well for me. Even though the runtime config file is big, it’s not really that big once you get to know it, and I can have unified config for both development and release time (just using .env in development for convenience).

This is a beautiful little package! Thank you for sharing your approach – it works well!