Collections

< Value Generation | Up: Foundational Effects | Index | Concurrency (Familiar Patterns) >

Effectful iteration over lists. When you need to map over a collection and each iteration uses effects (reading state, writing logs, etc.), use FxList or FxFasterList.

FxList

Full-featured effectful map with support for all effects, including Yield and Suspend:

comp do
  results <- FxList.fx_map([1, 2, 3], fn item ->
    comp do
      count <- State.get()
      _ <- State.put(count + 1)
      item * 2
    end
  end)
  results
end
|> State.with_handler(0, output: fn result, state -> {result, {:final_state, state}} end)
|> Comp.run!()
#=> {[2, 4, 6], {:final_state, 3}}

Each iteration sees the state from previous iterations - the state threads through sequentially.

Note: For large iteration counts (10,000+), use Yield-based coroutines instead of FxList for better performance. See the FxList module docs for details.

FxFasterList

A ~2x faster variant using Enum.reduce_while. The API is identical:

comp do
  results <- FxFasterList.fx_map([1, 2, 3], fn item ->
    comp do
      count <- State.get()
      _ <- State.put(count + 1)
      item * 2
    end
  end)
  results
end
|> State.with_handler(0, output: fn result, state -> {result, {:final_state, state}} end)
|> Comp.run!()
#=> {[2, 4, 6], {:final_state, 3}}

When to use which

< Value Generation | Up: Foundational Effects | Index | Concurrency (Familiar Patterns) >