reactor/lib/reactor.ex

220 lines
6.1 KiB
Elixir

defmodule Reactor do
alias Reactor.{Dsl, Error.Validation.StateError, Executor, Step}
@moduledoc """
Reactor is a dynamic, concurrent, dependency resolving saga orchestrator.
## Usage
You can construct a reactor using the `Reactor` Spark DSL:
```elixir
defmodule HelloWorldReactor do
@moduledoc false
use Reactor
input :whom
step :greet, Greeter do
argument :whom, input(:whom)
end
return :greet
end
```
iex> Reactor.run(HelloWorldReactor, %{whom: "Dear Reader"})
{:ok, "Hello, Dear Reader!"}
or you can build it programmatically:
iex> reactor = Builder.new()
...> {:ok, reactor} = Builder.add_input(reactor, :whom)
...> {:ok, reactor} = Builder.add_step(reactor, :greet, Greeter, whom: {:input, :whom})
...> {:ok, reactor} = Builder.return(reactor, :greet)
...> Reactor.run(reactor, %{whom: nil})
{:ok, "Hello, World!"}
"""
defstruct context: %{},
id: nil,
inputs: [],
intermediate_results: %{},
middleware: [],
plan: nil,
return: nil,
state: :pending,
steps: [],
undo: []
use Spark.Dsl, default_extensions: [extensions: [Dsl]]
@type context :: %{optional(atom) => any}
@type context_arg :: Enumerable.t({atom, any})
@typedoc """
Specify the maximum number of asynchronous steps which can be run in parallel.
Defaults to the result of `System.schedulers_online/0`. Only used if
`async?` is set to `true`.
"""
@type max_concurrency_option :: {:max_concurrency, pos_integer()}
@typedoc """
Specify the amount of execution time after which to halt processing.
Note that this is not a hard limit. The Reactor will stop when the first step
completes _after_ the timeout has expired.
Defaults to `:infinity`.
"""
@type timeout_option :: {:timeout, pos_integer() | :infinity}
@typedoc """
The maximum number of iterations which after which the Reactor will halt.
Defaults to `:infinity`.
"""
@type max_iterations_option :: {:max_iterations, pos_integer() | :infinity}
@typedoc """
How long to wait for asynchronous steps to complete when halting.
Defaults to 5000ms.
"""
@type halt_timeout_option :: {:halt_timeout, pos_integer() | :infinity}
@typedoc """
When set to `false` forces the Reactor to run every step synchronously,
regardless of the step configuration.
Defaults to `true`.
"""
@type async_option :: {:async?, boolean}
@typedoc """
Use a `Reactor.Executor.ConcurrencyTracker.pool_key` to allow this Reactor to
share it's concurrency pool with other Reactor instances.
If you do not specify one then the Reactor will initialise a new pool and
place it in it's context for any child Reactors to re-use.
Only used if `async?` is set to `true`.
"""
@type concurrency_key_option :: {:concurrency_key, reference()}
@type options ::
Enumerable.t(
max_concurrency_option
| timeout_option
| max_iterations_option
| halt_timeout_option
| async_option
| concurrency_key_option
)
@type state :: :pending | :executing | :halted | :failed | :successful
@type inputs :: %{optional(atom) => any}
@type t :: %Reactor{
context: context,
id: any,
inputs: [atom],
intermediate_results: %{any => any},
middleware: [Reactor.Middleware.t()],
plan: nil | Graph.t(),
undo: [{Step.t(), any}],
return: any,
state: state,
steps: [Step.t()]
}
@doc "A guard which returns true if the value is a Reactor struct"
@spec is_reactor(any) :: Macro.t()
defguard is_reactor(reactor) when is_struct(reactor, __MODULE__)
@option_schema [
max_concurrency: [
type: :pos_integer,
required: false,
doc: "The maximum number of processes to use to run the Reactor"
],
timeout: [
type: {:or, [:pos_integer, {:literal, :infinity}]},
required: false,
default: :infinity,
doc: "How long to allow the Reactor to run for"
],
max_iterations: [
type: {:or, [:pos_integer, {:literal, :infinity}]},
required: false,
default: :infinity,
doc: "The maximum number of times to allow the Reactor to loop"
],
async_option: [
type: :boolean,
required: false,
default: true,
doc: "Whether to allow the Reactor to start processes"
],
concurrency_key_option: [
type: :reference,
required: false,
hide: true
]
]
@doc """
Attempt to run a Reactor.
## Arguments
* `reactor` - The Reactor to run, either a Reactor DSL module, or a Reactor
struct.
* `inputs` - A map of values passed in to satisfy the Reactor's expected
inputs.
* `context` - An arbitrary map that will be merged into the Reactor context
and passed into each step.
## Options
#{Spark.Options.docs(@option_schema)}
"""
@doc spark_opts: [{4, @option_schema}]
@spec run(t | module, inputs, context_arg, options) :: {:ok, any} | {:error, any} | {:halted, t}
def run(reactor, inputs \\ %{}, context \\ %{}, options \\ [])
def run(reactor, inputs, context, options) when is_atom(reactor) do
if Spark.Dsl.is?(reactor, Reactor) do
run(reactor.reactor(), inputs, context, options)
else
{:error,
ArgumentError.exception(message: "Module `#{inspect(reactor)}` is not a Reactor module")}
end
end
def run(reactor, inputs, context, options)
when is_reactor(reactor) and reactor.state in ~w[pending halted]a do
Executor.run(reactor, inputs, context, options)
end
def run(reactor, _inputs, _context, _options) do
{:error,
StateError.exception(
reactor: reactor,
state: reactor.state,
expected: ~w[pending halted]a
)}
end
@doc "Raising version of `run/4`."
@spec run!(t | module, inputs, context_arg, options) :: any | no_return
def run!(reactor, inputs \\ %{}, context \\ %{}, options \\ [])
def run!(reactor, inputs, context, options) do
case run(reactor, inputs, context, options) do
{:ok, value} -> value
{:error, reason} -> raise reason
end
end
end