Sha256: 587d56f239a58db8a8b4d70257a23d954380df34d5cfdb3e5903d7e892fb457e
Contents?: true
Size: 1.52 KB
Versions: 7
Compression:
Stored size: 1.52 KB
Contents
class Trailblazer::Operation
module Policy
# Step: This generically `call`s a policy and then pushes its result to `options`.
# You can use any callable object as a policy with this step.
class Eval
def initialize(name:nil, path:nil)
@name = name
@path = path
end
# incoming low-level {Task API}.
# outgoing Task::Binary API.
def call((options, flow_options), **circuit_options)
condition = options[ @path ] # this allows dependency injection.
result = condition.( [options, flow_options], **circuit_options )
options["policy.#{@name}"] = result["policy"] # assign the policy as a skill.
options["result.policy.#{@name}"] = result
# flow control
signal = result.success? ? Trailblazer::Activity::Right : Trailblazer::Activity::Left # since we & this, it's only executed OnRight and the return boolean decides the direction, input is passed straight through.
return signal, [ options, flow_options ]
end
end
# Adds the `yield` result to the pipe and treats it like a
# policy-compatible object at runtime.
def self.step(condition, options, &block)
name = options[:name]
path = "policy.#{name}.eval"
task = Eval.new( name: name, path: path )
extension = Trailblazer::Activity::TaskWrap::Merge.new(
Trailblazer::Operation::Wrap::Inject::Defaults(
path => condition
)
)
{ task: task, id: path, extension: [extension] }
end
end
end
Version data entries
7 entries across 7 versions & 2 rubygems