Pipeable
A DSL for workflows built atop native Function Composition which leverages the Railway Pattern. This allows you to write a sequence of steps that cleanly read from top-to-bottom or left-to-right resulting in a single success or a failure. This allows you to avoid relying on exceptions for expensive control flows and/or complex conditional logic in general.
- Features
- Requirements
- Setup
- Usage
- Pipe
- Steps
- Basic
- alt
- amap
- as
- bind
- check
- fmap
- insert
- map
- merge
- tee
- to
- try
- use
- validate
- Advanced
- Procs
- Lambdas
- Methods
- Custom
- Basic
- Superpipes
- Containers
- Composition
- Development
- Architecture
- Style Guide
- Debugging
- Troubleshooting
- Type Errors
- No Method Errors
- Tests
- Benchmarks
- License
- Security
- Code of Conduct
- Contributions
- Developer Certificate of Origin
- Versions
- Community
- Credits
Features
-
Built atop native Function Composition.
-
Adheres to the Railway Pattern.
-
Provides built-in and customizable domain-specific steps.
-
Provides chainable pipes which can be used to build more complex workflows.
-
Compatible with the Containable, Infusible, and Dry Monads gems.
Requirements
-
Ruby.
-
A strong understanding of Function Composition and Method Parameters And Arguments.
Setup
To install with security, run:
# 💡 Skip this line if you already have the public certificate installed.
gem cert --add <(curl --compressed --location https://alchemists.io/gems.pem)
gem install pipeable --trust-policy HighSecurity
To install without security, run:
gem install pipeable
You can also add the gem directly to your project:
bundle add pipeable
Once the gem is installed, you only need to require it:
require "pipeable"
Usage
You can turn any object into a pipe by requiring and including this gem as follows:
require "csv"
require "pipeable"
class Demo
include Pipeable
def initialize client: CSV
@client = client
end
def call data
pipe data,
check(/Book.+Price/, :match?),
:parse,
map { |item| "#{item[:book]}: #{item[:price]}" }
end
private
attr_reader :client
def parse result
result.fmap do |data|
client.instance(data, headers: true, header_converters: proc { |key| key.downcase.to_sym })
.to_a
.map(&:to_h)
end
end
end
The above allows Demo#call
to be a sequence of steps which may pass or fail due to all steps using Dry Monads for input and output. This is the essence of the Railway Pattern.
To execute the above example, you’d only need to pass CSV content to it:
Demo.new.call <<~CSV
Book,Author,Price,At
Mystics,urGoh,10.50,2022-01-01
Skeksis,skekSil,20.75,2022-02-13
CSV
The computed result is a success with each book listing a price:
Success ["Mystics: 10.50", "Skeksis: 20.75"]
Pipe
Once you’ve included the Pipeable
module within your class, the #pipe
method is available to you and is how you build a sequence of steps for processing. The method signature is:
pipe(input, *steps)
The first argument is your input which can be a Ruby primitive or a monad. Regardless, the input will be automatically wrapped as a Success
 — but only if not a Result
to begin with — before passing to the first step. From there, all steps are required to answer a monad in order to adhere to the Railway Pattern.
Behind the scenes, the #pipe
method is syntactic sugar built atop Function Composition which means if this code were to be rewritten:
pipe csv,
check(/Book.+Price/, :match?),
:parse,
map { |item| "#{item[:book]}: #{item[:price]}" }
…​then the above would look like the following (as rewritten in native Ruby):
(
check(/Book.+Price/, :match?) >>
method(:parse) >>
map { |item| "#{item[:book]}: #{item[:price]}" }
).call Success(csv)
Visually, the pipe can be diagramed as follows:
The problem with native function composition is that it reads backwards by passing input at the end of all sequential steps. With the #pipe
method, you have the benefit of allowing your eyes to read from top to bottom while not having to type multiple forward composition operators.
Steps
There are several ways to compose steps for your pipe. As long as all steps succeed, you’ll get a successful response. Otherwise, the first step to fail will pass the failure down by skipping all subsequent steps (unless you dynamically turn the failure into a success). Each step can be initialized and called:
-
#initialize
: Arguments vary per step but can be positional, keyword, and/or block arguments. This is how you customize the behavior of each step. -
#call
: Expects a Dry MonadsResult
object as input. The output is either the same or newResult
object for consumption by the next step in the pipe. Additionally, each step will either unwrap theResult
or pass theResult
through depending on the step’s implementation (as detailed below).
Basic
The following are the basic (default) steps for building custom pipes for which you can mix and match within your own implementation.
alt
Allows you to operate on a failure and produce either a success or another failure. This is a convenience wrapper to native Dry Monads #or
functionality.
Accepts a failure while answering either a success or failure. Example:
pipe %i[a b c], alt { |object| Success object.join("-") } # Success [:a, :b, :c]
pipe Failure("Danger!"), alt { Success "Resolved" } # Success "Resolved"
pipe Failure("Danger!"), alt { |object| Failure "Big #{object}" } # Failure "Big Danger!"
amap
Allows you to unwrap a failure, make a modification, and wrap the modification as a new failure. This is a convenience wrapper to native Dry Monads #alt_map
functionality.
Accepts and answers a failure. Example:
pipe Failure("Danger"), amap { |object| "#{object}!" } # Failure "Danger!"
pipe Success("Pass"), amap { |object| "#{object}!" } # Success "Pass"
as
Allows you to message an object as a different result. The first argument is the method but additional positional and/or keyword arguments can be passed along if the method accepts them.
Accepts and answers a success. Example:
pipe :a, as(:inspect) # Success ":a"
pipe %i[a b c], as(:dig, 1) # Success :b
pipe Failure("Danger!"), as(:inspect) # Failure "Danger!"
bind
Allows you to perform operations upon success only. You are then responsible for answering a success or failure accordingly. This is a convenience wrapper to native Dry Monads #bind
functionality.
Accepts a success while answering either a success or failure. Example:
pipe %i[a b c], bind { |object| Success object.join("-") } # Success "a-b-c"
pipe %i[a b c], bind { |object| Failure object } # Failure [:a, :b, :c]
pipe Failure("Danger!"), bind { |object| Success object.join("-") } # Failure "Danger!"
check
Allows you to check if an object matches the proof (with message). The first argument is your proof while the second argument is the message to send to your proof. A check only passes if the messaged object evaluates to true
or Success
. When successful, the object is passed through as a Success
. When false, the object is passed through as a Failure
.
Accepts a success while answering a success or failure depending on whether unwrapped object checks against the proof. Example:
pipe :a, check(%i[a b], :include?) # Success :a
pipe :a, check(%i[b c], :include?) # Failure :a
pipe Failure("Danger!"), check(%i[a b], :include?) # Failure "Danger!"
fmap
Allows you to unwrap a success, make a modification, and wrap the modification as a new success. This is a convenience wrapper to native Dry Monads #fmap
functionality.
Accepts and answers a success. Example:
pipe %i[a b c], fmap { |object| object.join "-" } # Success "a-b-c"
pipe Failure("Danger!"), fmap { |object| object.join "-" } # Failure "Danger!"
insert
Allows you to insert an element after an object (default behavior) as a single array. This step wraps native Array#insert functionality. If the object is not an array, it will be cast as one. You can use the :at
key to specify where you want insertion to happen. This step is most useful when needing to assemble positional arguments for passing as an array to a subsequent step.
Accepts and answers a success. Example:
pipe :a, insert(:b) # Success [:a, :b]
pipe :a, insert(:b, at: 0) # Success [:b, :a]
pipe %i[a c], insert(:b, at: 1) # Success [:a, :b, :c]
pipe Failure("Danger!"), insert(:b) # Failure "Danger!"
map
Allows you to map over an object (enumerable) by wrapping native Enumerable#map functionality.
Accepts and answers a success. Example:
pipe %i[a b c], map(&:inspect) # Success [":a", ":b", ":c"]
pipe Failure("Danger!"), map(&:inspect) # Failure "Danger!"
merge
Allows you to merge an object with additional attributes as a single hash. This step wraps native Hash#merge functionality. If the input is not a hash, then the object will be merged with step
as the key. The default step
key can be renamed to a different key by using the :as
key. Like the insert step, this step is most useful when assembling keyword arguments and/or a hash for a subsequent steps.
Accepts and answers a success. Example:
pipe({a: 1}, merge(b: 2)) # Success {a: 1, b: 2}
pipe "test", merge(b: 2) # Success {step: "test", b: 2}
pipe "test", merge(as: :a, b: 2) # Success {a: "test", b: 2}
pipe Failure("Danger!"), merge(b: 2) # Failure "Danger!"
tee
Allows you to run an operation and ignore the response while input is passed through as output. This behavior is similar in nature to the tee program in Bash.
Accepts either a success or failure and passes the result through while allowing you to execute arbitrary behavior. Example:
pipe "test", tee(Kernel, :puts, "Example.")
# Example.
# Success "test"
pipe Failure("Danger!"), tee(Kernel, :puts, "Example.")
# Example.
# Failure "Danger!"
to
Allows you to delegate to an object which doesn’t have a callable interface and may or may not answer a result. If the response is not a monad, it’ll be automatically wrapped as a Success
.
Accepts a success while sending the unwrapped object to the given object’s corresponding method. The object is expected to answer either a plain Ruby object which will be automatically wrapped as a success or a Dry Monads Result
. Example:
Model = Struct.new :label do
include Dry::Monads[:result]
def self.for(**) = Success new(**)
end
pipe({label: "Test"}, to(Model, :for)) # Success #<struct Model label="Test">
pipe Failure("Danger!"), to(Model, :for) # Failure "Danger!"
try
Allows you to try an operation which may fail while catching any exceptions as a failure for further processing. You can catch a single exception by providing the exception as a single value or multiple exceptions as an array of values.
Accepts and answers a success if there are no exceptions. Otherwise, captures any error as a failure. Example:
pipe "test", try(:to_json, catch: JSON::ParserError)
# Success "\"test\""
pipe "test", try(:to_json, catch: [JSON::ParserError, StandardError])
# Success "\"test\""
pipe "test", try(:invalid, catch: NoMethodError)
# Failure(#<NoMethodError: undefined method `invalid' for an instance of String>)
pipe Failure("Danger!"), try(:to_json, catch: JSON::ParserError)
# Failure "Danger!"
use
Allows you to use another pipe to build a superpipe, use an object that adheres to the Command Pattern, or any function which answers a Dry Monads Result
object. In other words, you can use use any object which responds to #call
that answers a Dry Monads Result
object. This is great for chaining multiple pipes together (i.e. superpipes).
Accepts a success while sending the unwrapped object to the command (or pipe) for further processing. A Dry Monads Result
is expected to be answered by the command. Example:
function = -> number { Success number * 3 }
pipe 3, use(function) # Success 9
pipe Failure("Danger!"), use(function) # Failure "Danger!"
validate
Allows you to use an contract for validating an object. This is especially useful when using Dry Schema, Dry Validation, or any contract that responds to #call
and answers a Result
.
By default, the :as
key’s value is nil
. Use :to_h
, for example, as the value for automatic casting to a Hash
. You can also pass in any value to the :as
key which is a valid method that the contract’s result will respond to.
Accepts a success and rewraps as a success if the :as
keyword is supplied. Otherwise, any failure is immediately passed through. Example:
schema = Dry::Schema.Params { required(:label).filled :string }
pipe({label: "Test"}, validate(schema))
# Success label: "Test"
pipe({label: "Test"}, validate(schema, as: nil))
# Success #<Dry::Schema::Result{:label=>"Test"} errors={} path=[]>
pipe Failure("Danger!"), validate(schema)
# Failure "Danger!"
💡 Ensure you enable the Dry Monads extension for Dry Schema and/or Dry Validation when using this step since this step expects the contract to respond to the #to_monad
message.
Advanced
Several options are available should you need to advance beyond the basic steps. Each is described in detail below.
Procs
You can always use a Proc
as a custom step. Example:
include Dry::Monads[:result]
include Pipeable
pipe :a,
insert(:b),
proc { Success "input_ignored" },
as(:to_sym)
# Yields: Success :input_ignored
Lambdas
In addition to procs, lambdas can be used too. Example:
include Pipeable
pipe :a,
insert(:b),
-> result { result.fmap { |items| items.join "_" } },
as(:to_sym)
# Yields: Success :a_b
Methods
Methods, in addition to procs and lambdas, are the preferred way to add custom steps due to the concise syntax. Example:
class Demo
include Pipeable
def call(input) = pipe input, insert(:b), :join, as(:to_sym)
private
def join(result) = result.fmap { |items| items.join "_" }
end
Demo.new.call :a # Success :a_b
All methods can be referenced by symbol as shown via :join
above. Using a symbol is syntactic sugar for Object#method so :join
(symbol) is the same as using method(:join)
. Both work but the former requires less typing.
Custom
If you’d like to define permanent and reusable steps, you can register a custom step which requires you to:
-
Define a custom step as a class, lambda, or proc.
-
Register your custom step along side the existing default steps.
Here’s what this would look like:
module CustomSteps
class Join < Pipeable::Steps::Abstract
def initialize(delimiter = "_", **)
super(**)
@delimiter = delimiter
end
def call(result) = result.fmap { |items| items.join delimiter }
private
attr_reader :delimiter
end
end
Pipeable::Steps::Container.register :join, CustomSteps::Join
include Pipeable
pipe :a, insert(:b), join, as(:to_sym)
# Success :a_b
pipe :a, insert(:b), join(""), as(:to_sym)
# Success :ab
A lambda or proc can be used too (albeit in limited capacity). Here’s a version of the above using a lambda:
module CustomSteps
Join = -> result { result.fmap { |items| items.join "_" } }
end
Pipeable::Steps::Container.register :join, CustomSteps::Join
include Pipeable
puts pipe(:a, insert(:b), join, as(:to_sym))
# Success :a_b
Superpipes
Superpipes, as first hinted at in the use
step above, are a combination of pipeable objects chained together as individual steps. This allows you to reuse existing pipeable objects in new and interesting ways. Here’s an contrived, but simple, example of what a superpipe looks like when built from pipeable objects:
class One
include Pipeable
def initialize label = "one"
@label = label
end
def call(item) = pipe item, insert(label, at: 0)
private
attr_reader :label
end
class Two
include Pipeable
def initialize label = "two"
@label = label
end
def call(item) = pipe item, insert(label)
private
attr_reader :label
end
class Three
include Pipeable
def initialize one: One.new, two: Two.new
@one = one
@two = two
end
def call(item) = pipe item, use(one), use(two)
private
attr_reader :one, :two
end
Notice, One
and Two
are normal pipeable objects with individual steps while Three
injects both One
and Two
as dependencies and then subsequently pipes them together in the #call
method via the use
step. This is the power of a superpipe. …​and, yes, a superpipe can be an individual step in some other object. Turtles all the way down (or up). 😉
Again, the above is contrived but hopefully illustrates how you can build more complex architectures from smaller pipes.
Containers
Should you not want the basic steps, need custom steps, or a hybrid of default and custom steps, you can define your own container — using the Containable gem — and provide the container as an argument to .[]
when including pipeable behavior. Example:
require "containable"
module CustomContainer
extend Containable
register :echo, -> result { result }
register :insert, Pipeable::Steps::Insert
end
include Pipeable[CustomContainer]
pipe :a, echo, insert(:b)
# Yields: Success [:a, :b]
The above is a hybrid example where the CustomContainer
registers a custom echo
step along with the default insert
step to make a new container. This is included when passed in as an argument via .[]
(i.e. include Pipeable[CustomContainer]
).
Whether you use default, custom, or hybrid steps, you have maximum flexibility when using containers.
Composition
Should you ever need to make a plain old Ruby object functionally composable, then you can include the Pipeable::Composable
module which will give you the necessary #>>
, #<<
, and #call
methods where you only need to implement the #call
method.
Development
To contribute, run:
git clone https://github.com/bkuhlmann/pipeable
cd pipeable
bin/setup
You can also use the IRB console for direct access to all objects:
bin/console
Architecture
The architecture of this gem is built on top of the following concepts and gems:
-
Function Composition: Made possible through the use of the
#>>
and#<<
methods on the Method and Proc objects. -
Containable: Allows related dependencies to be grouped together for injection as desired.
-
Dry Monads: Critical to ensuring the entire pipeline of steps adhere to the Railway Pattern and leans heavily on the
Result
object. -
Marameters: Through the use of the
.categorize
method, dynamic message passing is possible by inspecting the object’s method parameters.
Style Guide
-
Pipes
-
Use a single method (i.e.
#call
) which is public and adheres to the Command Pattern so multiple pipes can be piped together (i.e. superpipes) if desired.
-
-
Steps
-
Inherit from the
Abstract
class to gain monad, composition, and dependency behavior. This allows subclasses to have direct access to the base positional, keyword, and block arguments. These variables are prefixed withbase_*
in order to not conflict with subclasses which might only want to use non-prefixed variables for convenience. -
All filtered arguments — in other words, unused arguments — need to be passed up to the superclass from the subclass (i.e.
super(*positionals, **keywords, &block)
). Doing so allows the superclass (i.e.Abstract
) to provide access tobase_positionals
,base_keywords
, andbase_block
for use if desired by the subclass. -
The
#call
method must define a single positionalresult
parameter since a monad will be passed as an argument. Example:def call(result) = # Implementation
. -
Each block within the
#call
method should use theobject
parameter to be consistent. More specific parameters likeoperation
orcontract
should be used to improve readability when context allows. Example:def call(result) = result.bind { |object| # Implementation }
. -
Use implicit blocks sparingly. Most of the default steps shy away from using blocks because the code becomes more complex. Use private methods, custom steps, and/or separate pipes if the code becomes too complex because you might have a smaller object which needs extraction.
-
Debugging
If you need to debug (i.e. Debug) your pipe, use a lambda. Example:
pipe data,
check(/Book.+Price/, :match?),
-> result { binding.break; result }, # Breakpoint
:parse
The above breakpoint will allow you inspect the result of the #check
step and/or build a modified result for passing to the subsequent :parse
method step.
Troubleshooting
The following might be of aid to as you implement your own pipes.
Type Errors
If you get a TypeError: Step must be functionally composable and answer a monad
, it means:
-
The step must be a
Proc
,Method
, or any object which responds to#>>
,#<<
, and#call
. -
The step doesn’t answer a result monad (i.e.
Success object
orFailure object
).
No Method Errors
If you get a NoMethodError: undefined method success?
exception, this might mean that you forgot to add a comma after one of your steps. Example:
# Valid
pipe "https://www.wikipedia.org",
to(client, :get),
try(:parse, catch: HTTP::Error)
# Invalid
pipe "https://www.wikipedia.org",
to(client, :get) # Missing comma.
try(:parse, catch: HTTP::Error)
Tests
To test, run:
bin/rake
Benchmarks
To view/compare performance, run:
bin/benchmark
💡 You can view current benchmarks at the end of the above file if you don’t want to manually run them.
Credits
-
Built with Gemsmith.
-
Engineered by Brooke Kuhlmann.