Stack-based composition combinators in JavaScript

Usage no npm install needed!

<script type="module">
  import dashkiteKatana from 'https://cdn.skypack.dev/@dashkite/katana';



Daisho (stack- and context-based) composition combinators in JavaScript.

import {pipe} from "@dashkite/joy/function"
import {push, mpush, pop, stack} from "@dashkite/katana/sync"

pipe [
  push -> 3
  push -> 4
  mpush add
  pop (sum) -> assert.equal sum, 7
  (stack) -> assert.equal stack.length, 0

Table Of Contents


npm i @dashkite/katana

Browser-compatible. Use with your favorite bundler or import directly.


Function composition is a powerful tool in theory, but in practice, it's often difficult for non-trivial scenarios because the arguments and return values of a given set of functions may not be amenable to simple composition. Stack-based composition provides generic context—the stack—and a set of combinators for adapting ordinary functions for use with it. This simplifies composition, even across libraries that were not designed to be used together. This is a key advantage of composition over chaining, which requires that each function be expressly added to an object as a method.

Daisho Data Structure

However, stack-based composition can lead to code that is difficult to reason about. Combining a stack with a context object makes possible variety of compositional scenarios. Context-based composition is the basis for method-chaining, or fluent, programming, popularized by jQuery, where the target object serves as the context. We call this hybrid stack/object data structure a daisho, because of its dual nature. We can use the stack for simple composition and the context for complex composition. In combination, we may use the stack to compute results we place into the context for later use.


Stack operations always apply a function, using the arity of the function to determine how many elements from the stack to pass into the function and possibly to remove from the stack. Applying a unary function will result in passing the top of the stack into the function. Applying a binary function will result in passing the first two elements from the stack into the function, and so on. Context operations do not apply a function, but simply move data to and from the context.

There are synchronous and asynchrouns variants for operations that apply a function. By default, when importing Katana, you get the asynchronous versions. These are bit slower since they yield control of the event loop after each operation that applies a function (since the function may return a promise). You may load the synchronous versions using a subpath:

import {push, pop} from "@dashkite/katana/sync"

You can load both variants using the wildcard import:

import * as ks from "@dashkite/katana/sync"
import * as ka from "@dashkite/katana/async"

Keep in mind that the async variants that apply a function will return a promise.


Operations that mutate the given daisho operate on and return a clone. However, keep in mind that the values within it are not cloned (that is, it is not a deep clone).

Creating A Daisho

Daisho.create object → daisho

Daisho.create iterable → daisho

Daisho.create iterable, object → daisho

Daisho.create object, iterable → daisho

You may create a Daisho using an iterable, object, or both, in any order. The stack will be constructed from an iterable using Array.from. If you pass in an array, however, it will be used directly.

Stack Operations

Functions prefixed with an m will alter the stack based on the arity of the given function, ex: mpop, will not only pass arity elements to the given function, but will subsequently remove those elements from the stack (instead of just the first element).


push f, daisho → daisho

Calls f with arity f elements from the top of the stack. The result is added to the top of the stack.

pop | mpop

pop f, daisho → daisho

Calls f with arity f elements from the top of the stack. The top of the stack is removed.


peek f, daisho → daisho

Calls f with arity f elements from the top of the stack. The stack is unchanged.

poke | mpoke

poke f, daisho → daisho

Calls f with arity f elements from the top of the stack. The top of the stack is replaced with the result.


pushn array<function>, daisho → daisho

Like push, but for an array of functions, pushing the result of each onto the stack.


discard daisho → daisho

Discard the element at the top of the stack. Equvalent to pop -> but faster and there’s no need for a synchronous variant.

Context Operations

The read and write functions operate on the context.


read name, daisho → daisho

Reads a property from the context and pushes it.


write name, daisho → daisho

Writes the element at the top of the stack to the context.


assign f, daisho → daisho

Applies the function, which should take and return a daisho as an argument and sets the context based on the returned context. Useful for performing a computation and writing the result to the context while discarding any changes to the stack.


test predicate, action

Calls predicate with arity predicate elements from the top of the stack and calls action with the stack if the result is true.

branch conditions

Given an array of conditions whose elements are predicate, action pairs, calls predicatewitharity predicateelements from the top of the stack and callsaction` with the stack if the result is true, for each pair, until one is true.

The array may optionally take a last element as a default action (whose predicate is implicitly true).



daisho → stack

Pushes the entire stack onto the top of the stack.


daisho → context

Pushes the context onto the top of the stack.


daisho → value

Returns the top of the stack.