A repo of one-liners and mini libraries that I've found useful
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.vscode
src
.babelrc
.eslintrc
.flowconfig
.gitignore
README.md
package-lock.json
package.json

README.md

Little Libs

First make sure you have Node.js installed (I recommend using NVM to do so). Run npm install within this directory and then npm test to see everything pass, feel free to open an issue if it doesn't.

Look at the test.js files within each section to get an idea of how to use the functions, or read on for more detail.

zip

const zip = <A,B>(xs: A[], ys: B[]): [A, B][] => (
  xs.map((x, i) => [x, ys[i]])
)

zip takes two arrays and returns them zipped up as pairs. If the second array is longer than the first it'll be truncated and if it's shorter it'll be extended with undefined values.

pluck

const pluck = (key: string, xs: Object[]): [] => (
  xs.map((x) => x[key])
)

pluck takes a key and an array of objects and returns an array of values plucked from the objects by their key.

flatten

const flatten = (xs: []): [] => (
  xs.reduce((flat, x) => (
    flat.concat(Array.isArray(x) ? flatten(x) : [x])
  ), [])
)

flatten is a recursive function that takes nested arrays of any depth and returns a flattened version.

range

const range = (length: number): number[] => [...Array(length).keys()]

range takes a length and returns an array of that size populated by values ranging from 0 to the given number (exclusive).

partition

const partition = <T>(xs: T[], fn: (x: T) => boolean): [T[], T[]] => (
  xs.reduce((p, x) => (
    fn(x) ? [p[0].concat([x]), p[1]] : [p[0], p[1].concat([x])]
  ), [[], []])
)

partition takes an array and a function to partition it by. Each value is tested by the function and if true is placed into the first partition and if false into the second.

partial

const partial = default (fn: Function, ...args: any[]): Function => (
  (...argsN: any[]) => (
    fn(...args, ...argsN)
  )
)

partial takes one function and any number of arguments and returns another function that takes any further arguments and returns the result of applying both sets of arguments to the original function. For more detail, see the Wikipedia page on partial application.

get-in

const getIn = (object: Object, path: string[], notFound: any = null) => (
  path.reduce((obj, seg) => (obj && obj[seg] || notFound), object)
)

getIn takes an object, a path (as an array of strings) to follow through the object and an optional value (defaulting to null) to be returned if the path doesn't resolve. It borrows heavily from the Clojure function of the same name.

assoc-in

const assocIn = (object: Object, [key, ...keys]: string[], value: any): Object => (
  {...object, [key]: keys.length ? assocIn(object[key], keys, value) : value}
)

assocIn takes an object, a path (as above) and a value and recursively builds up a new object that's merged with the original. It's for updating (or creating) nested values in objects and is again borrowed from a Clojure core function of the same name.

mapcat

const mapcat = <T, U>(fn: (x: T) => U[], xs: T[]): U[] => (
  [].concat(...xs.map(fn))
)

mapcat takes an array and a function that returns an array and maps this function over the given array, concatenating the results into a single array. Similarities will be found in Clojure's own mapcat.

fnull

const fnull = (fn: Function, ...args: any[]): Function => (
  (...argsN: any[]) => {
    const newArgs = args.reduce(([newArgs, [next, ...rest]], arg) => (
      [[...newArgs, arg === null ? next : arg], arg === null ? rest : [next, ...rest]]
    ), [[], argsN])[0]
    return fn(...newArgs)
  }
)

fnull takes a function and arguments to be passed to that function and returns a new function for any further arguments. If any of the first set of arguments is null they'll be replaced in-order by arguments from the second set. The idea for this function is once again borrowed from Clojure's core library, this time fnil.

chunkify

const chunkify = <T>(size: number, xs: T[]) =>
  xs.reduce(
    (xss: T[][], x: T) =>
      xss.length === 0 || xss[xss.length - 1].length === size
        ? xss.concat([[x]])
        : [
            ...xss.slice(0, xss.length - 1),
            xss[xss.length - 1].concat([x]),
            ...xss.slice(xss.length)
          ],
    []
  )

chunkify takes a number and an array and "chunks" the array into an array of arrays of that size. If the number of elements don't fit neatly into equal-sized arrays then the last array will be the size of the remaining number of elements.