@aeinbu/groupby
    TypeScript icon, indicating that this package has built-in type declarations

    1.1.0 • Public • Published

    @aeinbu/groupby

    Lightweight and non-intrusive functions for doing "group by" and "distinct" (aka. "unique") transformations on collections using reducers

    • The functions are lightweight, as they are small and don't make you bring along lots of functions you don't need.
    • The functions are flexible since you can choose how to best represent your grouped data
      • An array of key-value pairs that are easy to further process using .map(...), .filter(...) and other built-in array functions
      • A dictionary in the shape of a standard javascript object or an ES2015 built-in Map
    • These functions run just as well in a browser as they do in nodejs.
    • The package is non-intrusive, as it doesn't extend arrays adding new and non-standard methods to them.
      • This is done by having functions that you use with .reduce(...) on arrays.
    • @types are included for better discoverability during development when you use editors with javascript/typescript intellisence support

    How to install npm (scoped)

    With npm

    npm install @aeinbu/groupby

    ...or with yarn

    yarn add @aeinbu/groupby

    How to use

    import { groupBy, toDictionary } from "@aeinbu/groupby"
    
    const people = [
        {name: "Tony", residence: "Rome"},
        {name: "Mary", residence: "Rome"},
        {name: "Peter", residence: "London"},
        {name: "Peter", residence: "Rome"},
        {name: "Elisabeth", residence: "London"},
        {name: "Francois", residence: "Paris"}
    ]
    
    const resultsAsArray = people.reduce(
        groupBy(
            x => x.residence,  // First lambda is used to extraxt the key values, so this will group by the residence
            x => x.name        // Second lambda to determine object values to put in the groups, so in this example the group will contain all names for people in a residence city
            // This paramter is optional, and if it is skipped, the whole item is selected as the value
        ),
        []
    )
    // Result (Note how the output is a new array):
    // [
    //     {"key": "Rome", values: ["Tony", "Mary", "Peter"]},
    //     {"key: "London", values: ["Peter", "Elisabeth"]},
    //     {"key: "Paris", values: ["Francois"]}
    // ]
    
    const resultsAsDictionary = resultsAsArray.reduce(
        toDictionary(
            x => x.name,   // First lambda to determine the property name
            x => x.values  // Second lambda to determine where to find the value to set that property to
            // This paramter is optional, an if omitted it will default to `x => x.values´ which would match the default output of `groupBy` above
        ),
        {}
    )
    // Result (See how the array is transformed to an object with properties for each key):
    // {
    //     "Rome": ["Tony", "Mary", "Peter"],
    //     "London": ["Peter", "Elisabeth"],
    //     "Paris": ["Francois"]
    // }

    Since the reduction from groupBy is an array, the above two transforms can be chained and shortened (using a default parameter in toDictionary), like this:

    import { groupBy, toDictionary } from "@aeinbu/groupby"
    
    //...
    
    const chainedResults = people
        .reduce(groupBy(x => x.name, x => x.residence), [])
        .reduce(toDictionary(x => x.name), {})
    // With the same results as above:
    // {
    //     "Rome": ["Tony", "Mary", "Peter"],
    //     "London": ["Peter", "Elisabeth"],
    //     "Paris": ["Francois"]
    // }

    There is also a toMap reducer, so you can create a ES2015 Map instead of an ordinary object:

    import { groupBy, toMap } from "@aeinbu/groupby"
    
    //...
    
    const chainedResults = people
        .reduce(groupBy(x => x.name, x => x.residence), [])
        .reduce(toMap(x => x.name), new Map())
    // With the same results as above:
    // {
    //     "Rome": ["Tony", "Mary", "Peter"],
    //     "London": ["Peter", "Elisabeth"],
    //     "Paris": ["Francois"]
    // }

    Also, look in the tests directory for more examples. The tests demonstrate at least another dosen different ways to use this library

    Semantic versioning

    This package follows semantic versioning (See semver.org for more info)

    License

    This package is published under the MIT License. (See LICENSE file for more info)

    Install

    npm i @aeinbu/groupby

    DownloadsWeekly Downloads

    10

    Version

    1.1.0

    License

    MIT

    Unpacked Size

    16.1 kB

    Total Files

    15

    Last publish

    Collaborators

    • aeinbu