Skip to content

pbeshai/tidy

Repository files navigation

tidy.js

CircleCI npm

Tidy up your data with JavaScript! Inspired by dplyr and the tidyverse, tidy.js attempts to bring the ergonomics of data manipulation from R to javascript (and typescript). The primary goals of the project are:

  • Readable code. Tidy.js prioritizes making your data transformations readable, so future you and your teammates can get up and running quickly.

  • Standard transformation verbs. Tidy.js is built using battle-tested verbs from the R community that can handle any data wrangling need.

  • Work with plain JS objects. No wrapper classes needed — all tidy.js needs is an array of plain old-fashioned JS objects to get started. Simple in, simple out.

Secondarily, this project aims to provide acceptable types for the functions provided.

Quick Links

Related work

Be sure to check out a very similar project, Arquero, from UW Data.

Getting started

To start using tidy, your best bet is to install from npm:

npm install @tidyjs/tidy
# or
yarn add @tidyjs/tidy

Then import the functions you need:

import { tidy, mutate, arrange, desc } from '@tidyjs/tidy'

Note if you're just trying tidy in a browser, you can use the UMD version hosted on jsdelivr (codesandbox example):

<script src="https://d3js.org/d3-array.v2.min.js"></script>
<script src="https://cdn.jsdelivr.net/npm/@tidyjs/tidy/dist/umd/tidy.min.js"></script>
<script>
  const { tidy, mutate, arrange, desc } = Tidy;
  // ...
</script>  

And use them on an array of objects:

const data = [
  { a: 1, b: 10 }, 
  { a: 3, b: 12 }, 
  { a: 2, b: 10 }
]

const results = tidy(
  data, 
  mutate({ ab: d => d.a * d.b }),
  arrange(desc('ab'))
)

The output is:

[
  { a: 3, b: 12, ab: 36},
  { a: 2, b: 10, ab: 20},
  { a: 1, b: 10, ab: 10}
]

All tidy.js code is wrapped in a tidy flow via the tidy() function. The first argument is the array of data, followed by the transformation verbs to run on the data. The actual functions passed to tidy() can be anything so long as they fit the form:

(items: object[]) => object[]

For example, the following is valid:

tidy(
  data, 
  items => items.filter((d, i) => i % 2 === 0),
  arrange(desc('value'))
)

All tidy verbs fit this style, with the exception of exports from groupBy, discussed below.

Grouping data with groupBy

Besides manipulating flat lists of data, tidy provides facilities for wrangling grouped data via the groupBy() function.

import { tidy, summarize, sum, groupBy } from '@tidyjs/tidy'

const data = [
  { key: 'group1', value: 10 }, 
  { key: 'group2', value: 9 }, 
  { key: 'group1', value: 7 }
]

tidy(
  data,
  groupBy('key', [
    summarize({ total: sum('value') })
  ])
)

The output is:

[
  { "key": "group1", "total": 17 },
  { "key": "group2", "total": 9 },
]

The groupBy() function works similarly to tidy() in that it takes a flow of functions as its second argument (wrapped in an array). Things get really fun when you use groupBy's third argument for exporting the grouped data into different shapes.

For example, exporting data as a nested object, we can use groupBy.object() as the third argument to groupBy().

const data = [
  { g: 'a', h: 'x', value: 5 },
  { g: 'a', h: 'y', value: 15 },
  { g: 'b', h: 'x', value: 10 },
  { g: 'b', h: 'x', value: 20 },
  { g: 'b', h: 'y', value: 30 },
]

tidy(
  data,
  groupBy(
    ['g', 'h'], 
    [
      mutate({ key: d => `\${d.g}\${d.h}`})
    ], 
    groupBy.object() // <-- specify the export
  )
);

The output is:

{
  "a": {
    "x": [{"g": "a", "h": "x", "value": 5, "key": "ax"}],
    "y": [{"g": "a", "h": "y", "value": 15, "key": "ay"}]
  },
  "b": {
    "x": [
      {"g": "b", "h": "x", "value": 10, "key": "bx"},
      {"g": "b", "h": "x", "value": 20, "key": "bx"}
    ],
    "y": [{"g": "b", "h": "y", "value": 30, "key": "by"}]
  }
}

Or alternatively as { key, values } entries-objects via groupBy.entriesObject():

tidy(data,
  groupBy(
    ['g', 'h'], 
    [
      mutate({ key: d => `\${d.g}\${d.h}`})
    ], 
    groupBy.entriesObject() // <-- specify the export
  )
);

The output is:

[
  {
    "key": "a",
    "values": [
      {"key": "x", "values": [{"g": "a", "h": "x", "value": 5, "key": "ax"}]},
      {"key": "y", "values": [{"g": "a", "h": "y", "value": 15, "key": "ay"}]}
    ]
  },
  {
    "key": "b",
    "values": [
      {
        "key": "x",
        "values": [
          {"g": "b", "h": "x", "value": 10, "key": "bx"},
          {"g": "b", "h": "x", "value": 20, "key": "bx"}
        ]
      },
      {"key": "y", "values": [{"g": "b", "h": "y", "value": 30, "key": "by"}]}
    ]
  }
]

It's common to be left with a single leaf in a groupBy set, especially after running summarize(). To prevent your exported data having its values wrapped in an array, you can pass the single option to it.

tidy(input,
  groupBy(['g', 'h'], [
    summarize({ total: sum('value') })
  ], groupBy.object({ single: true }))
);

The output is:

{
  "a": {
    "x": {"total": 5, "g": "a", "h": "x"},
    "y": {"total": 15, "g": "a", "h": "y"}
  },
  "b": {
    "x": {"total": 30, "g": "b", "h": "x"},
    "y": {"total": 30, "g": "b", "h": "y"}
  }
}

Visit the API reference docs to learn more about how each function works and all the options they take. Be sure to check out the levels export, which can let you mix-and-match different export types based on the depth of the data. For quick reference, other available groupBy exports include:

  • groupBy.entries()
  • groupBy.entriesObject()
  • groupBy.grouped()
  • groupBy.levels()
  • groupBy.object()
  • groupBy.keys()
  • groupBy.map()
  • groupBy.values()

Developing

clone the repo:

git clone [email protected]:pbeshai/tidy.git

install dependencies:

yarn

initialize lerna:

lerna bootstrap

build tidy:

yarn run build

test all of tidy:

yarn run test

test:watch a single package

yarn workspace @tidyjs/tidy test:watch

Conventional commits

This library uses conventional commits, following the angular convention. Prefixes are:

  • build: Changes that affect the build system or external dependencies (example scopes: yarn, npm)
  • ci: Changes to our CI configuration files and scripts (e.g. CircleCI)
  • chore
  • docs: Documentation only changes
  • feat : A new feature
  • fix: A bug fix
  • perf: A code change that improves performance
  • refactor: A code change that neither fixes a bug nor adds a feature
  • revert
  • style: Changes that do not affect the meaning of the code (white-space, formatting, missing semi-colons, etc)
  • test: Adding missing tests or correcting existing tests

Docs website

start the local site:

yarn start:web

build the site:

yarn build:web

deploy the site via github-pages:

USE_SSH=true GIT_USER=pbeshai yarn workspace @tidyjs/tidy-website deploy

Ideally we can automate this via github actions one day!


Shout out to Netflix

I want to give a big shout out to Netflix, my current employer, for giving me the opportunity to work on this project and to open source it. It's a great place to work and if you enjoy tinkering with data-related things, I'd strongly recommend checking out our analytics department. – Peter Beshai