Arquero is a JavaScript library for query processing and transformation of array-backed data tables. Following the relational algebra and inspired by the design of dplyr, Arquero provides a fluent API for manipulating column-oriented data frames. Arquero supports a range of data transformation tasks, including filter, sample, aggregation, window, join, and reshaping operations.
- Fast: process data tables with million+ rows.
- Flexible: query over arrays, typed arrays, array-like objects, or Apache Arrow columns.
- Full-Featured: perform a variety of wrangling and analysis tasks.
- Extensible: add new column types or functions, including aggregate & window operations.
- Lightweight: small size, minimal dependencies.
To get up and running, start with the Introducing Arquero tutorial, part of the Arquero notebook collection.
Arquero is Spanish for "archer": if datasets are arrows, Arquero helps their aim stay true.
The core abstractions in Arquero are data tables, which model each column as an array of values, and verbs that transform data and return new tables. Verbs are table methods, allowing method chaining for multi-step transformations. Though each table is unique, many verbs reuse the underlying columns to limit duplication.
import { all, desc, op, table } from 'arquero';
// Average hours of sunshine per month, from https://usclimatedata.com/.
const dt = table({
'Seattle': [69,108,178,207,253,268,312,281,221,142,72,52],
'Chicago': [135,136,187,215,281,311,318,283,226,193,113,106],
'San Francisco': [165,182,251,281,314,330,300,272,267,243,189,156]
});
// Sorted differences between Seattle and Chicago.
// Table expressions use arrow function syntax.
dt.derive({
month: d => op.row_number(),
diff: d => d.Seattle - d.Chicago
})
.select('month', 'diff')
.orderby(desc('diff'))
.print();
// Is Seattle more correlated with San Francisco or Chicago?
// Operations accept column name strings outside a function context.
dt.rollup({
corr_sf: op.corr('Seattle', 'San Francisco'),
corr_chi: op.corr('Seattle', 'Chicago')
})
.print();
// Aggregate statistics per city, as output objects.
// Reshape (fold) the data to a two column layout: city, sun.
dt.fold(all(), { as: ['city', 'sun'] })
.groupby('city')
.rollup({
min: d => op.min(d.sun), // functional form of op.min('sun')
max: d => op.max(d.sun),
avg: d => op.average(d.sun),
med: d => op.median(d.sun),
// functional forms permit flexible table expressions
skew: ({sun: s}) => (op.mean(s) - op.median(s)) / op.stdev(s) || 0
})
.objects()
To use in the browser, you can load Arquero from a content delivery network:
<script src="https://cdn.jsdelivr.net/npm/arquero@latest"></script>
Arquero will be imported into the aq
global object.
Alternatively, you can build and import arquero.min.js
from the build
directory, or build your own application package.
Arquero uses modern JavaScript features, and so will not work with some outdated browsers. To use Arquero with older browsers including Internet Explorer, set up your project with a transpiler such as Babel.
First install arquero
as a dependency, via npm install arquero --save
or yarn add arquero
. Arquero assumes Node version 12 or higher.
Import using CommonJS module syntax:
const aq = require('arquero');
Import using ES module syntax, import all exports into a single object:
import * as aq from 'arquero';
Import using ES module syntax, with targeted imports:
import { op, table } from 'arquero';
To build and develop Arquero locally:
- Clone https://github.com/uwdata/arquero.
- Run
yarn
to install dependencies for all packages. If you don't have yarn installed, see https://yarnpkg.com/en/docs/install. - Run
yarn test
to run test cases, andyarn build
to build output files.