qix-faker

Powered by faker.js

Generate fake Qlik Engine data in the shape of hypercubes and listobjects.

Great for prototyping with picasso.js, nebula.js and any other time when a real Qlik Engine is not available.

Usage

node.js

npm install qix-faker

const { hypercube, listobject } = require('qix-faker');

const hc = hypercube({
  dimensions: [f => f.commerce.product()],
  measures: [f => f.commerce.price(100, 5000, 2, '$')],
  numRows: 3,
});

const lo = listobject({
  dimension: [f => f.commerce.product()],
  numRows: 5,
});

API

qixFaker.hypercube(options)

• options <object>
  • dimensions <Array<field>> - Dimensions.
  • measures <Array<field>> - Measures.
  • numRows <number> - Number of rows to generate.
  • seed <number> - Set the same number to generate consistent results.

qixFaker.hypercube({
  numRows: 10,
  seed: 5,
  dimensions: [
    {
      value: (faker, rowIdx) => faker.date.month(),
    },
  ],
});

qixFaker.listobject(options)

• options <object>
  • dimension <field>.
  • numRows <number> - Number of rows to generate.
  • seed <number> - Set the same number to generate consistent results.

qixFaker.listobject({
  numRows: 10,
  seed: 5678,
  dimension: f => d.name.firstName(),
});

types

field <function | object>

A field configuration can take on two different shapes:

<function(faker, rowIndex)>

A function which is provided the faker instance as first parameter, and the rowIndex as the second. See the faker api for details on params of each method.

hypercube({
  dimensions: [(faker, idx) => faker.commerce.product()],
  measures: [(faker, idx) => faker.finance.amount(100, 3000, 2, '$')],
});

<object>

Using an object provides more control:

• value: <function(faker, rowIndex)> - Same function as above.
• maxCardinalRatio <number> - A value between 0 - 1 to limit the uniqueness of dimension values.
• attrDims <Array<field>> - Attribute dimensions.
• attrExps <Array<field> - Attribute expression.
• override <object> - Set custom properties on the field.

hypercube({
  numRows: 100,
  dimensions: [
    {
      value: f => f.address.city(),
      maxCardinalRatio: 0.4,
      attrDim: [f => f.commerce.color()],
      attrExps: [f => f.random.number()]
      override: {
        qFallbackTitle: 'City',
        qLocked: true,
      },
    },
  ],
});

In the above example, 100 rows of data is generated using the city dataset provided in faker, there is however no guarantee that all 100 rows will be unique.

In some cases though, it might be desirable to limit the uniqueness of the generated values; maxCardinalRatio provides a way to do just that.

Assume we want to generate some data containing cities grouped by country, by setting maxCardinalRatio to a low number we can create such a dataset.

hypercube({
  numRows: 80,
  dimensions: [
    {
      value: f => f.address.country(),
      maxCardinalRatio: 0.1,
    },
    {
      value: f => f.address.city(),
    },
  ],
});

In the above example, the first 8 rows (numRows * ratio) of country will be the same, while city will be randomized, thus creating groups.