AntV

AntV

Powered by GitBook
Have an account? Sign in

Scale

Last updated 4 months ago
Edit on GitHub
Contents

Scales are a convenient abstraction for a fundamental task in visualization: mapping a dimension of abstract data to a visual representation. Different data types have different type of scales.

  • For continuous quantitative data, you typically want a linear scale.

  • For time series data, a timeCat scale

  • For discrete ordinal (ordered) or categorical (unordered) data, a cat scale specifies an explicit mapping from a set of data values to a corresponding set of visual attributes (such as colors, shapes).

  • A identity scale is for constant.

Scales have no intrinsic visual representation. However, most scales can generate and format ticks for reference marks to aid in the construction of axes.

F2 will generate scales for the data automatically. But still offers two methods to configure.Here is an example：

const data = [
  { State: 'WY', Age: 'Under 5 Years', Population: 25635 },
  { State: 'WY', Age: '5 to 13 Years', Population: 1890 },
  { State: 'WY', Age: '14 to 17 Years', Population: 9314 },
  { State: 'DC', Age: 'Under 5 Years', Population: 30352 },
  { State: 'DC', Age: '5 to 13 Years', Population: 20439 },
  { State: 'DC', Age: '14 to 17 Years', Population: 10225 },
  { State: 'VT', Age: 'Under 5 Years', Population: 38253 },
  { State: 'VT', Age: '5 to 13 Years', Population: 42538 },
  { State: 'VT', Age: '14 to 17 Years', Population: 15757 },
  { State: 'ND', Age: 'Under 5 Years', Population: 51896 },
  { State: 'ND', Age: '5 to 13 Years', Population: 67358 },
  { State: 'ND', Age: '14 to 17 Years', Population: 18794 },
  { State: 'AK', Age: 'Under 5 Years', Population: 72083 },
  { State: 'AK', Age: '5 to 13 Years', Population: 85640 },
  { State: 'AK', Age: '14 to 17 Years', Population: 22153 }
];

In the data, 'State' is a cat scale, 'Age' is a cat scale and 'Population' is a linear scale. Now we can set the maximum and minimum values ​​of Population

chart.source(data, {
  Population: {
    min: 1000,
    max: 100000
  }
});

Or

chart.scale('Population', {
  min: 1000,
  max: 100000
});

Common Configuration

chart.scale('fieldName', {
  // configurations
})

The common configuration properties of scale are described below：

Name

Type

Default

Description

type

String

null

Specify the type of the scale, supported scale types are: identify, linear, cat, timeCat.

formatter

Function

null

It is used to format the text of the scale point in the axis and will affect the display of the data on axis, legend, and tooltip.

range

Array

[ 0, 1 ]

Used to control the drawing range of data on the coordinate, in the format of [ min, max ], and both min and max are data in the range 0 to 1.

alias

String

null

Alias of the data field, if it is set, we will display the alias value in the chart.

tickCount

Number

​

Number of tick points in the axis, different scale types have different default values.

ticks

Array

null

Used to specify the scale text on the axis. When the user sets ticks, it will display according to ticks.

values

Array

null

Set the data collection of the scale, the default will automatically read from the data source, see Cat scale for detail usage.

chart.source(data, {
  value: {
    min: 0,
    ticks: [ 0, 500, 1000, 1300, 1600 ]
  }
});

Except for the above general configuration properties, different scale types have different configuration items.

Linear

Name

Type

Default

Description

nice

Boolean

true

used to optimize the range of values so that ticks on axes are evenly distributed. For example, the range [ 3, 97 ] will be optimized to [ 0, 100 ], if nice is set to be true.

min

Number

​

the minimum value of the data set

max

Number

​

the maximum value of the data set

tickInterval

Number

​

Used to specify the distance between the scale points of the axis, which is the difference between the original data. Note thattickCountandtickInterval cannot both be defined.

tickInterval explanation
  const data = [
    { year: '1951 年', sales: 38 },
    { year: '1952 年', sales: 52 },
    { year: '1956 年', sales: 61 },
    { year: '1957 年', sales: 145 },
    { year: '1958 年', sales: 48 },
    { year: '1959 年', sales: 38 },
    { year: '1960 年', sales: 38 },
    { year: '1962 年', sales: 38 },
  ];
  const chart = new F2.Chart({
    id: 'mountNode',
    pixelRatio: window.devicePixelRatio
  });
​
  chart.source(data, {
    sales: {
      tickInterval: 80 // set tickInterval
    }
  });
  chart.interval().position('year*sales');
  chart.render();

Cat

Cat scale has no other configuration items. But we will introduce the usage of values ​​here.

1. When you need to specify the display order of the categorical data.

We use the following data to draw a bar chart:

  const data = [
    { name: 'A', value: 38 },
    { name: 'B', value: 52 },
    { name: 'C', value: 61 },
  ];
  const chart = new F2.Chart({
    id: 'mountNode',
    pixelRatio: window.devicePixelRatio
  });
​
  chart.source(data);
  chart.interval().position('name*value').color('name');
  chart.render();

Now we can change the display order of the bar by setting values, including legend:

chart.source(data, {
  name: {
    type: 'cat', // In fact, you can also do not need to set, F2 will automatically recognize
    values: [ 'C', 'B', 'A' ]
  }
});

After set this, the bar chart will be rendered as follow:

2. Convert index values ​​to corresponding data. But the original value of the field must be indexed value, start from 0.

  const data = [
    { month: 0, tem: 7, city: 'Tokyo' },
    { month: 1, tem: 6.9, city: 'Tokyo' },
    { month: 2, tem: 9.5, city: 'Tokyo' },
    { month: 3, tem: 14.5, city: 'Tokyo' },
    { month: 4, tem: 18.2, city: 'Tokyo' },
    { month: 5, tem: 21.5, city: 'Tokyo' },
    { month: 6, tem: 25.2, city: 'Tokyo' }
  ];
  const chart = new F2.Chart({
    id: 'mountNode',
    width: 375,
    height: 260,
    pixelRatio: window.devicePixelRatio
  });
  chart.source(data, {
    month: {
      type: 'cat',
      values: ['Jan.', 'Feb.', 'Mar.', 'Apr.', 'May.', 'Jun.', 'Jul.']
    }
  });
  chart.interval().position('month*tem');
  chart.render();

3. Control the display of data.

Case 1: we just have 3 items in data, but we can set 10 ticks on the axis:

  const data = [
    { name: 'A', value: 38 },
    { name: 'B', value: 52 },
    { name: 'C', value: 61 },
  ];
  const chart = new F2.Chart({
    id: 'mountNode',
    pixelRatio: window.devicePixelRatio
  });
​
  chart.source(data, {
    name: {
      values: [ 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K' ]
    }
  });
  chart.interval().position('name*value');
  chart.render();

Case 2: we just have 3 items in data, but we can just want to display one:

  const data = [
    { name: 'A', value: 38 },
    { name: 'B', value: 52 },
    { name: 'C', value: 61 },
  ];
  const chart = new F2.Chart({
    id: 'mountNode',
    pixelRatio: window.devicePixelRatio
  });
​
  chart.source(data, {
    name: {
      values: [ 'B' ]
    }
  });
  chart.interval().position('name*value');
  chart.render();

TimeCat

For time series data, we will sort the data set by default.

Name

Type

Default

Description

mask

String

'YYYY-MM-DD'

To Format the display of date, we use fecha lib, the format tokens can see here​

NOTE: mask and formmater cannot both be specified, if both are defined, the formatter attribute will be used in priority.

Previous
Geometry
Next
Coordinate
Contents