Axes are an integral part of a chart. They are used to determine how data maps to a pixel value on the chart. In a cartesian chart, there is 1 or more X-axis and 1 or more Y-axis to map points onto the 2-dimensional canvas. These axes are known as ['cartesian axes'](./cartesian/).
In a radial chart, such as a radar chart or a polar area chart, there is a single axis that maps points in the angular and radial directions. These are known as ['radial axes'](./radial/).
The default `scaleId`'s for carterian charts are `'x'` and `'y'`. For radial charts: `'r'`.
Each dataset is mapped to a scale for each axis (x, y or r) it requires. The scaleId's that a dataset is mapped to, is determined by the `xAxisID`, `yAxisID` or `rAxisID`.
If the ID for an axis is not specified, first scale for that axis is used. If no scale for an axis is found, a new scale is created.
Some examples:
The following chart will have `'x'` and `'y'` scales:
```js
let chart = new Chart(ctx, {
type: 'line'
});
```
The following chart will have scales `'x'` and `'myScale'`:
```js
let chart = new Chart(ctx, {
type: 'bar',
data: {
datasets: [{
data: [1, 2, 3]
}]
},
options: {
scales: {
myScale: {
type: 'logarithmic',
position: 'right', // `axis` is determined by the position as `'y'`
}
}
}
});
```
The following chart will have scales `'xAxis'` and `'yAxis'`:
```js
let chart = new Chart(ctx, {
type: 'bar',
data: {
datasets: [{
yAxisID: 'yAxis'
}]
},
options: {
scales: {
xAxis: {
// The axis for this scale is determined from the first letter of the id as `'x'`
// It is recommended to specify `position` and / or `axis` explicitly.
These are only the common tick options supported by all axes. Please see specific axis documentation for all of the available tick options for that axis.
Given the number of axis range settings, it is important to understand how they all interact with each other.
The `suggestedMax` and `suggestedMin` settings only change the data values that are used to scale the axis. These are useful for extending the range of the axis while maintaining the auto fit behaviour.
```javascript
let minDataValue = Math.min(mostNegativeValue, options.suggestedMin);
let maxDataValue = Math.max(mostPositiveValue, options.suggestedMax);
```
In this example, the largest positive value is 50, but the data maximum is expanded out to 100. However, because the lowest data value is below the `suggestedMin` setting, it is ignored.
In contrast to the `suggested*` settings, the `min` and `max` settings set explicit ends to the axes. When these are set, some data points may not be visible.
By default data is not stacked. If the `stacked` option of the value scale (y-axis on horizontal chart) is `true`, positive and negative values are stacked separately. Additionally a `stack` option can be defined per dataset to further divide into stack groups [more...](../general/data-structures/#dataset-configuration).
For some charts, you might want to stack positive and negative values together. That can be achieved by specifying `stacked: 'single'`.
There are a number of config callbacks that can be used to change parameters in the scale at different points in the update process. The options are supplied at the top level of the axis options.
For example, to set the minimum value of 0 for all linear scales, you would do the following. Any linear scales created after this time would now have a minimum of 0.