README
@visx/xychart
In contrast to other visx
packages which are low-level, this package seeks to abstract some of the
complexity of common visualization engineering, and exposes a high-level x,y (cartesian
coordinate) chart API. However, it is implemented using modularized React.context
layers for
theme, canvas dimensions, x/y/color scales, data, events, and tooltips which allows for more
expressivity and advanced use cases.
Out of the box it supports the following:
- * many common
<*Series />
types (animated or not) such as lines, bars, etc. - *
<Axis />
(animated or not) - *
<Grid />
(animated or not) - *
<Annotation />
(animated or not) - *
<Tooltip />
- *
theme
ing
The following illustrates basic usage to create an animated line chart with a bottom Axis
, Grid
,
and Tooltip
:
import {
AnimatedAxis, // any of these can be non-animated equivalents
AnimatedGrid,
AnimatedLineSeries,
XYChart,
Tooltip,
} from '@visx/xychart';
const data1 = [
{ x: '2020-01-01', y: 50 },
{ x: '2020-01-02', y: 10 },
{ x: '2020-01-03', y: 20 },
];
const data2 = [
{ x: '2020-01-01', y: 30 },
{ x: '2020-01-02', y: 40 },
{ x: '2020-01-03', y: 80 },
];
const accessors = {
xAccessor: d => d.x,
yAccessor: d => d.y,
};
const render = () => (
<XYChart height={300} xScale={{ type: 'band' }} yScale={{ type: 'linear' }}>
<AnimatedAxis orientation="bottom" />
<AnimatedGrid columns={false} numTicks={4} />
<AnimatedLineSeries dataKey="Line 1" data={data1} {...accessors} />
<AnimatedLineSeries dataKey="Line 2" data={data2} {...accessors} />
<Tooltip
snapTooltipToDatumX
snapTooltipToDatumY
showVerticalCrosshair
showSeriesGlyphs
renderTooltip={({ tooltipData, colorScale }) => (
<div>
<div style={{ color: colorScale(tooltipData.nearestDatum.key) }}>
{tooltipData.nearestDatum.key}
</div>
{accessors.xAccessor(tooltipData.nearestDatum.datum)}
{', '}
{accessors.yAccessor(tooltipData.nearestDatum.datum)}
</div>
)}
/>
</XYChart>
);
See sections below for more detailed guidance and advanced usage, or explore the comprehensive API below.
Basic usage
Installation
npm install --save @visx/xychart react-spring
Note: react-spring
is a required peerDependency
for importing Animated*
components.
Series types
The following Series
types are currently supported and we are happy to review or consider
additional Series types in the future.
Component name | Description | Usage |
---|---|---|
(Animated)AreaSeries | Connect data points with a <path /> , with a color fill to the zero baseline |
<AreaSeries /> |
(Animated)BarSeries | Render a <rect /> for each data point |
<BarSeries /> |
(Animated)BarGroup | Group multiple child <BarSeries /> values together |
<BarGroup><BarSeries /><BarSeries />...</BarGroup> |
(Animated)BarStack | Stack multiple child <BarSeries /> values together |
<BarStack><BarSeries /><BarSeries />...</BarStack> |
(Animated)GlyphSeries | Render a Glyph (any shape, defaults to <circle /> ) for each data point, e.g., a scatter plot |
<GlyphSeries renderGlyph={() => ...} /> |
(Animated)LineSeries | Connect data points with a <path> |
<GlyphSeries /> |
All Series
have animated and non-animated variants to give you more control over your bundle size,
support missing (null
) data, and can be rendered vertically or horizontally.
Theming
Default lightTheme
and darkTheme
themes are exported from @visx/xychart
and the utility
buildChartTheme
is exported to support easy creation of custom themes.
import { buildChartTheme, XYChart } from '@visx/xychart';
import { TextProps as SVGTextProps } from '@visx/text/lib/Text'; // just for types
const customTheme = buildChartTheme({
// colors
backgroundColor: string; // used by Tooltip, Annotation
colors: string[]; // categorical colors, mapped to series via `dataKey`s
// labels
svgLabelBig?: SVGTextProps;
svgLabelSmall?: SVGTextProps;
htmlLabel?: HTMLTextStyles;
// lines
xAxisLineStyles?: LineStyles;
yAxisLineStyles?: LineStyles;
xTickLineStyles?: LineStyles;
yTickLineStyles?: LineStyles;
tickLength: number;
// grid
gridColor: string;
gridColorDark: string; // used for axis baseline if x/yxAxisLineStyles not set
gridStyles?: CSSProperties;
});
() => <XYChart theme={customTheme} />
Tooltips
@visx/tooltip
Tooltip
s are integrated into @visx/xychart
, and should be rendered as a child of
XYChart
(or a child where TooltipContext
is provided).
Tooltip
positioning is handled by the Tooltip
itself, based on TooltipContext
. Tooltip
is rendered inside a Portal
, avoiding clipping by parent DOM elements with higher z-index
contexts. See the API below for a full list of props
to support additional behavior, such as
snapping to data point positions and rendering cross-hairs.
Tooltip
content is controlled by the specified prop.renderTooltip
which has access to:
tooltipData.nearestDatum
– the globally closestDatum
, across allSeries
'sdataKey
stooltipData.datumByKey
– the closestDatum
for eachSeries
'sdataKey
; this enables "shared tooltips" where you can render the nearest data point for eachSeries
.- a shared
colorScale
which mapsSeries
'sdataKey
s totheme
colors
Event handlers
The following PointerEvent
s (handling both MouseEvent
s and TouchEvent
s) are currently
supported. They may be set on individual Series
components (e.g.,
<BarSeries onPointerMove={() => ...} />
), or at the chart level (e.g.,
<XYChart onPointerMove={() => {}} />
) in which case they are invoked once for every *Series
.
To disable event emitting for any Series
set <*Series enableEvents=false />
. The
onFocus/onBlur
handlers enable you to make your chart events and Tooltip
s accessible via
keyboard interaction. Note that the current implementation requires your target browser to support
the SVG 2.0
spec for tabIndex
on SVG
elements.
Below, HandlerParms
has the following type signature:
type EventHandlerParams<Datum> = {
datum: Datum; // nearest Datum to event, for Series with `dataKey=key`
distanceX: number; // x distance between event and Datum, in px
distanceY;: number; // y distance between event and Datum, in px
event: React.PointerEvent | React.FocusEvent; // the event
index: number; // index of Datum in Series `data` array
key: string; // `dataKey` of Series to which `Datum` belongs
svgPoint: { x: number; y: number }; // event position in svg-coordinates
};
Prop name | Signature | XYChart support |
*Series support |
---|---|---|---|
onPointerMove |
(params: EventHandlerParams<Datum>) => void |
✅ | ✅ |
onPointerOut |
(event: React.PointerEvent) => void |
✅ | ✅ |
onPointerUp |
(params: EventHandlerParams<Datum>) => void |
✅ | ✅ |
onFocus |
(params: EventHandlerParams<Datum>) => void |
❌ | ✅ |
onBlur |
(event: React.TouchEvent) => void |
❌ | ✅ |
Annotations
Composable @visx/annotations
annotations are integrated into @visx/xychart
and use its theme and
dimension context. These components allow for annotation of individual points using
AnnotationCircleSubject
, or x- or y-thresholds using AnnotationLineSubject
.
import React from 'react';
import {
Annotation,
AnnotationLabel,
AnnotationConnector,
AnnotationCircleSubject,
Grid,
LineSeries,
XYChart,
} from '@visx/xychart';
const data = [
{ x: '2020-01-01', y: 50 },
{ x: '2020-01-02', y: 10 },
{ x: '2020-01-03', y: 20 },
{ x: '2020-01-04', y: 5 },
];
const labelXOffset = -40;
const labelYOffset = -50;
const chartConfig = {
xScale: { type: 'band' },
yScale: { type: 'linear' },
height: 300,
margin: { top: 10, right: 10, bottom: 10, left: 10 },
};
export default () => (
<XYChart {...chartConfig}>
<Grid numTicks={3} />
<LineSeries dataKey="line" data={data} xAccessor={d => d.x} yAccessor={d => d.y} />
<Annotation
dataKey="line" // use this Series's accessor functions, alternatively specify x/yAccessor here
datum={data[2]}
dx={labelXOffset}
dy={labelYOffset}
>
{/** Text label */}
<AnnotationLabel
title="Title"
subtitle="Subtitle deets"
showAnchorLine={false}
backgroundFill="rgba(0,150,150,0.1)"
/>
{/** Draw circle around point */}
<AnnotationCircleSubject />
{/** Connect label to CircleSubject */}
<AnnotationConnector />
</AnimatedAnnotation>
</XYChart>
);
Advanced usage
Examples
XYChart
is implemented using modularized React.context
layers for scales, canvas dimensions,
data, events, and tooltips which enables more advanced usage than many other chart-level
abstractions.
By default XYChart
renders all context providers if a given context is not available, but you can
share context across multiple XYChart
s to implement functionality such as linked tooltips, shared
themes, or shared data.
DataContext
This context provides chart canvas dimensions (width
, height
, and margin
), x/y/color scales,
and a data registry. The data registry includes data from all child *Series
, and x/y/color scales
are updated accordingly accounting for canvas dimensions.
ThemeContext
This context provides an XYChart
theme, its used by all visual elements that compose a chart, and
can be used to render custom visual elements that are on theme.
EventEmitterContext
This context provides an event publishing / subscription object which can be used via the
useEventEmitter
hook. Series
and XYChart
events, including tooltip updates, are emitted and
handled with through this context.
import React, { useState } from 'react';
import { useEventEmitter, EventEmitterProvider } from '@visx/xychart';
const eventSourceId = 'optional-source-id-filter';
const EmitEvent = () => {
const emit = useEventEmitter();
return <button onPointerUp={event => emit('pointerup', event, eventSourceId)}>emit event</button>;
};
const SubscribeToEvent = () => {
const [clickCount, setClickCount] = useState(0);
const allowedEventSources = [eventSourceId];
useEventEmitter('pointerup', () => setClickCount(clickCount + 1), allowedEventSources);
return <div>Emitted {clickCount} events</div>;
};
export default function Example() {
return (
<EventEmitterProvider>
<EmitEvent />
<SubscribeToEvent />
</EventEmitterProvider>
);
}
TooltipContext
This context provides access to @visx/tooltip
s useTooltip
state, including whether the tooltip
is visible (tooltipOpen
), tooltlip position (tooltipLeft
, tooltipTop
),
tooltipData: { nearestDatum, datumByKey }
described above, and functions to update context
(hideTooltip
, showTooltip
, and updateTooltip
).
ResizeObserver
dependency
⚠️ The Tooltip
and AnnotationLabel
components rely on
ResizeObserver
s. If your
browswer target needs a polyfill, you can either polute the window
object or inject it cleanly
using the resizeObserverPolyfill
prop for these components.
Examples
❌ Error: This browser does not support ResizeObserver out of the box
// no polyfill, no browser support
() => <XYChart {...}><Tooltip /></XYChart>
✅ No errors
// no polyfill, target browser supports ResizeObserver
() => <XYChart {...}><Tooltip /></XYChart>
// import the polyfill in the needed module, or set it on `window` object
import ResizeObserver from 'resize-observer-polyfill';
() => <XYChart {...}><Tooltip /></XYChart> // 😎
// cleanly pass polyfill to component that needs it
import ResizeObserver from 'resize-observer-polyfill';
() => (
<XYChart {...}>
<Tooltip resizeObserverPolyfill={ResizeObserver} />
</XYChart>
)