@formily/react-schema-renderer
v1.3.17
Published
English | [简体中文](./README.zh-cn.md)
Downloads
1,265
Readme
@formily/react-schema-renderer
English | 简体中文
Schema rendering engine, the package mainly relies on @formily/react, its responsibilities are very simple, the core does two things:
- Parse the Form Schema protocol, recursive rendering
- Manage custom components
Installation
npm install --save @formily/react-schema-renderer
Contents
- Usage
- Advanced tutorial
- API
- Classes
- new Schema(json : ISchema)
get
merge
getEmptyValue
getSelfProps
getExtendsRules
getExtendsRequired
getExtendsEditable
getExtendsVisible
getExtendsDisplay
getExtendsTriggerType
getExtendsProps
getExtendsComponent
getExtendsComponentProps
getExtendsRenderer
getExtendsEffect
setProperty
setProperties
setArrayItems
getOrderProperties
mapProperties
toJSON
fromJSON
isObject
isArray()
- new Schema(json : ISchema)
- Components
- Interfaces
Usage
Quick start
If you are developing directly based on @formily/react-schema-renderer, then you must register your custom component before development. Go to the renderer, otherwise our JSON-Schema protocol can't render the form. And so:
import React from 'react'
import {
SchemaForm,
registerFormField,
connect
} from '@formily/react-schema-renderer'
registerFormField(
'string',
connect()(({ value, onChange }) => {
return <input value={value} onChange={onChange} />
})
)
export default () => {
return (
<SchemaForm
schema={{
type: 'object',
properties: {
input: {
type: 'string'
}
}
}}
onChange={console.log} //{input:'this is your input string'}
/>
)
}
Great workmanship, this is the easiest way to use, the core is to register components, and then use the protocol to render.**One thing to note is that we used the connect function when registering components,The function of this connect function is to allow any component to be registered in the SchemaForm as long as it supports the value/onChange API.At the same time, the connect function also shields the Field API, so components that use the connect function cannot be more powerfully extended. The detailed connect API will be introduced later. ** At the same time, one thing to note is that if we want to access a set of component libraries, most of the component libraries in the industry have their Form and FormItem components.Their core is used to control styles, FormItem controls form local styles, and Form controls global form styles,So in a production environment, we need to register the Form and FormItem components so that the style is consistent with the original solution.How to register specifically, we will introduce in detail later.
JSON Schema driver form rendering
Speaking of JSON Schema, the above example has been covered. Of course, it is not complicated enough. Let's look at a more complicated example:
import React from 'react'
import { SchemaForm } from '@formily/react-schema-renderer'
registerFormField(
'string',
connect()(({ value, onChange }) => {
return <input value={value} onChange={onChange} />
})
)
registerFormField('array', () => {
return //...
})
export default () => {
return (
<SchemaForm
schema={{
type: 'object',
properties: {
array: {
type: 'array',
items: {
type: 'object',
properties: {
input: {
type: 'string'
}
}
}
}
}
}}
onChange={console.log} //{array:[{input:'This is your input string'}]}
/>
)
}
The above code is a pseudo-code because we didn't register a custom component of type array. Here we don't talk about how to register custom components of the array type. Our core is to analyze how JSON Schema drives form rendering.Note:In SchemaForm, recursive rendering of properties of an object is built-in, but there is no recursive rendering of items with a built-in array.The main reason is that the recursive rendering of an array involves many style requirements, which is not convenient for built-in. So it’s best to leave it to the developer to implement it yourself. So, we will detail how to implement the recursive rendering requirements of the auto-increment list.
JSchema driver form rendering
JSchema describes JSON Schema in a more elegant way in jsx. We can implement a version of JSchema for the above example:
import React from 'react'
import {
SchemaMarkupForm as SchemaForm,
SchemaMarkupField as Field
} from '@formily/react-schema-renderer'
export default () => {
return (
<SchemaForm
onChange={console.log} //{array:[{input:'This is your input string'}]}
>
<Field type="array" name="array">
<Field type="object">
<Field type="string" name="input" />
</Field>
</Field>
</SchemaForm>
)
}
As you can see, using JSchema to describe JSON Schema in your code is more elegant than JSON, greatly improving code maintainability.
Non-single instance registration component
In the previous example, we used the registerFormField API to register a custom component, which is the way to register a singleton,Its main advantage is the convenience, but there are also some problems, that is, the single case is easily contaminated, especially in the SPA page.If the developers of different pages are different, because sharing the same memory environment, then A developers may register custom components with the same name as B developers, which can easily lead to online failures.Therefore, we recommend users to use non-single registration methods:
import React, { useMemo } from 'react'
import {
SchemaForm,
registerFormField,
connect
} from '@formily/react-schema-renderer'
const StringField = connect()(({ value, onChange }) => {
return <input value={value} onChange={onChange} />
})
const useFields = () =>
useMemo(() => {
string: StringField
})
export default () => {
return (
<SchemaForm
fields={useFields()}
schema={{
type: 'object',
properties: {
input: {
type: 'string'
}
}
}}
onChange={console.log} //{input:'this is your input string'}
/>
)
}
In the above example, we mainly pass the custom component in the props dimension of SchemaForm, so that we can guarantee the instance-level registration. This form is very friendly to the SPA.At the same time,It should be noted that we abstracted a React Hook of useFields, which is mainly used to solve the React Virtual DOM recalculation when the component is rendered multiple times, thus avoiding the repeated rendering of the form component.
Advanced tutorial
How to access third-party component libraries?
Because @formily/react-schema-renderer is a base library, no component library is integrated by default. so in actual business development, If you want to customize it based on it, you have to face the problem of accessing third-party component libraries. How to access the third-party component library, let's divide it into the following steps:
- Access the Form/FormItem component
- Access component library form component
- Implement form layout components
- Implement an auto-increment list component
Let's take a step by step access to third-party component libraries! Here we mainly use the antd component library as an example.
How to access the Form/FormItem component?
The access method currently provides a global registration mechanism and a singleton registration mechanism. Global registration is mainly registered using two APIs, registerFormComponent and registerFormItemComponent.Singleton registration is to upload formComponent and formItemComponent directly in the SchemaForm property. If it is a SPA scene, it is recommended to use the singleton registration method. Let's look at the example below:
import React from 'react'
import {
SchemaForm,
registerFormComponent,
registerFormItemComponent
} from '@formily/react-schema-renderer'
import { Form } from 'antd'
export const CompatFormComponent = ({ children, ...props }) => {
return <Form {...props}>{children}</Form> //Very simple to use the Form component, props is the props of the SchemaForm component, here will be directly transparent
}
export const CompatFormItemComponent = ({ children, ...props }) => {
const messages = [].concat(props.errors || [], props.warnings || [])
let status = ''
if (props.loading) {
status = 'validating'
}
if (props.invalid) {
status = 'error'
}
if (props.warnings && props.warnings.length) {
status = 'warning'
}
return (
<Form.Item
{...props}
label={props.schema.title}
help={
messages.length ? messages : props.schema && props.schema.description
}
validateStatus={status}
>
{children}
</Form.Item>
)
}
/***
Global registration method
registerFormComponent(CompatFormComponent)
registerFormItemComponent(CompatFormItemComponent)
***/
//Single case registration method
export default () => {
return (
<SchemaForm
formComponent={CompatFormComponent}
formItemComponent={CompatFormItemComponent}
/>
)
}
We can see that extending the overall or partial style of the form can be easily solved by simply extending the Form/FormItem component. It should be noted here that the props received by the FormItem component are a bit complicated. Don't worry, the detailed props API will be listed later. Now we just need to know how it is probably registered.
How to access form components?
all components of the component library are atomic components, most of them are compatible with the value/onChange specification, so we can quickly access the components of the component library with the connect function. Usually, if we need access to the component library component, we probably do three things:
- Handle state maps, map the loading/error state inside the formily to the component properties, of course, the premise is that the component must support styles such as loading or error.
- Handle detail styles, map the editable state inside the formily to a PreviewText component for more friendly and cleaner presentation data.
- Handle component enumeration, let's think about it, JSON Schema, each node should support the enum attribute, If we have the enum attribute, we'd better show it in the form of Select, so we need to handle the component enumeration state.
Let's take an example of InputNumber:
import React from 'react'
import { connect, registerFormField } from '@formily/react-schema-renderer'
import { InputNumber } from 'antd'
const mapTextComponent = (
Target: React.JSXElementConstructor<any>,
props: any = {},
fieldProps: any = {}
): React.JSXElementConstructor<any> => {
const { editable } = fieldProps
if (editable !== undefined) {
if (editable === false) {
return PreviewText
}
}
if (Array.isArray(props.dataSource)) {
return Select
}
return Target
}
const mapStyledProps = (
props: IConnectProps,
fieldProps: MergedFieldComponentProps
) => {
const { loading, errors } = fieldProps
if (loading) {
props.state = props.state || 'loading'
} else if (errors && errors.length) {
props.state = 'error'
}
}
const acceptEnum = (component: React.JSXElementConstructor<any>) => {
return ({ dataSource, ...others }) => {
if (dataSource) {
return React.createElement(Select, { dataSource, ...others })
} else {
return React.createElement(component, others)
}
}
}
registerFormField(
'number',
connect({
getProps: mapStyledProps, //Process state map
getComponent: mapTextComponent //Process details
})(acceptEnum(InputNumber)) //Process enumeration states
)
In this example, we use the connect function in depth. Connect is a HOC. During the rendering phase, it can add some intermediate processing logic to the component rendering process to help dynamic distribution. Of course, connect has a lot of APIs, which will be covered in detail later.
How to handle form layout?
JSON Schema describes the form of data structure, which is supported by nature. But the form ends up at the UI level. Unfortunately, at the UI level, we have a lot of components that are not a specific data node of JSON Schema. It's just a UI node. So, how to do to describe complex layouts in JSON Schema?
Now formily's approach is to abstract a concept called virtual nodeAfter the user specifies a JSON Schema x-component as a virtual node at the code level, whether it is rendering, data processing, or final data submission, as long as the node is virtual, it will not be treated as a normal data node. So, with the concept of this virtual node, we can describe various complex layouts in JSON Schema. Let's try to write a layout component:
import React from 'react'
import { SchemaForm, registerVirtualBox } from '@formily/react-schema-renderer'
import { Card } from 'antd'
registerVirtualBox('card', ({ children, ...props }) => {
return <Card {...props.schema.getExtendsComponentProps()}>{children}</Card>
})
export default () => {
return (
<SchemaForm
schema={{
type: 'object',
properties: {
layout_card: {
//layout_cardThe name of this property, whatever you change will not affect the final submitted data structure
type: 'object',
'x-component': 'card',
'x-component-props': {
title: 'This is Card'
},
properties: {
array: {
type: 'array',
items: {
type: 'object',
properties: {
input: {
type: 'string'
}
}
}
}
}
}
}
}}
/>
)
}
From this pseudo-code, we can see that the card is a normal Object Schema node, just need to specify an x-component as card, so that it can match the card registered by registerVirtualBox, it will achieve the effect of virtual node. So, no matter what you change the name of the property in JSON Schema, it will not affect the final submitted data structure.It should be noted here that x-component-props is directly passed to the callback function parameters of registerVirtualBox. This is the use of the JSON Schema form, and we also use JSchema:
import React from 'react'
import { SchemaForm, createVirtualBox } from '@formily/react-schema-renderer'
import { Card } from 'antd'
const Card = createVirtualBox('card', ({ children, ...props }) => {
return <Card {...props}>{children}</Card>
})
export default () => {
return (
<SchemaForm>
<Card title="This is Card">
<Field type="array" name="array">
<Field type="object">
<Field type="string" name="input" />
</Field>
</Field>
</Card>
</SchemaForm>
)
}
From this example we can see that with the createVirtualBox API you can quickly create a layout component that you can use directly in JSchema.In fact, the internal implementation of createVirtualBox is very simple, just use registerVitualBox and SchemaMarkupField:
import React from 'react'
import { SchemaMarkupField as Field } from '@formily/react-schema-renderer'
export function createVirtualBox<T = {}>(
key: string,
component?: React.JSXElementConstructor<React.PropsWithChildren<T>>
) {
registerVirtualBox(
key,
component
? ({ props, schema, children }) => {
return React.createElement(component, {
...schema.getExtendsComponentProps(),
children
})
}
: () => <Fragment />
)
const VirtualBox: React.FC<T & { name?: string }> = ({
children,
name,
...props
}) => {
return (
<Field
type="object"
name={name}
x-component={key}
x-props={props}
x-component-props={props}
>
{children}
</Field>
)
}
return VirtualBox
}
The way to register the layout components described above is a singleton registration. If we need registration in the form of an instance, it is still similar to the way we said earlier.
import React from 'react'
import { Card as AntdCard } from 'antd'
const Card = ({children,...props})=>{
return <AntdCard {...props.schema.getExtendsComponentProps()}>{children}</Card>
}
export default ()=>{
return (
<SchemaForm
virtualFields={{
card:Card
}}
schema={{
type:"object",
properties:{
layout_card:{//layout_cardThe name of this property, whatever you change will not affect the final submitted data structure
type:"object",
"x-component":"card",
"x-component-props":{
title:"This is Card"
},
properties:{
array:{
type:"array",
items:{
type:"object",
properties:{
input:{
type:"string"
}
}
}
}
}
}
}
}}
/>
)
}
How to implement recursive rendering components?
What is a recursive rendering component? In fact, implement the components of properties and items in JSON Schema,sucn astype:"string"
This kind of node belongs to the atomic node and is not a recursive rendering component. The layout component mentioned above, it is also a recursive rendering component, but it fixes the rendering mode, so it can be easily registered. So, most of us want to implement recursive rendering of the scene,Perhaps more of a scene in`type:"array"will implement recursive rendering. Below we will detail the implementation of the auto-increment list component.
How to implement the auto-increment list component?
The self-increment list has several main features:
- Independent style
- Support for recursive rendering of subcomponents
- Support for adding, deleting, moving up, down moving array items
- Cannot be wrapped with the connect function because the Field API must be called
In order to help you better understand how to implement the auto-increment list component, we will not implement the specific style, and more is to teach you how to implement recursive rendering and array item operations. Let us look at the pseudo-code:
import React, { Fragment } from 'react'
import {
registerFormField,
SchemaField,
FormPath
} from '@formily/react-schema-renderer'
//No need to pack with connet
registerFormField('array', ({ value, path, mutators }) => {
const emptyUI = (
<button
onClick={() => {
mutators.push()
}}
>
Add Element
</button>
)
const listUI = value.map((item, index) => {
return (
<div key={index}>
<SchemaField path={FormPath.parse(path).concat(index)} />
<button
onClick={() => {
mutators.remove(index)
}}
>
remove
</button>
<button
onClick={() => {
mutators.moveDown(index)
}}
>
move down
</button>
<button
onClick={() => {
mutators.moveUp(index)
}}
>
move up
</button>
</div>
)
})
return value.length == 0 ? emptyUI : listUI
})
To implement a self-incrementing list component with recursive rendering is super simple. On the contrary, it would be a bit cumbersome to implement related styles. In short, the core is to use the SchemaField component and the mutators API, the specific API will be described in detail later.
How to implement ultra-complex custom components?
This problem is no solution in the old Formily.It happens that because our business complexity is high to a certain extent, we are limited by this problem, so we must find a way to solve this problem. We can define what is a super complex custom component:
There are a large number of form components inside the component, and there are also a large number of linkage relationships inside.
There is a private server dynamic rendering scheme inside the component.
There is a complex layout structure inside the component
In these three points, meet the characteristics of ultra-complex custom components, For this scenario, why can't we solve the problem by properly encapsulating the form of a custom component? In fact, it is mainly limited to verification, there is no way to check the whole, so we need an ability to aggregate a large amount of field processing logic. Let's take a look at the specific solution:
import React, { Fragment } from 'react'
import {
registerFormField,
SchemaField,
FormPath,
InternalField,
useFormEffects,
FormEffectHooks
} from '@formily/react-schema-renderer'
import { Input, Form } from 'antd'
const FormItem = ({ component, ...props }) => {
return (
<InternalField {...props}>
{({ state, mutators }) => {
const messages = [].concat(state.errors || [], state.warnings || [])
let status = ''
if (state.loading) {
status = 'validating'
}
if (state.invalid) {
status = 'error'
}
if (state.warnings && state.warnings.length) {
status = 'warning'
}
return (
<Form.Item
{...props}
help={messages.length ? messages : props.help && props.help}
validateStatus={status}
>
{React.createElement(component, {
...state.props,
value: state.value,
onChange: mutators.change,
onBlur: mutators.blur,
onFocus: mutators.focus
})}
</Form.Item>
)
}}
</InternalField>
)
}
//No need to pack with connet
registerFormField('complex', ({ path }) => {
useFormEffects(({ setFieldState }) => {
FormEffectHooks.onFieldValueChange$('ccc').subscribe(({ value }) => {
if (value === '123') {
setFieldState('ddd', state => {
state.value = 'this is linkage relationship'
})
}
})
})
return (
<>
<FormItem name={FormPath.parse(path).concat('aaa')} component={Input} />
<FormItem name={FormPath.parse(path).concat('bbb')} component={Input} />
<FormItem name="ccc" component={Input} />
<FormItem name="ddd" component={Input} />
</>
)
})
In this pseudo-code, we mainly use two core APIs, mainly useFormEffects and InternalField,useFormEffects gives developers a place to write the effects logic locally, so that the effects logic can be easily reused. InternalField is the @formily/react Field component. You can take a look at the @formily/react documentation. SchemaForm also uses @formily/react internally, it can share the same Context, so we can easily use InternalField inside a custom component. Also note that when using InternalField directly, the name we registered is the root level name, if you want to reuse the path of the current custom component, you can use FormPath to resolve the path, and then concat.
API
The overall API fully inherits @formily/core and @formily/react, and only the specific APIs of @formily/react-schema-renderer are listed below.
connect
Custom component registration bridge, a high-level component function (HOC) that is used primarily to quickly access most component library components (components that implement the value/onChange API)
Signature
connect(options?: IConnectOptions): (component : React.JSXElementConstructor<any>)=>(fieldProps:ISchemaFieldComponentProps)=>JSX.Element
Usage
import { registerFormField, connect } from '@formily/react-schema-renderer'
import { Select } from 'antd'
registerFormField('select', connect()(Select))
registerFormField
Register custom component functions
Signature
registerFormField(
name:string,
component: React.JSXElementConstructor<ISchemaFieldComponentProps>
)
registerFormFields
Bulk registration of custom components
Signature
registerFormFields(
fieldsMap: {
[key : string]: React.JSXElementConstructor<ISchemaFieldComponentProps>
}
)
registerFormComponent
Register Form Style Component
Signature
registerFormComponent<Props>(
component:React.JSXElementConstructor<Props>
)
registerFormItemComponent
Register the FormItem style component
Signature
registerFormItemComponent(
component:React.JSXElementConstructor<ISchemaFieldComponentProps>
)
registerVirtualBox
Register the virtual box component, mainly used to describe the layout in JSON Schema
Signature
registerVirtualBox(
name:string,
component:React.JSXElementConstructor<ISchemaVirtualFieldComponentProps>
)
createVirtualBox
Create a virtual box component, the returned component can be used directly in SchemaMarkupForm
Signature
createVirtualBox<Props>(
name:string,
component:React.JSXElementConstructor<Props>
) : React.FC<Props>
Usage
import React from 'react'
import {
SchemaMarkupForm as SchemaForm,
SchemaMarkupField as Field,
createVirtualBox
} from '@formily/react-schema-renderer'
import { Card } from 'antd'
const FormCard = createVirtualBox('card', props => {
return <Card {...props} />
})
export default () => (
<SchemaForm>
<FormCard title="This is card">
<Field name="aaa" type="string" />
</FormCard>
</SchemaForm>
)
createControllerBox
Creating a virtual box component, the returned component can be used directly in SchemaMarkupForm, it is different from createVirtualBox mainly because the component receives the props The props received by createVirtualBox is the simplest component props createControllerBox receives
ISchemaVirtualFieldComponentProps
Signature
createControllerBox<Props>(
name:string,
component:React.JSXElementConstructor<ISchemaVirtualFieldComponentProps>
) : React.FC<Props>
Usage
import React from 'react'
import {
SchemaMarkupForm as SchemaForm,
SchemaMarkupField as Field,
createControllerBox
} from '@formily/react-schema-renderer'
import { Card } from 'antd'
const FormCard = createControllerBox('card', ({ schema, children }) => {
return <Card {...schema.getExtendsComponentProps()}>{children}</Card>
})
export default () => (
<SchemaForm>
<FormCard title="This is card">
<Field name="aaa" type="string" />
</FormCard>
</SchemaForm>
)
getRegistry
Get the registry, all components registered through the registerFormXXX API are managed in the registry
Signature
getRegistry(): ISchemaFormRegistry
cleanRegistry
Clear the registry and clear all components registered via the registerFormXXX API
Signature
cleanRegistry(): void
Classes
The whole Class inherits @formily/core completely, such as FormPath and FormLifeCyle. Only the classes specific to @formily/react-schema-renderer are listed below.
new Schema(json : ISchema)
The Schema parsing engine, given a data that satisfies the JSON Schema, we will parse it into the corresponding Schema instance, and we can quickly process some things with some tool methods. At the same time, the Schema Class provides a unified protocol difference smoothing capability to ensure Seamless and smooth upgrade when the protocol is upgraded
Attributes
| Attribute name | Description | Type |
| -------------------- | -------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- |
| title | Field title | React.ReactNode
|
| name | The parent property name of the field | string
|
| description | Field description | React.ReactNode
|
| default | Field default | any
|
| readOnly | Whether read-only and editable | boolean
|
| type | Field Type | 'string' | 'object' | 'array' | 'number' | string
|
| enum | Enumerate data | Array<string | number | { label: React.ReactNode, value: any }>
|
| const | Check if the field value is equal to the value of const | any
|
| multipleOf | Check if the field value can be divisible by the value of multipleOf | number
|
| maximum | Check the maximum value (greater than) | number
|
| exclusiveMaximum | Check the maximum value (greater than or equal to) | number
|
| minimum | Check minimum value (less than) | number
|
| exclusiveMinimum | Minimum value (less than or equal to) | number
|
| maxLength | Check maximum length | number
|
| minLength | Check minimum length | number
|
| pattern | Regular check rule | string | RegExp
|
| maxItems | Maximum number of entries | number
|
| minItems | Minimum number of entries | number
|
| uniqueItems | Whether to check for duplicates | boolean
|
| maxProperties | Maximum number of attributes | number
|
| minProperties | Minimum number of attributes | number
|
| required | Required | boolean
|
| format | Regular rule type | InternalFormats
|
| properties | Object property | {[key : string]:Schema}
|
| items | Array description | Schema | Schema[]
|
| additionalItems | Extra array element description | Schema
|
| patternProperties | Dynamically match the schema of an attribute of an object | {[key : string]:Schema}
|
| additionalProperties | Schema matching the extra attributes of the object | Schema
|
| editable | Whether the field is editable | boolean
|
| visible | Whether the data and style is visible | boolean
|
| display | Whether the style is visible | boolean
|
| x-props | Field extension attribute | { [name: string]: any }
|
| x-index | Field order | number
|
| x-rules | Field check rule | ValidatePatternRules
|
| x-component | Field UI component | string
|
| x-component-props | Field UI component properties | {}
|
| x-render | Field extension rendering function | <T = ISchemaFieldComponentProps>(props: T & { renderComponent: () => React.ReactElement}) => React.ReactElement
|
| x-effect | Field side effect trigger | (dispatch: (type: string, payload: any) => void,option?:object) => { [key: string]: any }
|
Method
get
Get the Schema child node according to the data path
Signature
get(path?: FormPathPattern): Schema
Usage
const schema = new Schema({
type: 'object',
properties: {
array: {
type: 'array',
items: {
type: 'object',
properties: {
input: {
type: 'string'
}
}
}
}
}
})
schema.get('array[0].input') //{type:"string"}
merge
Merge Schema
Signature
merge(spec:ISchema): Schema
Usage
const schema = new Schema({
type: 'object',
properties: {
array: {
type: 'array',
items: {
type: 'object',
properties: {
input: {
type: 'string'
}
}
}
}
}
})
schema.merge({
title: 'root object'
})
/**
{
type:"object",
title:"root object",
properties:{
array:{
type:'array',
items:{
type:'object',
properties:{
input:{
type:'string'
}
}
}
}
}
}
**/
getEmptyValue
Get the null value of the current Schema based on the type of Schema
Signature
getEmptyValue() : '' | [] | {} | 0
Usage
const schema = new Schema({
type: 'object',
properties: {
array: {
type: 'array',
items: {
type: 'object',
properties: {
input: {
type: 'string'
}
}
}
}
}
})
schema.get('array.0.input').getEmptyValue() // ''
schema.get('array.0').getEmptyValue() // {}
schema.get('array').getEmptyValue() // []
schema.getEmptyValue() // {}
getSelfProps
Get no nested Schema properties (does not include nested properties like properties/items)
Signature
getSelfProps() : ISchema
Usage
const schema = new Schema({
type: 'object',
properties: {
array: {
type: 'array',
items: {
type: 'object',
properties: {
input: {
type: 'string'
}
}
}
}
}
})
schema.getSelfProps() // { type:"object" }
getExtendsRules
Get the extended check rule. This method is more complicated. It will parse the attributes and x-rules attributes of all the check types of the current Schema, and finally merge them into a unified rules structure.
Signature
getExtendsRules() : ValidateArrayRules
Usage
const schema = new Schema({
type:"string",
required:true,
maxLength:10
"x-rules":{
pattern:/^\d+$/
}
})
schema.getExtendsRules() // [{required:true},{max:10},{pattern:/^\d+$/}]
getExtendsRequired
Obtaining whether it is required or not is actually reading the required attribute of the Schema. Why is it encapsulated into a method to ensure that the user is not aware of the protocol upgrade? We only need to ensure backward compatibility of the method.
Signature
getExtendsRequired(): void | boolean
Usage
const schema = new Schema({
type:"string",
required:true,
maxLength:10
"x-rules":{
pattern:/^\d+$/
}
})
schema.getExtendsRequired() // true
getExtendsEditable
Get the editable state of the Schema, consistent with the getExtendsEditable ability, also to smooth out the protocol differences
Signature
getExtendsEditable() : void | boolean
Usage
const schema1 = new Schema({
type: 'string',
editable: false
})
schema1.getExtendsEditable() // false
const schema2 = new Schema({
type: 'string',
readOnly: true
})
schema2.getExtendsEditable() // false
const schema3 = new Schema({
type: 'string',
'x-props': {
editable: false
}
})
schema3.getExtendsEditable() // false
const schema4 = new Schema({
type: 'string',
'x-component-props': {
editable: false
}
})
schema4.getExtendsEditable() // false
getExtendsVisible
Get data and style visible property
签名
getExtendsVisible(): boolean
getExtendsDisplay
Get style visible property
签名
getExtendsDisplay() : boolean
getExtendsTriggerType
Get the triggerType, which is consistent with the getExtendsTriggerType capability, and is capable of providing protocol differences.
Signature
getExtendsTriggerType() : 'onBlur' | 'onChange' | string
Usage
const schema1 = new Schema({
type: 'string',
'x-props': {
triggerType: 'onBlur'
}
})
schema1.getExtendsTriggerType() // onBlur
const schema2 = new Schema({
type: 'string',
'x-component-props': {
triggerType: 'onBlur'
}
})
schema2.getExtendsTriggerType() // onBlur
const schema3 = new Schema({
type: 'string',
'x-item-props': {
triggerType: 'onBlur'
}
})
schema3.getExtendsTriggerType() // onBlur
getExtendsProps
Get the x-props attribute
Signature
getExtendsProps() : {}
getExtendsComponent
Get the x-props attribute
Signature
getExtendsComponent() : string
getExtendsComponentProps
Get the x-component-props property, which is the component property of the x-component
Signature
getExtendsComponentProps() : {}
getExtendsRenderer
Get the x-render attribute
Signature
getExtendsRenderer() : <T = ISchemaFieldComponentProps>(
props: T & {
renderComponent: () => React.ReactElement
}
) => React.ReactElement
getExtendsEffect
Get the x-effect attribute
Signature
getExtendsEffect() : (
dispatch: (type: string, payload: any) => void,
option?: object
) => { [key: string]: any }
setProperty
Set properties for the current schema
Signature
setProperty(key: string, schema: ISchema): Schema
setProperties
Set properties for the current Schema batch
Signature
setProperties(properties: {[key : string]:ISchema}) : {[key : string]:Schema}
setArrayItems
Set the items property to the current schema
Signature
setArrayItems(schema:Ischema) : Schema
getOrderProperties
Give all properties in x-index order
Signature
getOrderProperties() : {schema:Schema,key:string}[]
mapProperties
Traverse the properties of the schema in order (x-index)
Signature
mapProperties(callback?: (schema: Schema, key: string) => any):any[]
toJSON
Output no loop dependent json data structure
Signature
toJSON() : ISchema
fromJSON
Generate a Schema object based on a json parsing
Signature
fromJSON(json : ISchema) : Schema
isObject
Determine whether the current schema is an object type
Signature
isObject() : boolean
isArray()
Determine if the current schema is an array type
Signature
isArray() : boolean
Components
The whole component is completely inherited @formily/react, only the components specific to @formily/react-schema-renderer are listed below.
<SchemaForm/>
The core JSON Schema rendering component
Attributes
interface ISchemaFormProps<
Value = any,
DefaultValue = any,
FormEffectPayload = any,
FormActions = ISchemaFormActions | ISchemaFormAsyncActions
> {
//Form value
value?: Value
//Form default value
defaultValue?: DefaultValue
//Form default value, weakly controlled
initialValues?: DefaultValue
//Form actions
actions?: FormActions
//Form effects
effects?: IFormEffect<FormEffectPayload, FormActions>
//Form instance
form?: IForm
//Form change callback
onChange?: (values: Value) => void
//Form submission callback
onSubmit?: (values: Value) => void | Promise<Value>
//Form reset callback
onReset?: () => void
//Form validation failure callback
onValidateFailed?: (valideted: IFormValidateResult) => void
//Form child node
children?: React.ReactElement
//Whether to open the dirty check
useDirty?: boolean
//Is it editable
editable?: boolean | ((name: string) => boolean)
//Whether to enable pessimistic verification, if the first verification fails, stop the remaining verification
validateFirst?: boolean
//Form Schema object
schema?: ISchema
//Instance level registration custom component
fields?: ISchemaFormRegistry['fields']
//Instance level registration virtual box component
virtualFields?: ISchemaFormRegistry['virtualFields']
//Instance level registration Form style component
formComponent?: ISchemaFormRegistry['formComponent']
//Instance level registration FormItem style component
formItemComponent?: ISchemaFormRegistry['formItemComponent']
}
<SchemaField/>
Based on a Data Path, it automatically finds and renders the internal components of the Schema node, mainly used to implement recursive rendering within the custom component.
Attributes
interface ISchemaFieldProps {
//Data path
path?: FormPathPattern
}
<SchemaMarkupForm/>
Let SchemaForm support the Form component of jsx tabbed notation, which needs to be used together with SchemaMarkupField
Attributes
interface ISchemaFormProps<
Value = any,
DefaultValue = any,
FormEffectPayload = any,
FormActions = ISchemaFormActions | ISchemaFormAsyncActions
> {
//Form value
value?: Value
//Form default value
defaultValue?: DefaultValue
//Form default value, weakly controlled
initialValues?: DefaultValue
//Form actions
actions?: FormActions
//Form effects
effects?: IFormEffect<FormEffectPayload, FormActions>
//Form instance
form?: IForm
//Form change callback
onChange?: (values: Value) => void
//Form submission callback
onSubmit?: (values: Value) => void | Promise<Value>
//Form reset callback
onReset?: () => void
//Form validation failure callback
onValidateFailed?: (valideted: IFormValidateResult) => void
//Form child node
children?: React.ReactElement
//Whether to open the dirty check
useDirty?: boolean
//Is it editable
editable?: boolean | ((name: string) => boolean)
//Whether to enable pessimistic verification, if the first verification fails, stop the remaining verification
validateFirst?: boolean
//Form Schema object
schema?: ISchema
//Instance level registration custom component
fields?: ISchemaFormRegistry['fields']
//Instance level registration virtual box component
virtualFields?: ISchemaFormRegistry['virtualFields']
//Instance level registration Form style component
formComponent?: ISchemaFormRegistry['formComponent']
//Instance level registration FormItem style component
formItemComponent?: ISchemaFormRegistry['formItemComponent']
}
Usage
import {
SchemaMarkupForm as SchemaForm,
SchemaMarkupField as Field
} from '@formily/react-schema-renderer'
export default () => {
return (
<SchemaForm>
<Field name="aa" type="string" />
</SchemaForm>
)
}
<SchemaMarkupField/>
The SchemaForm supports the Field component of jsx tabbed notation, which needs to be used together with SchemaMarkupForm
Attributes
type IMarkupSchemaFieldProps = ISchema
<InternalForm/>
The core Form is the same as the Form component in @formily/react
<InternalField/>
The core Field, like the Field component in @formily/react, is mainly used in complex custom components.
Interfaces
Inheriting @formily/react and @formily/core's Interfaces as a whole, only @formily/react-schema-renderer are listed below.
IConnectOptions
Connect parameters for registered components
interface IConnectOptions {
valueName?: string //value name
eventName?: string //get value callback name
defaultProps?: {} //component default props
getValueFromEvent?: (event?: any, value?: any) => any //get value from event callback.
getProps?: (
//props transformer
componentProps: {},
fieldProps: MergedFieldComponentProps
) => {} | void
getComponent?: (
//component transformer
Target: any,
componentProps: {},
fieldProps: MergedFieldComponentProps
) => React.JSXElementConstructor<any>
}
ISchemaFieldComponentProps
It is very important to customize the properties received by the component. As long as it involves developing custom components, you need to understand the protocol.
interface ISchemaFieldComponentProps {
//State name, FieldState
displayName?: string
//Data path
name: string
//Node path
path: string
//Whether it has been initialized
initialized: boolean
//Whether it is in the original state, the state is true only when value===intialValues
pristine: boolean
//Whether it is in a legal state, as long as the error length is greater than 0, the valid is false.
valid: boolean
//Whether it is in an illegal state, as long as the error length is greater than 0, the valid is true.
invalid: boolean
//Is it in check state?
validating: boolean
//Whether it is modified, if the value changes, the property is true and will be true for the entire lifetime of the field.
modified: boolean
//Whether it is touched
touched: boolean
//Whether it is activated, it will be triggered to true when the field triggers the onFocus event, and false when onBlur is triggered.
active: boolean
//Whether it has been accessed, when the field triggers the onBlur event, it will be triggered to true
visited: boolean
//Is it visible, note: if the status is false, the value of the field will not be submitted, and the UI will not display
visible: boolean
//Whether to show, note: if the status is false, then the value of the field will be submitted, the UI will not display, similar to the form hidden field
display: boolean
//Is it editable
editable: boolean
//Whether it is in the loading state, note: if the field is in asynchronous check, loading is true
loading: boolean
//The field has multiple parameters. For example, when the field onChange is triggered, multi-parameter data is passed to the event callback, then the values of all parameters are stored here.
values: any[]
//Field error message
errors: string[]
//Field alert message
warnings: string[]
//Field value, which is constant equal to values[0]
value: any
//Initial value
initialValue: any
//Verification rules, specific type descriptions refer to the following documents
rules: ValidatePatternRules[]
//Required or not
required: boolean
//Has it been mounted?
mounted: boolean
//Has it been unmounted?
unmounted: boolean
//Field extension property, under the SchemaForm ISchema structure
props: ISchema
//The schema object of the current field
schema: Schema
//Current operation data set of the field
mutators: IMutators
//Form instance
form: IForm
//Recursive rendering function
renderField: (
addtionKey: string | number,
reactKey?: string | number
) => React.ReactElement
}
ISchemaVirtualFieldComponentProps
The attributes received by the virtual field component, as long as it involves registering the virtual field, you need to understand the protocol.
interface ISchemaVirtualFieldComponentProps {
//State name, VirtualFieldState
displayName: string
//Field data path
name: string
//Field node path
path: string
//Whether it has been initialized
initialized: boolean
//Is it visible, note: if the status is false, the UI will not be displayed, and the data will not be submitted (because it is a VirtualField)
visible: boolean
//Whether to show, note: if the state is false, the UI will not be displayed, and the data will not be submitted (because it is a VirtualField)
display: boolean
//Has it been mounted?
mounted: boolean
//Has it been unmounted?
unmounted: boolean
//Field extension attribute
props: ISchema
//The schema object of the current field
schema: Schema
//Form instance
form: IForm
//Child element
children: React.ReactElement[]
//Recursive rendering function
renderField: (
addtionKey: string | number,
reactKey?: string | number
) => React.ReactElement
}
ISchemaFormRegistry
The component registry, whether it is a normal field, a virtual field, or a Form/FormItem will be registered here.
interface ISchemaFormRegistry {
fields: {
[key: string]: React.JSXElementConstructor<ISchemaFieldComponentProps>
}
virtualFields: {
[key: string]: React.JSXElementConstructor<
ISchemaVirtualFieldComponentProps
>
}
formItemComponent: React.JSXElementConstructor<ISchemaFieldComponentProps>
formComponent: string | React.JSXElementConstructor<any>
}
ISchema
Schema protocol object, mainly used to constrain a json structure to satisfy the Schema protocol
interface ISchema {
/** base json schema spec**/
title?: React.ReactNode
description?: React.ReactNode
default?: any
readOnly?: boolean
writeOnly?: boolean
type?: 'string' | 'object' | 'array' | 'number' | string
enum?: Array<string | number | { label: React.ReactNode; value: any }>
const?: any
multipleOf?: number
maximum?: number
exclusiveMaximum?: number
minimum?: number
exclusiveMinimum?: number
maxLength?: number
minLength?: number
pattern?: string | RegExp
maxItems?: number
minItems?: number
uniqueItems?: boolean
maxProperties?: number
minProperties?: number
required?: string[] | boolean
format?: string
/** nested json schema spec **/
properties?: {
[key: string]: ISchema
}
items?: ISchema | ISchema[]
additionalItems?: ISchema
patternProperties?: {
[key: string]: ISchema
}
additionalProperties?: ISchema
/** extend json schema specs */
visible?: boolean //Field initial visible status(Whether the data is visible)
display?: boolean //Field initial display status(Whether the style is visible)
editable?: boolean
['x-props']?: { [name: string]: any }
['x-index']?: number
['x-rules']?: ValidatePatternRules
['x-linkages']?: Array<{
target: FormPathPattern
type: string
[key: string]: any
}>
['x-mega-props']?: { [name: string]: any }
['x-component']?: string
['x-component-props']?: { [name: string]: any }
['x-render']?: <T = ISchemaFieldComponentProps>(
props: T & {
renderComponent: () => React.ReactElement
}
) => React.ReactElement
['x-effect']?: (
dispatch: (type: string, payload: any) => void,
option?: object
) => { [key: string]: any }
}
ISchemaFormActions
The core actions inherit @formily/react's IFormActions, mainly adding the getSchema API
interface ISchemaFormActions extends IFormActions {
getSchema(): Schema
getFormSchema(): Schema
}
ISchemaFormAsyncActions
Core actions inherit @formily/react's IFormAsyncActions, mainly adding the getSchema API
interface ISchemaFormAsyncActions extends IFormAsyncActions {
getSchema(): Promise<Schema>
getFormSchema(): Promise<Schema>
}
IFormValidateResult
Calibration result
interface IFormValidateResult {
errors: Array<{
path: string
messages: string[]
}>
warnings: Array<{
path: string
messages: string[]
}>
}
InternalFormats
Built-in format checksum
type InternalFormats =
| 'url'
| 'email'
| 'ipv6'
| 'ipv4'
| 'idcard'
| 'taodomain'
| 'qq'
| 'phone'
| 'money'
| 'zh'
| 'date'
| 'zip'
| string
ValidateDescription
Original check description
interface ValidateDescription {
// built-in rules,ref: string rules
format?: InternalFormats
// custom validation
validator?: CustomValidator
// required
required?: boolean
// pattern
pattern?: RegExp | string
// max length
max?: number
// maximum
maximum?: number
// exclusiveMaximum
exclusiveMaximum?: number
// exclusiveMinimum
exclusiveMinimum?: number
// minimum
minimum?: number
// min
min?: number
// length
len?: number
// whitespace
whitespace?: boolean
// enum
enum?: any[]
// error message
message?: string
[key: string]: any
}
ValidateResponse
type SyncValidateResponse =
| null
| string
| boolean
| {
type?: 'error' | 'warning'
message: string
}
type AsyncValidateResponse = Promise<SyncValidateResponse>
type ValidateResponse = SyncValidateResponse | AsyncValidateResponse
CustomValidator
Custom check function
type CustomValidator = (
value: any,
description?: ValidateDescription
) => ValidateResponse
ValidatePatternRules
Verification rule set
type ValidatePatternRules =
| InternalFormats
| CustomValidator
| ValidateDescription
| Array<InternalFormats | CustomValidator | ValidateDescription>