
Security News
AI Agent Lands PRs in Major OSS Projects, Targets Maintainers via Cold Outreach
An AI agent is merging PRs into major OSS projects and cold-emailing maintainers to drum up more work.
@betafcc/red
Advanced tools

Type-safe, composable, boilerplateless reducers

npm install --save @betafcc/red
import React, { useReducer } from 'react'
import { render } from 'react-dom'
import { red } from '@betafcc/red'
const inputApp = red
.withState({ input: '' })
.handle({ setInput: (state, input: string) => ({ input }) })
const todoApp = inputApp
.withState({ todos: [] as Array<{ msg: string; done: boolean }> })
.handle({
add: (state, msg: string) => ({ input: '', todos: [...s.todos, { msg, done: false }] }),
complete: (state, id: number) => ({ todos: state.todos.map((e, i) => i !== id ? e : { ...e, done: true }) }),
})
export const TodoApp = () => {
const [{ input, todos }, { setInput, add, complete }] = todoApp.useHook(useReducer)
return <>
<input value={input} onChange={e => setInput(e.target.value)} />
<button onClick={_ => add(input)}>add</button>
{todos.map((e, id) => <li key={id} style={e.done ? { textDecoration: 'line-through' } : {}}>
{e.msg}<button onClick={_ => complete(id)}>done</button>
</li>)}
</>
}
render(<TodoApp />, document.getElementById('root'))
If every action has this shape:
type ActionType<K extends string, P extends Array<unknown>> = {
type: K
payload: P
}
We can automatically provide action creators and strong-typed reducer from simple handlers definitions:
const app = red
.withState({
input: '',
todo: [] as Array<{msg: string, id: number, done: boolean}>
})
.handle({
setInput(state, input: string) {
return { input }
},
addTodo(state, msg: string, id: number) {
return { todo: [...state.todo, {msg, id, done: false}] }
},
completeTodo(state, id: number) {
return { todo: todo.map(t => t.id !== id ? t : {...t, done: true}) }
}
})
const {
intial, // the initial state
reducer, // the reducer made by combining the handlers
actions, // the action creators
} = red
The arguments in the handlers define the payload, and their keys define the 'type', the revealed signature is:
import { StateOf, ActionOf } from '@betafcc/red'
type State = StateOf<typeof red>
// { input: string, todo: Array<{msg: string, id: number, done: boolean}> }
type Action = ActionOf<typeof red>
// { type: 'setInput', payload: [string] } | { type: 'addTodo', payload: [string, number] } | { type: 'completeTodo', payload: [number] }
And you also have action creators that matches the signature:
const { addTodo, completeTodo } = app.actions
addTodo('buy milk', 0) // { type: 'addTodo', payload: ['buy milk', 0] }
completeTodo(0) // { type: 'completeTodo', payload: [number] }
If your prefer to define the actions yourself, you can use ActionType helper and withActions method:
import { ActionType } from '@betafcc/red'
type Action =
| ActionType<'setInput', [string]>
| ActionType<'addTodo', [string, number]>
| ActionType<'completeTodo', [number]>
const app = red
.withState({ input: '', todo: [] as Array<{msg: string, id: number, done: boolean}>})
.withActions<Action>({ // annotate here
setInput(state, input) { // so arguments will have infered type, no need to annotate here
return {input}
},
addTodo(state, msg, id) {
return {todo: [...state.todo, {msg, id, done: false}]}
},
completeTodo(state, id) {
return {todo: todo.map(t => t.id !== id ? t : {...t, done: true})}
}
})
The simplest way to use is to extract the generated reducer, initial and the actions creators:
const {reducer, initial, actions} = app
const App = () => {
const [state, dispatch] = React.useReducer(reducer, initial)
return <>
{state.count}
<button onClick={_ => dispatch(actions.increment())}>+</button>
</>
}
But you can use red.useHook for some extra magic:
const App = () => {
// the action creators will become dispatchers
const [state, {increment}] = app.useHook(React.useReducer)
return <>
{state.count}
<button onClick={_ => increment()}>+</button>
</>
}
You can use red.merge to combine apps together:
const inputApp = red
.withState({input: ''})
.handle({setInput: (s, value: string) => ({input: value})})
const todoApp = red
.withState({todos: [] as Array<{id: number, done: boolean, msg: string}>})
.handle({add: (s, msg: string, id: number) => ({todos: [...s.todos, {id, msg, done: false}]}) })
const app = red.merge(inputApp).merge(todoApp)
// same as
const app = inputApp.merge(todoApp)
// same as
const app = red
.withState({input: ''})
.handle({setInput: (s, value: string) => ({input: value})})
.withState({todos: [] as Array<{id: number, done: boolean, msg: string}>})
.handle({add: (s, msg: string, id: number) => ({todos: [...s.todos, {id, msg, done: false}]}) })
Or you can combine them by namespacing with red.combine, similar to redux's combineReducers:
const inputApp = red.withState({input: ''}).handle({setInput: (s, input: string) => ({input})})
const counterApp = red.withState({count: 0}).handle({increment: s => ({count: s.count + 1})})
const app = red.combine({
ui: inputApp,
counter: counterApp
})
// equivalent to:
const app = red.withState({
ui: {input: ''},
counter: {count: 0}
}).handle({
// note that you have to manually namespace the state
setInput: (s, input: string) => ({...s, ui: {input}}),
increment: s => ({...s, counter: {count: s.counter.count + 1}})
})
FAQs
Type safe reducers without boilerplate
We found that @betafcc/red demonstrated a not healthy version release cadence and project activity because the last version was released a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Security News
An AI agent is merging PRs into major OSS projects and cold-emailing maintainers to drum up more work.

Research
/Security News
Chrome extension CL Suite by @CLMasters neutralizes 2FA for Facebook and Meta Business accounts while exfiltrating Business Manager contact and analytics data.

Security News
After Matplotlib rejected an AI-written PR, the agent fired back with a blog post, igniting debate over AI contributions and maintainer burden.