Question

我已经做到了，但是我正在追求一种“最佳实践”方式。

使用https://icanhazdadjoke API来显示一个随机笑话，该笑话每隔x秒就会更新一次。有更好的方法吗？

最终我想添加停止，启动，重置功能，并觉得这种方式可能不是最好的。

我可以使用任何中间件吗？

Redux动作

// action types
import axios from 'axios';
export const FETCH_JOKE = 'FETCH_JOKE';
export const FETCH_JOKE_SUCCESS = 'FETCH_JOKE_SUCCESS';
export const FETCH_JOKE_FAILURE = 'FETCH_JOKE_FAILURE';


function fetchJoke() {
  return {
    type: FETCH_JOKE
  };
}

function fetchJokeSuccess(data) {
  return {
    type: FETCH_JOKE_SUCCESS,
    data
  };
}

function fetchJokeFail(error) {
  return {
    type: FETCH_JOKE_FAILURE,
    error
  };
}

export function fetchJokeCall(){
  return function(dispatch){
    dispatch(fetchJoke());
    return axios.get('https://icanhazdadjoke.com', { headers: { 'Accept': 'application/json' }})
    .then(function(result){
      dispatch(fetchJokeSuccess(result.data))
    })
    .catch(error => dispatch(fetchJokeFail(error)));
  }
}

Redux减速器

import {combineReducers} from 'redux';
import {FETCH_JOKE, FETCH_JOKE_SUCCESS, FETCH_JOKE_FAILURE} from '../actions';

const defaultStateList = {
  isFetching: false,
  items:[],
  error:{}
};

const joke = (state = defaultStateList, action) => {
  switch (action.type){
  case FETCH_JOKE:
    return {...state, isFetching:true};
  case FETCH_JOKE_SUCCESS:
    return {...state, isFetching:false, items:action.data};
  case FETCH_JOKE_FAILURE:
    return {...state, isFetching:false, error:action.data};
  default:
    return state;
  }
};

const rootReducer = combineReducers({
  joke
});

export default rootReducer;

笑话组件

import React, { Component } from 'react';
import { connect } from 'react-redux';
import PropTypes from 'prop-types';
import { fetchJokeCall } from '../actions';


class Joke extends Component {
  componentDidMount() {
    this.timer = setInterval(()=>  this.props.fetchJokeCall(), 1000);
  }
  componentWillUnmount() {
    clearInterval(this.timer)
    this.timer = null;
  }
  render() {
    return (
      <div>
        {this.props.joke.joke}
      </div>
    );
  }
}

Joke.propTypes = {
  fetchJokeCall: PropTypes.func,
  joke: PropTypes.array.isRequired
};

function mapStateToProps(state) {
  return {
    joke: state.joke.items,
    isfetching: state.joke.isFetching
  };
}

export default connect(mapStateToProps, { fetchJokeCall })(Joke);

Answer 1

Redux-Sagas更好，我们也在应用程序中使用它，这就是使用Redux-Sagas进行轮询的方式

只是给您一个想法，这就是您可以做到的，您还需要了解Redux-Sagas的工作方式

操作

export const FETCH_JOKE = 'FETCH_JOKE';
export const FETCH_JOKE_SUCCESS = 'FETCH_JOKE_SUCCESS';
export const FETCH_JOKE_FAILURE = 'FETCH_JOKE_FAILURE';
export const START_POLLING = 'START_POLLING';
export const STOP_POLLING = 'STOP_POLLING';

function startPolling() {
      return {
        type: START_POLLING
      };
    }

function stopPolling() {
      return {
        type: STOP_POLLING
      };
    }

function fetchJoke() {
  return {
    type: FETCH_JOKE
  };
}

function fetchJokeSuccess(data) {
  return {
    type: FETCH_JOKE_SUCCESS,
    data
  };
}

function fetchJokeFail(error) {
  return {
    type: FETCH_JOKE_FAILURE,
    error
  };
}

减速器

import {combineReducers} from 'redux';
import {FETCH_JOKE, FETCH_JOKE_SUCCESS, FETCH_JOKE_FAILURE, START_POLLING, STOP_POLLING } from '../actions';

const defaultStateList = {
  isFetching: false,
  items:[],
  error:{},
  isPolling: false,
};

const joke = (state = defaultStateList, action) => {
  switch (action.type){
  case FETCH_JOKE:
    return {...state, isFetching:true};
  case FETCH_JOKE_SUCCESS:
    return {...state, isFetching:false, items:action.data};
  case FETCH_JOKE_FAILURE:
    return {...state, isFetching:false, error:action.data};
  case START_POLLING:
    return {...state, isPolling: true};
  case STOP_POLLING:
    return {...state, isPolling: false};
  default:
    return state;
  }
};

const rootReducer = combineReducers({
  joke
});

export default rootReducer;

Sagas

import { call, put, takeEvery, takeLatest, take, race } from 'redux-saga/effects'

import {FETCH_JOKE, FETCH_JOKE_SUCCESS, FETCH_JOKE_FAILURE, START_POLLING, STOP_POLLING } from '../actions';

import axios from 'axios';



function delay(duration) {
  const promise = new Promise(resolve => {
    setTimeout(() => resolve(true), duration)
  })
  return promise
}

function* fetchJokes(action) {
  while (true) {
    try {
      const { data } = yield call(() => axios({ url: ENDPOINT }))
      yield put({ type: FETCH_JOKE_SUCCESS, data: data })
      yield call(delay, 5000)
    } catch (e) {
      yield put({ type: FETCH_JOKE_FAILURE, message: e.message })
    }
  }
}

function* watchPollJokesSaga() {
  while (true) {
    const data = yield take(START_POLLING)
    yield race([call(fetchJokes, data), take(STOP_POLLING)])
  }
}

export default function* root() {
  yield [watchPollJokesSaga()]
}

如果您想了解更多内容，还可以使用Redux-Observable，请阅读article

Answer 2

这是一个穷人的方式。我认为这不是最好的方法，但不需要任何额外的库。

动作

// action types
import axios from 'axios';

export const START_POLLING_JOKE = 'START_POLLING_JOKE';
export const STOP_POLLING_JOKE = 'STOP_POLLING_JOKE';
export const FETCH_JOKE = 'FETCH_JOKE';
export const FETCH_JOKE_SUCCESS = 'FETCH_JOKE_SUCCESS';
export const FETCH_JOKE_FAILURE = 'FETCH_JOKE_FAILURE';

const defaultPollingInterval = 60000

function startPollingJoke(interval = defaultPollingInterval) {
  return function (dispatch) {
    const fetch = () => dispatch(fetchJoke())
    dispatch({
      type: START_POLLING_JOKE,
      interval,
      fetch,
    })
  }
}

function stopPollingJoke() {
  return {
    type: STOP_POLLING_JOKE
  }
}

function fetchJoke() {
  return {
    type: FETCH_JOKE
  };
}

function fetchJokeSuccess(data) {
  return {
    type: FETCH_JOKE_SUCCESS,
    data
  };
}

function fetchJokeFail(error) {
  return {
    type: FETCH_JOKE_FAILURE,
    error
  };
}

export function pollJokeCall(interval = defaultPollingInterval) {
  return function (dispatch) {
    dispatch(fetchJoke())
    dispatch(startPollingJoke(interval))
  }
}

export function fetchJokeCall() {
  return function(dispatch){
    dispatch(fetchJoke());
    return axios.get('https://icanhazdadjoke.com', { headers: { 'Accept': 'application/json' }})
    .then(function(result){
      dispatch(fetchJokeSuccess(result.data))
    })
    .catch(error => dispatch(fetchJokeFail(error)));
  }
}

减速器

import {combineReducers} from 'redux';
import {
  START_POLLING_JOKE,
  STOP_POLLING_JOKE,
  FETCH_JOKE, 
  FETCH_JOKE_SUCCESS, 
  FETCH_JOKE_FAILURE,
} from '../actions';

const defaultStateList = {
  isFetching: false,
  items:[],
  error:{},
  pollingId: null,
  polling: false,
};

const joke = (state = defaultStateList, action) => {
  switch (action.type){
  case START_POLLING_JOKE: 
    clearInterval(state.pollingId)
    return {
      ...state,
      polling: true,
      pollingId: setInterval(action.fetch, action.interval),
    }
  }
  case STOP_POLLING_JOKE: 
    clearInterval(state.pollingId)
    return {...state, polling: false, pollingId: null}
  case FETCH_JOKE:
    return {...state, isFetching:true};
  case FETCH_JOKE_SUCCESS:
    return {...state, isFetching:false, items:action.data};
  case FETCH_JOKE_FAILURE:
    return {...state, isFetching:false, error:action.data};
  default:
    return state;
  }
};

const rootReducer = combineReducers({
  joke
});

export default rootReducer;

组件（可能有一个错误，因为我不习惯对组件进行分类）

import React, { Component } from 'react';
import { connect } from 'react-redux';
import PropTypes from 'prop-types';
import { pollJokeCall, stopPollingJoke } from '../actions';


class Joke extends Component {
  componentDidMount() {
    this.props.pollJokeCall()
  }
  componentWillUnmount() {
    this.props.stopPollingJoke()
  }
  render() {
    return (
      <div>
        {this.props.joke.joke}
      </div>
    );
  }
}

Joke.propTypes = {
  pollJokeCall: PropTypes.func,
  stopPollingJoke: PropTypes.func,
  joke: PropTypes.array.isRequired,
};

function mapStateToProps(state) {
  return {
    joke: state.joke.items,
    isfetching: state.joke.isFetching
  };
}

export default connect(mapStateToProps, { pollJokeCall, stopPollingJoke })(Joke);

Answer 3

我制作了一个小的（5kb gzipped）助手来创建基于 redux-thunk 存储的轮询。这个想法是有一个逻辑来防止注册相同的轮询两次，在迭代之间有回调等等。

https://www.npmjs.com/package/redux-polling-thunk

Answer 4

redux-saga很棒，我一直在将它与redux一起使用。它提供了出色的api来执行诸如延迟，轮询，节流，竞赛条件，任务取消之类的事情。因此，使用redux-saga，您可以添加一个观察者，以保持池状态

function* pollSagaWorker(action) {
  while (true) {
    try {
      const { data } = yield call(() => axios({ url: ENDPOINT }));
      yield put(getDataSuccessAction(data));
      yield call(delay, 4000);
    } catch (err) {
      yield put(getDataFailureAction(err));
    }
  }
}

Answer 5

我一直在解决几乎相同的问题，除了我不担心启动和停止民意测验。由于某些原因，while循环不断冻结我的应用程序，因此我放弃了它，而是像这样设置了我的传奇。

import { all, takeLatest, call, put } from 'redux-saga/effects';
import axios from 'axios';

import { API_CALL_REQUEST, API_CALL_SUCCESS, API_CALL_FAILURE, API_CALL_FETCHED } from 
'../actions/giphy';

function apiFetch() {
  let randomWord = require('random-words');
  let API_ENDPOINT = `https://api.giphy.com/v1/gifs/search? 
                   api_key=MYKEY&q=${randomWord()}&limit=12`;
  return axios({
    method: "get",
    url: API_ENDPOINT
 });
}

export function* fetchImages() {
  try {
   const res = yield call(apiFetch)
   const images = yield res.data
   yield put({type: API_CALL_SUCCESS, images})

} catch (e) {
  yield put({type: API_CALL_FAILURE, e})
  console.log('Error fetching giphy data')
 }
}

export default function* giphySaga() {
  yield all([
    takeLatest(API_CALL_REQUEST, fetchImages),
 ]);
}

然后在我的组件中添加了它。

 componentDidMount() {
   this.interval = setInterval(() => {
   this.props.dispatch({type: 'API_CALL_REQUEST'});
   }, 5000);
  }

componentWillUnmount() {
  clearInterval(this.interval)
}

它正在工作，但是希望获得一些有关如何改进它的反馈。

每x秒反应一次Redux API轮询

5 个答案: