JavaScript：如何对嵌套数组进行分组

Question

I am trying to display data using SectionList in React Native. I have written the code below displaying what I am trying to accomplish.

I want the data to first be grouped together by date , and inside of that date, I need them grouped by location. A regular JavaScript solution will work. It's important that it has a title and data key.

My input data is in this format:

[ { game_id: 1171,
    date: '2018-11-17',
    location: 'Plaza'
   },
  { game_id: 1189,
    date: '2018-11-17',
    location: 'Field - Kickball'
   },
   { game_id: 1489,
    date: '2018-11-16',
    location: 'Field - Kickball'
   },
   { game_id: 1488,
    date: '2018-11-16',
    location: 'Field - Soccer'
   }
]

I need the obtain the following output from the data array above:

data = [{
    title: "2018-11-17",
    data: [{
            title: "Field - Kickball",
            data: [{
                game_id: 1189,
                date: '2018-11-17',
                location: 'Field - Kickball'
            }]
        },
        {
            title: "Plaza",
            data: [{
                game_id: 1171,
                date: '2018-11-17',
                location: 'Plaza'
            }]
        }
    ]
    },
    {
        title: "2018-11-16",
        data: [{
                title: "Field - Kickball",
                data: [{
                    game_id: 1489,
                    date: '2018-11-16',
                    location: 'Field - Kickball'
                }]
            },
            {
                title: "Field - Soccer",
                data: [{
                    game_id: 1488,
                    date: '2018-11-16',
                    location: 'Field - Soccer'
                }]
            }
        ]
    }
]

I have tried this already:

const games = [data here]
var groups = _(games)
.groupBy(x => x.date)
        .map(value => {
            return _.groupBy(value, 'location')
            .map(({key, value}) => ({title: key, data: value}))
        })

        .map((value, key) => {
            return ({title: value[Object.keys(value)[0]][0].date, data: value})
        })

Answer 1

There are a few ways this can be achieved, however a simple approach that does not require third party dependencies like Underscore or Lodash, can implemented with the built-in Array#reduce() method as shown below.

Please see the documentation in the following code snippet for details on how this solution works:

 const input = [ { game_id: 1171, date: '2018-11-17', location: 'Plaza' }, { game_id: 1189, date: '2018-11-17', location: 'Field - Kickball' }, { game_id: 1489, date: '2018-11-16', location: 'Field - Kickball' }, { game_id: 1488, date: '2018-11-16', location: 'Field - Soccer' } ]; /* Reduce input data to required nested sub-array structure */ const data = input.reduce((result, item) => { /* Construct item to be inserted into sub array for this item.date in resulting data object */ const resultItem = { title: item.location, data: [{ game_id: item.game_id, date: item.date, location: item.location }] }; /* Find existing item in result array with title that matches date */ const resultDateList = result.find(i => i.title === item.date); if (resultDateList) { /* If matching sublist found, add constructed item to it's data array */ resultDateList.data.push(resultItem); } else { /* If not matching sublist found, add a new one to the result for this item and pre-populate the data array with new item*/ result.push({ title: item.date, data: [resultItem] }); } return result; }, []) console.log(data) 

Hope that helps!

Answer 2

You can do something like this with ES6 and without lodash:

 let arr = [ { game_id: 1171, date: '2018-11-17', location: 'Plaza' }, { game_id: 1189, date: '2018-11-17', location: 'Field - Kickball' }, { game_id: 1489, date: '2018-11-16', location: 'Field - Kickball' }, { game_id: 1488, date: '2018-11-16', location: 'Field - Soccer' } ] let groupByfield = (data, field) => data.reduce((r, c) => { let key = c[field] r[key] = r[key] || {title: key, data: []} r[key].data = [...(r[key].data || []), c] return r }, {}) let result = Object.values(groupByfield(arr, 'date')) .map(x => ({ title: x.title, data: Object.values(groupByfield(x.data, 'location')) }) ) console.log(result) 

The idea is to create your custom groupBy function and then use it for your groupings.

We are using Array.reduce , Array.map and Object.values

Answer 3

If you want a standard solution, then you could first reduce to an object, and return the values of that object, and then group again on the output :)

 function groupBy( arr, prop ) { return Object.values( arr.reduce( ( aggregate, item ) => { const val = item[prop]; if (!aggregate[val]) { aggregate[val] = { [prop]: val, data: [] }; } aggregate[val].data.push( item ); return aggregate; }, {} ) ); } const games = [ { game_id: 1171, date: '2018-11-17', location: 'Plaza' }, { game_id: 1189, date: '2018-11-17', location: 'Field - Kickball' }, { game_id: 1489, date: '2018-11-16', location: 'Field - Kickball' }, { game_id: 1488, date: '2018-11-16', location: 'Field - Soccer' } ]; const grouped = groupBy( games, 'date' ) .map( item => ({ ...item, data: groupBy( item.data, 'location' ) }) ); console.log( grouped ); 

Note that I just use the prop that got extracted as a target property for the grouping, if you want title instead, just change the [prop]: val to 'title': val and then you can make your second grouping a litle bit easier :)

Answer 4

Generate a function using _.flow() that can group an array by a field, and convert it to the format of { title, data }. The function should also accept a transformer for the data. Now you can use it recursively to group multiple times.

 const { identity, flow, partialRight: pr, groupBy, map } = _ const groupByKey = (key, transformer = identity) => flow( pr(groupBy, key), pr(map, (data, title) => ({ title, data: transformer(data) })) ) const data = [{"game_id":1171,"date":"2018-11-17","location":"Plaza"},{"game_id":1189,"date":"2018-11-17","location":"Field - Kickball"},{"game_id":1489,"date":"2018-11-16","location":"Field - Kickball"},{"game_id":1488,"date":"2018-11-16","location":"Field - Soccer"}] const result = groupByKey('date', groupByKey('location'))(data) console.log(result) 

 <script src="https://cdnjs.cloudflare.com/ajax/libs/lodash.js/4.17.11/lodash.js"></script> 

And the same idea with lodash/fp:

 const { identity, flow, groupBy, map, get } = _ const groupByKey = (key, transformer = identity) => flow( groupBy(key), map(data => ({ title: get([0, key], data), data: transformer(data) })) ) const data = [{"game_id":1171,"date":"2018-11-17","location":"Plaza"},{"game_id":1189,"date":"2018-11-17","location":"Field - Kickball"},{"game_id":1489,"date":"2018-11-16","location":"Field - Kickball"},{"game_id":1488,"date":"2018-11-16","location":"Field - Soccer"}] const result = groupByKey('date', groupByKey('location'))(data) console.log(result) 

 <script src='https://cdn.jsdelivr.net/g/lodash@4(lodash.min.js+lodash.fp.min.js)'></script>