我使用@FurkanO answer并创建了一个可以用于任何对象类型的泛型函数，我还用TypeScript写了这个函数，我更喜欢它，因为它有自动补全功能。

实现:

1. Javascript:

export const flatListToTree = (flatList, idPath, parentIdPath, childListPath, isParent) => {
  const rootParents = [];
  const map = {};
  for (const item of flatList) {
    if (!item[childListPath]) item[childListPath] = [];
    map[item[idPath]] = item;
  }
  for (const item of flatList) {
    const parentId = item[parentIdPath];
    if (isParent(item)) {
      rootParents.push(item);
    } else {
      const parentItem = map[parentId];
      parentItem[childListPath].push(item);
    }
  }
  return rootParents;
};

2. TypeScript:我假设“T”类型有一个属性的孩子列表，你可以改变“childListPath”是一个字符串而不是“keyof T”如果你有不同的用例。

export const flatListToTree = <T>(
  flatList: T[],
  idPath: keyof T,
  parentIdPath: keyof T,
  childListPath: keyof T,
  isParent: (t: T) => boolean,
) => {
  const rootParents: T[] = [];
  const map: any = {};
  for (const item of flatList) {
    if (!(item as any)[childListPath]) (item as any)[childListPath] = [];
    map[item[idPath]] = item;
  }
  for (const item of flatList) {
    const parentId = item[parentIdPath];
    if (isParent(item)) {
      rootParents.push(item);
    } else {
      const parentItem = map[parentId];
      parentItem[childListPath].push(item);
    }
  }
  return rootParents;
};

使用方法:

  const nodes = [
    { id: 2, pid: undefined, children: [] },
    { id: 3, pid: 2 },
    { id: 4, pid: 2 },
    { id: 5, pid: 4 },
    { id: 6, pid: 5 },
    { id: 7, pid: undefined },
    { id: 8, pid: 7 },
  ];
  
  const result = flatListToTree(nodes, "id", "pid", "children", node => node.pid === undefined);

我喜欢@WilliamLeung的纯JavaScript解决方案，但有时你需要在现有数组中进行更改，以保持对对象的引用。

function listToTree(data, options) {
  options = options || {};
  var ID_KEY = options.idKey || 'id';
  var PARENT_KEY = options.parentKey || 'parent';
  var CHILDREN_KEY = options.childrenKey || 'children';

  var item, id, parentId;
  var map = {};
    for(var i = 0; i < data.length; i++ ) { // make cache
    if(data[i][ID_KEY]){
      map[data[i][ID_KEY]] = data[i];
      data[i][CHILDREN_KEY] = [];
    }
  }
  for (var i = 0; i < data.length; i++) {
    if(data[i][PARENT_KEY]) { // is a child
      if(map[data[i][PARENT_KEY]]) // for dirty data
      {
        map[data[i][PARENT_KEY]][CHILDREN_KEY].push(data[i]); // add child to parent
        data.splice( i, 1 ); // remove from root
        i--; // iterator correction
      } else {
        data[i][PARENT_KEY] = 0; // clean dirty data
      }
    }
  };
  return data;
}

Exapmle: https://jsfiddle.net/kqw1qsf0/17/

Based on @FurkanO's answer, I created another version that does not mutate the origial data (like @Dac0d3r requested). I really liked @shekhardtu's answer, but realized it had to filter through the data many times. I thought a solution could be to use FurkanO's answer by copying the data first. I tried my version in jsperf, and the results where unfortunately (very) bleak... It seems like the accepted answer is really a good one! My version is quite configurable and failsafe though, so I share it with you guys anyway; here is my contribution:

function unflat(data, options = {}) {
    const { id, parentId, childrenKey } = {
        id: "id",
        parentId: "parentId",
        childrenKey: "children",
        ...options
    };
    const copiesById = data.reduce(
        (copies, datum) => ((copies[datum[id]] = datum) && copies),
        {}
    );
    return Object.values(copiesById).reduce(
        (root, datum) => {
            if ( datum[parentId] && copiesById[datum[parentId]] ) {
                copiesById[datum[parentId]][childrenKey] = [ ...copiesById[datum[parentId]][childrenKey], datum ];
            } else {
                root = [ ...root, datum ];
            }
            return root
        }, []
    );
}

const data = [
    {
        "account": "10",
        "name": "Konto 10",
        "parentAccount": null
    },{
        "account": "1010",
        "name": "Konto 1010",
        "parentAccount": "10"
    },{
        "account": "10101",
        "name": "Konto 10101",
        "parentAccount": "1010"
    },{
        "account": "10102",
        "name": "Konto 10102",
        "parentAccount": "1010"
    },{
        "account": "10103",
        "name": "Konto 10103",
        "parentAccount": "1010"
    },{
        "account": "20",
        "name": "Konto 20",
        "parentAccount": null
    },{
        "account": "2020",
        "name": "Konto 2020",
        "parentAccount": "20"
    },{
        "account": "20201",
        "name": "Konto 20201",
        "parentAccount": "2020"
    },{
        "account": "20202",
        "name": "Konto 20202",
        "parentAccount": "2020"
    }
];

const options = {
    id: "account",
    parentId: "parentAccount",
    childrenKey: "children"
};

console.log(
    "Hierarchical tree",
    unflat(data, options)
);

通过options参数，可以配置将哪个属性用作id或父id。也可以配置children属性的名称，如果有人想要“childNodes”:[]或其他什么。

OP可以简单地使用默认选项:

input.People = unflat(input.People);

如果父对象id是假的(null, undefined或其他假的值)或父对象不存在，我们认为该对象是根节点。

一个更简单的从列表到树的函数

NPM安装列表到树精简版

listToTree(列表)

来源:

function listToTree(data, options) {
    options = options || {};
    var ID_KEY = options.idKey || 'id';
    var PARENT_KEY = options.parentKey || 'parent';
    var CHILDREN_KEY = options.childrenKey || 'children';

    var tree = [],
        childrenOf = {};
    var item, id, parentId;

    for (var i = 0, length = data.length; i < length; i++) {
        item = data[i];
        id = item[ID_KEY];
        parentId = item[PARENT_KEY] || 0;
        // every item may have children
        childrenOf[id] = childrenOf[id] || [];
        // init its children
        item[CHILDREN_KEY] = childrenOf[id];
        if (parentId != 0) {
            // init its parent's children object
            childrenOf[parentId] = childrenOf[parentId] || [];
            // push it into its parent's children object
            childrenOf[parentId].push(item);
        } else {
            tree.push(item);
        }
    };

    return tree;
}

斯菲德尔

也可以使用lodashjs(v4.x)

function buildTree(arr){
  var a=_.keyBy(arr, 'id')
  return _
   .chain(arr)
   .groupBy('parentId')
   .forEach(function(v,k){ 
     k!='0' && (a[k].children=(a[k].children||[]).concat(v));
   })
   .result('0')
   .value();
}

这是我在一个react项目中使用的

// ListToTree.js
import _filter from 'lodash/filter';
import _map from 'lodash/map';

export default (arr, parentIdKey) => _map(_filter(arr, ar => !ar[parentIdKey]), ar => ({
  ...ar,
  children: _filter(arr, { [parentIdKey]: ar.id }),
}));

用法:

// somewhere.js
import ListToTree from '../Transforms/ListToTree';

const arr = [
   {
      "id":"Bci6XhCLZKPXZMUztm1R",
      "name":"Sith"
   },
   {
      "id":"C3D71CMmASiR6FfDPlEy",
      "name":"Luke",
      "parentCategoryId":"ltatOlEkHdVPf49ACCMc"
   },
   {
      "id":"aS8Ag1BQqxkO6iWBFnsf",
      "name":"Obi Wan",
      "parentCategoryId":"ltatOlEkHdVPf49ACCMc"
   },
   {
      "id":"ltatOlEkHdVPf49ACCMc",
      "name":"Jedi"
   },
   {
      "id":"pw3CNdNhnbuxhPar6nOP",
      "name":"Palpatine",
      "parentCategoryId":"Bci6XhCLZKPXZMUztm1R"
   }
];
const response = ListToTree(arr, 'parentCategoryId');

输出:

[
   {
      "id":"Bci6XhCLZKPXZMUztm1R",
      "name":"Sith",
      "children":[
         {
            "id":"pw3CNdNhnbuxhPar6nOP",
            "name":"Palpatine",
            "parentCategoryId":"Bci6XhCLZKPXZMUztm1R"
         }
      ]
   },
   {
      "id":"ltatOlEkHdVPf49ACCMc",
      "name":"Jedi",
      "children":[
         {
            "id":"C3D71CMmASiR6FfDPlEy",
            "name":"Luke",
            "parentCategoryId":"ltatOlEkHdVPf49ACCMc"
         },
         {
            "id":"aS8Ag1BQqxkO6iWBFnsf",
            "name":"Obi Wan",
            "parentCategoryId":"ltatOlEkHdVPf49ACCMc"
         }
      ]
   }
]```