javascript - 对 JSON 数据进行排序并获取前 n 条记录

Question

我有大约数百万条记录的 json 数据。我必须做一个简单但棘手的功能。

What I have to do?

我必须从该 json 中收集出现次数最多的前 10 个项目及其计数。出现最多的意思是 json 中项目的最多数量。我还不确定，我将如何使用 count，我的意思是我可以将它添加到与属性相同的 json 对象中。

Here is what I did so far.

//my origional json, it's too big but adding some portion of it.
var jsonData = [    
  {
    "id": "5",
    "name": "#5"
  },
  {
    "id": "1",
    "name": "#1"
  },
  {
    "id": "2",
    "name": "#2"
  },
  {
    "id": "8",
    "name": "#8"
  },
  {
    "id": "1",
    "name": "#1"
  },
  {
    "id": "10",
    "name": "#10" 
  },  
  {
    "id": "2",
    "name": "#2"
  }];    

var top10Data = [];

//geting top 10 items
function getTop10Data() {
    var i = 0;
    while (i <= 20) {
        top10Data.push(getTop1Data());
        i++;
    }

    return true;
}

//getting top 1 data that has max count in json
function getTop1Data() {
    var store = jsonData, distribution = {}, max = 0, result = [];

    store.forEach(function (a) {
        distribution[a] = (distribution[a] || 0) + 1;
        if (distribution[a] > max) {
            max = distribution[a];
            result = [a];
            return;
        }
        if (distribution[a] === max) {
            result.push(a);
        }
    });

    //remove this item with it's all occurences, and push it to top10Data
    removeData(result); 

    return result;
}

//remove items from origional json. but this is not working properly as it removes only one item from top
function removeData(result) {
    var length = jsonData.length;
    for (var i = 0; i < length; i++) {
        if (jsonData[i].toppings === result[0].toppings) {
            jsonData.splice(jsonData[i], 1);            
        }
    }
}

My question.

我认为我的做法不正确，是否有更好的方法来处理这种情况。如果我的方法没问题，我在当前代码中缺少什么。

任何帮助将不胜感激。

Answer 1

您可以将数据减少到一个对象中，该对象保存每个项目的计数，并按字符串化项目进行索引。然后，如果没有相当数量的唯一对象，您可以按条目的出现次数对条目进行排序，然后对前 10 个进行切片。

var jsonData = [    
  {
    "id": "5",
    "name": "#5"
  },
  {
    "id": "1",
    "name": "#1"
  },
  {
    "id": "2",
    "name": "#2"
  },
  {
    "id": "8",
    "name": "#8"
  },
  {
    "id": "1",
    "name": "#1"
  },
  {
    "id": "10",
    "name": "#10" 
  },  
  {
    "id": "2",
    "name": "#2"
  }];
 const counts = jsonData.reduce((a, obj) => {
  const string = JSON.stringify(obj);
  a[string] = (a[string] || 0) + 1
  return a;
}, {});
const result = Object.entries(counts)
  .sort((a, b) => b[1] - a[1])
  .slice(0, 10)
  .map(([string, count]) => ({ count, obj: JSON.parse(string) }));
console.log(result);

要将计数添加到原始数据，请在构造 counts 对象后迭代数据:

var jsonData = [    
  {
    "id": "5",
    "name": "#5"
  },
  {
    "id": "1",
    "name": "#1"
  },
  {
    "id": "2",
    "name": "#2"
  },
  {
    "id": "8",
    "name": "#8"
  },
  {
    "id": "1",
    "name": "#1"
  },
  {
    "id": "10",
    "name": "#10" 
  },  
  {
    "id": "2",
    "name": "#2"
  }];
 const counts = jsonData.reduce((a, obj) => {
  const string = JSON.stringify(obj);
  a[string] = (a[string] || 0) + 1
  return a;
}, {});
jsonData.forEach((item) => {
  item.count = counts[JSON.stringify(item)];
});
console.log(jsonData);