使用Crossfilter，如何返回特定类型的所有id值的数组

Question

Forgive me, I'm not sure I'm approaching this problem correctly.

I have some data (many thousands of elements) with a type and an ID:

const data = [
  { type: 'foo', id: 1 },
  { type: 'foo', id: 3 },
  { type: 'foo', id: 5 },
  { type: 'baz', id: 8 },
  { type: 'baz', id: 10 },
  { type: 'bar', id: 11 },
  { type: 'bar', id: 13 },
  { type: 'bar', id: 17 },
  ...
];

With crossfilter , I want to filter by a type and return an array of all their ids.

For example: all the type 'bar' should return [10, 11, 13, 17]

My attempt was to group reduce. But I didn't get very far with it:

let ndx = crossfilter(data);
let d = ndx.dimension(d => d.type);
let reduceAdd = (p, v) => p.push(v);
let reduceRemove = (p, v) => p.filter(i => i !== v);
let reduceInitial = () => ([]);

And then something like:

d.group().reduce(reduceAdd, reduceRemove, reduceInitial)

Answer 1

You should use filter method in combination with map and destructing assignment .

 const data = [ { type: 'foo', id: 1 }, { type: 'foo', id: 3 }, { type: 'foo', id: 5 }, { type: 'baz', id: 8 }, { type: 'baz', id: 10 }, { type: 'bar', id: 11 }, { type: 'bar', id: 13 }, { type: 'bar', id: 17 }, ], type = 'bar'; console.log(data.filter(elem => elem.type == type).map(({id}) => id)); 

Answer 2

What you've got looks pretty much correct with me. You just have to query your group by saving it to a variable

var grp = d.group().reduce(reduceAdd, reduceRemove, reduceInitial)

and then query it like

grp.top(Infinity)

This will return an array of objects. The key of one of the objects will be bar and the value of that object will be the array of records where type is bar .

Answer 3

Using a single forEach() is more efficient in this case instead of using filter() and then map() because you have complexity of O(n) where n is the number of objects but using filter() and then map() there will be complexity of O(n+m) where m is the number of filtered records on which you do map() :

 const data = [ { type: 'foo', id: 1 }, { type: 'foo', id: 3 }, { type: 'foo', id: 5 }, { type: 'baz', id: 8 }, { type: 'baz', id: 10 }, { type: 'bar', id: 11 }, { type: 'bar', id: 13 }, { type: 'bar', id: 17 }, ]; let type = 'bar'; var res = []; data.forEach((obj)=> { if(obj.type===type){ res.push(obj.id); } }); console.log(res); 

If there are 8 objects then you are iterating 8 times in filter and then lets say you get 4 records in filter you will then iterate 4 times to get the id value in the resultant array. Total of 12 iterations. So, in such cases I prefer to support the usage of forEach() in which there will be only 8 iterations to get the same set of array.

Answer 4

 const data = [ { type: 'foo', id: 1 }, { type: 'foo', id: 3 }, { type: 'foo', id: 5 }, { type: 'baz', id: 8 }, { type: 'baz', id: 10 }, { type: 'bar', id: 11 }, { type: 'bar', id: 13 }, { type: 'bar', id: 17 }, ]; let type = 'bar'; let result = data.reduce((acc, {type, id})=>{ if(!acc[type]) acc[type]=[]; acc[type].push(id); return acc },{}) console.log(result[type]); 

Answer 5

You can also use Array.reduce here.

 const data = [ { type: 'foo', id: 1 }, { type: 'foo', id: 3 }, { type: 'foo', id: 5 }, { type: 'baz', id: 8 }, { type: 'baz', id: 10 }, { type: 'bar', id: 11 }, { type: 'bar', id: 13 }, { type: 'bar', id: 17 } ]; const filteredArray = data.reduce((result, obj) => { if (obj.type === "bar") { result.push(obj.id) } return result; }, []); console.log(filteredArray)