Skip to main content
🧙‍♂️ refine grants your wishes! Please give us a ⭐️ on GitHub to keep the magic going.
Version: 4.xx.xx

Handling Filters

refine expects an array of type CrudFilters to filter results based on some field’s values. So you can use more than one filter. Even the or operator can be used to combine multiple filters.

CrudFilters

CrudFilters is an array of objects with the following properties:

// Supported operators:
type CrudOperators =
| "eq"
| "ne"
| "lt"
| "gt"
| "lte"
| "gte"
| "in"
| "nin"
| "contains"
| "ncontains"
| "containss"
| "ncontainss"
| "between"
| "nbetween"
| "null"
| "nnull"
| "or"
| "startswith"
| "nstartswith"
| "startswiths"
| "nstartswiths"
| "endswith"
| "nendswith"
| "endswiths"
| "nendswiths"

// Supported filter types:
type LogicalFilter = {
field: string;
operator: Exclude<CrudOperators, "or">;
value: any;
};

type ConditionalFilter = {
operator: "or";
value: LogicalFilter[];
};

type CrudFilter = LogicalFilter | ConditionalFilter;
type CrudFilters = CrudFilter[];

LogicalFilters

LogicalFilter works with AND logic. For example, if you want to filter by name field and age field, you can use the following filter:

const filter = [
{
field: "name",
operator: "eq",
value: "John",
},
{
field: "age",
operator: "lt",
value: 30,
},
];

Here the query will look like: "name" = "John" AND "age" < 30

ConditionalFilters

ConditionalFilter works or / and operator and expects an array of LogicalFilter objects in the value property. For example, if you want to filter multiple OR by name field and age field, you can use the following filter:

const filter = [
{
operator: "or",
value: [
{
operator: "and",
value: [
{
field: "name",
operator: "eq",
value: "John Doe",
},
{
field: "age",
operator: "eq",
value: 30,
},
],
},
{
operator: "and",
value: [
{
field: "name",
operator: "eq",
value: "JR.Doe",
},
{
field: "age",
operator: "eq",
value: 1,
},
],
},
],
},
];

Here the query will look like: ("name" = John Doe AND "age" = 30) OR ("name" = JR.Doe AND "age" = 1)

Top level multiple conditional filters usage

If you create multiple Conditional Filters at the top level, you must add a key to it. Otherwise, you will get a warning in the console and your filters may not be combined correctly.

const filter = [
{
key: "parent",
operator: "or",
value: [
{
operator: "and",
value: [
{
field: "name",
operator: "eq",
value: "John Doe",
},
{
field: "age",
operator: "eq",
value: 30,
},
],
},
{
operator: "and",
value: [
{
field: "name",
operator: "eq",
value: "Jane Doe",
},
{
field: "age",
operator: "eq",
value: 28,
},
],
},
],
},
{
key: "children",
operator: "or",
value: [
{
operator: "and",
value: [
{
field: "name",
operator: "eq",
value: "JR John",
},
{
field: "age",
operator: "eq",
value: 1,
},
],
},
{
operator: "and",
value: [
{
field: "name",
operator: "eq",
value: "JR Jane",
},
{
field: "age",
operator: "eq",
value: 2,
},
],
},
],
},
];

Combining Filters

You can group multiple parameters in the same query using the logical filters or the conditional filters operators to filter results based on more than one criteria at the same time. This allows you to create more complex queries.

Example query: Find posts with 2 possible dates & a specific status

filter = [
{
operator: "or",
value: [
{
field: "createdAt",
operator: "eq",
value: "2022-01-01",
},
{
field: "createdAt",
operator: "eq",
value: "2022-01-02",
},
],
},
{
operator: "eq",
field: "status",
value: "published",
},
];

Here the query will look like:
"status" == "published" AND ("createdAt" == "2022-01-01" OR "createdAt" == "2022-01-02")

Handle filters in a data provider

dataProvider.ts
import { DataProvider } from "@refinedev/core";

const dataProvider = (): DataProvider => ({
getList: async ({ resource, pagination, filters, sorters }) => {
if (filters) {
filters.map((filter) => {
if (filter.operator !== "or" && filter.operator !== "and" && "field" in filter) {
// Handle your logical filters here
// console.log(typeof filter); // LogicalFilter
} else {
// Handle your conditional filters here
// console.log(typeof filter); // ConditionalFilter
}
});
}
},
});
TIP

Data providers that support or and and filtering logic are as follows: