useInfiniteList

Here is a basic example of how to use the useInfiniteList hook.

import React from "react";
import { useInfiniteList } from "@refinedev/core";

const PostList = () => {
  const {
    data,
    isError,
    isLoading,
    hasNextPage,
    fetchNextPage,
    isFetchingNextPage,
  } = useInfiniteList({
    resource: "categories",
    pagination: {
      pageSize: 4,
    },
  });

  if (isLoading) {
    return <p>Loading</p>;
  }
  if (isError) {
    return <p>Something went wrong</p>;
  }

  const allPages = [].concat(...(data?.pages ?? []).map((page) => page.data));

  return (
    <div>
      <ul>
        {allPages.map(({ id, title }) => (
          <li key={id}>
            {id}.{title}
          </li>
        ))}
      </ul>
      {hasNextPage && (
        <button onClick={() => fetchNextPage()} disabled={isFetchingNextPage}>
          {isFetchingNextPage ? "Loading more..." : "Load More"}
        </button>
      )}
    </div>
  );
};

The useInfiniteList hook supports pagination properties just like useList. To handle pagination, the useInfiniteList hook passes the pagination property to the getList method from the dataProvider.

Dynamically changing the pagination properties will trigger a new request. The fetchNextPage method will increase the pagination.current property by one and trigger a new request as well.

When the getList method is called via useInfiniteList, it should ideally return the total count of rows (rowCount). The way this count is obtained depends on the data provider in use:

• REST Providers: Commonly obtain the total count from the x-total-count header.
• GraphQL Providers: Often source the count from specific data fields like pageInfo.total.
• Other Providers: Follow their own practices for obtaining the total count.

If the data provider doesn't return a specific count, the getList method may fall back to using the length of the paginated data array as the rowCount.

For more information on how this works, refer to the getList method documentation.

import { useInfiniteList } from "@refinedev/core";

const postListQueryResult = useInfiniteList({
  resource: "posts",
  pagination: { current: 3, pageSize: 8 },
});

The useInfiniteList hook supports the sorting feature, which you can enable by passing the sorters property. useInfiniteList will then pass this property to the getList method from the dataProvider.

Dynamically changing the sorters property will trigger a new request.

import { useState } from "react";
import { useInfiniteList, HttpError } from "@refinedev/core";

interface IProduct {
  id: number;
  name: string;
  material: string;
}

const ProductList: React.FC = () => {
  const [order, setOrder] = useState<"asc" | "desc">("asc");

  const {
    data,
    isLoading,
    isError,
    hasNextPage,
    fetchNextPage,
    isFetchingNextPage,
  } = useInfiniteList<IProduct, HttpError>({
    resource: "products",
    sorters: [
      {
        field: "name",
        order,
      },
    ],
  });

  if (isLoading) {
    return <p>Loading</p>;
  }
  if (isError) {
    return <p>Something went wrong</p>;
  }

  const allPages = [].concat(...(data?.pages ?? []).map((page) => page.data));

  return (
    <div>
      <button
        onClick={() => setOrder((prev) => (prev === "asc" ? "desc" : "asc"))}
      >
        toggle sort
      </button>

      <ul>
        {allPages.map((product) => (
          <li key={product.id}>
            {product.name} - ({product.material})
          </li>
        ))}
      </ul>

      {hasNextPage && (
        <button onClick={() => fetchNextPage()} disabled={isFetchingNextPage}>
          {isFetchingNextPage ? "Loading more..." : "Load More"}
        </button>
      )}
    </div>
  );
};

The useInfiniteList hook supports the filtering feature, which you can enable by passing the filters property. useInfiniteList will then pass this property to the getList method from the dataProvider.

Dynamically changing the filters property will trigger a new request.

import { useState } from "react";
import { useInfiniteList, HttpError } from "@refinedev/core";

interface IProduct {
  id: number;
  name: string;
  material: string;
}

const ProductList: React.FC = () => {
  const [value, setValue] = useState("Cotton");

  const {
    data,
    isLoading,
    isError,
    hasNextPage,
    fetchNextPage,
    isFetchingNextPage,
  } = useInfiniteList<IProduct, HttpError>({
    resource: "products",
    filters: [
      {
        field: "material",
        operator: "eq",
        value,
      },
    ],
  });

  if (isLoading) {
    return <div>Loading...</div>;
  }

  if (isError) {
    return <div>Something went wrong!</div>;
  }

  const allPages = [].concat(...(data?.pages ?? []).map((page) => page.data));

  return (
    <div>
      <span> material: </span>
      <select value={value} onChange={(e) => setValue(e.target.value)}>
        {["Cotton", "Bronze", "Plastic"].map((material) => (
          <option key={material} value={material}>
            {material}
          </option>
        ))}
      </select>

      <ul>
        {allPages.map((product) => (
          <li key={product.id}>
            {product.name} - ({product.material})
          </li>
        ))}
      </ul>

      {hasNextPage && (
        <button onClick={() => fetchNextPage()} disabled={isFetchingNextPage}>
          {isFetchingNextPage ? "Loading more..." : "Load More"}
        </button>
      )}
    </div>
  );
};

This feature is only available if you use a Live Provider.

When the useInfiniteList hook is mounted, it will call the subscribe method from the liveProvider with some parameters such as channel, resource etc. This is useful when you want to subscribe to live updates.

This parameter will be passed to the getList method from the dataProvider as a parameter. It is usually used as an API endpoint path but it all depends on how you handle the resource in the getList method.

useInfiniteList({
  resource: "categories",
});

For more information, refer to the creating a data provider tutorial →

If you have multiple resources with the same name, you can pass the identifier instead of the name of the resource. It will only be used as the main matching key for the resource, data provider methods will still work with the name of the resource defined in the <Refine/> component.

For more information, refer to the identifier of the <Refine/> component documentation →

This prop allows you to specify which dataProvider if you have more than one. Just pass it like in the example:

useInfiniteList({
  dataProviderName: "second-data-provider",
});

filters will be passed to the getList method from the dataProvider as a parameter. It is used to send filter query parameters to the API.

useInfiniteList({
  filters: [
    {
      field: "title",
      operator: "contains",
      value: "Foo",
    },
  ],
});

For more information, refer to the CrudFilters interface →

sorters will be passed to the getList method from the dataProvider as a parameter. It is used to send sort query parameters to the API.

useInfiniteList({
  sorters: [
    {
      field: "title",
      order: "asc",
    },
  ],
});

For more information, refer to the CrudSorting interface →

pagination will be passed to the getList method from the dataProvider as a parameter. It is used to send pagination query parameters to the API.

You can pass the current page number to the pagination property.

useInfiniteList({
  pagination: {
    current: 2,
  },
});

You can pass the pageSize to the pagination property.

useInfiniteList({
  pagination: {
    pageSize: 20,
  },
});

This property can be "off", "client" or "server". It is used to determine whether to use server-side pagination or not.

useInfiniteList({
  pagination: {
    mode: "off",
  },
});

queryOptions is used to pass additional options to the useQuery hook. It is useful when you want to pass additional options to the useQuery hook.

useInfiniteList({
  queryOptions: {
    retry: 3,
  },
});

For more information, refer to the useQuery documentation→

meta is a special property that can be used to pass additional information to data provider methods for the following purposes:

• Customizing the data provider methods for specific use cases.
• Generating GraphQL queries using plain JavaScript Objects (JSON).

In the following example, we pass the headers property in the meta object to the create method. With similar logic, you can pass any properties to specifically handle the data provider methods.

useInfiniteList({
  meta: {
    headers: { "x-meta-data": "true" },
  },
});

const myDataProvider = {
  //...
  getList: async ({
    resource,
    pagination,
    sorters,
    filters,
    meta,
  }) => {
    const headers = meta?.headers ?? {};
    const url = `${apiUrl}/${resource}`;
    //...
    const { data } = await httpClient.get(`${url}`, { headers });

    return {
      data,
    };
  },
  //...
};

For more information, refer to the meta section of the General Concepts documentation→

NotificationProvider is required for this prop to work.

After data is fetched successfully, useInfiniteList can call open function from NotificationProvider to show a success notification. With this prop, you can customize the success notification.

useInfiniteList({
  successNotification: (data, values, resource) => {
    return {
      message: `${data.title} Successfully fetched.`,
      description: "Success with no errors",
      type: "success",
    };
  },
});

NotificationProvider is required for this prop to work.

After data fetching is failed, useInfiniteList will call the open function from NotificationProvider to show an error notification. With this prop, you can customize the error notification.

useInfiniteList({
  errorNotification: (data, values, resource) => {
    return {
      message: `Something went wrong when getting ${data.id}`,
      description: "Error",
      type: "error",
    };
  },
});

LiveProvider is required for this prop to work.

Determines whether to update data automatically ("auto") or not ("manual") if a related live event is received. It can be used to update and show data in Realtime throughout your app.

useInfiniteList({
  liveMode: "auto",
});

LiveProvider is required for this prop to work.

The callback function is executed when new events from a subscription have arrived.

useInfiniteList({
  onLiveEvent: (event) => {
    console.log(event);
  },
});

LiveProvider is required for this prop to work.

Params to pass to liveProvider's subscribe method.

If you want loading overtime for the request, you can pass the overtimeOptions prop to the this hook. It is useful when you want to show a loading indicator when the request takes too long. interval is the time interval in milliseconds. onInterval is the function that will be called on each interval.

Return overtime object from this hook. elapsedTime is the elapsed time in milliseconds. It becomes undefined when the request is completed.

const { overtime } = useInfiniteList({
  //...
  overtimeOptions: {
    interval: 1000,
    onInterval(elapsedInterval) {
      console.log(elapsedInterval);
    },
  },
});

console.log(overtime.elapsedTime); // undefined, 1000, 2000, 3000 4000, ...

// You can use it like this:
{
  elapsedTime >= 4000 && <div>this takes a bit longer than expected</div>;
}

Use pagination, hasPagination, sorters and filters instead.

Use pagination.mode instead.

Returns an object with TanStack Query's useInfiniteQuery return values.

For more information, refer to the useInfiniteQuery documentation→

overtime object is returned from this hook. elapsedTime is the elapsed time in milliseconds. It becomes undefined when the request is completed.

const { overtime } = useInfiniteList();

console.log(overtime.elapsedTime); // undefined, 1000, 2000, 3000 4000, ...

Some APIs use the cursor-pagination method for its benefits. This method uses a cursor object to determine the next set of data. The cursor can be a number or a string and is passed to the API as a query parameter.

Preparing the data provider:

Consumes data from data provider useInfiniteList with the getList method. First of all, we need to make this method in the data provider convenient for this API. The cursor data is kept in pagination and should be set to 0 by default.

getList: async ({ resource, pagination }) => {
    const { current } = pagination;
    const { data } = await axios.get(
        `https://api.fake-rest.refine.dev/${resource}?cursor=${current || 0}`,
    );

    return {
        data: data[resource],
        total: 0,
    };
},

After this process, we successfully retrieved the first page of data. Let's fill the cursor object for the next page.

getList: async ({ resource, pagination }) => {
    const { current } = pagination;
    const { data } = await axios.get(
        `https://api.fake-rest.refine.dev/${resource}?cursor=${current || 0}`,
    );

    return {
        data: data[resource],
        total: 0,
        cursor: {
            next: data.cursor.next,
            prev: data.cursor.prev,
        },
    };
},

By default, Refine expects you to return the cursor object, but is not required. This is because some APIs don't work that way. To fix this problem you need to override the getNextPageParam method and return the next cursor.

import { useInfiniteList } from "@refinedev/core";

const {
  data,
  error,
  hasNextPage,
  isLoading,
  fetchNextPage,
  isFetchingNextPage,
} = useInfiniteList({
  resource: "posts",
  queryOptions: {
    // When you override this method, you can access the `lastPage` and `allPages`.
    getNextPageParam: (lastPage, allPages) => {
      // return the last post's id
      const { data } = lastPage;
      const lastPost = data[data.length - 1];
      return lastPost.id;
    },
  },
});