Skip to main content

Optimistic Updates

Optimistic updates enable highly responsive and fast interfaces by avoiding network wait times. An update is optimistic by assuming the network is successful. In the case of any errors, Rest Hooks will then roll back any changes in a way that deals with all possible race conditions.

Partial updates​

One common use case is for quick toggles. Here we demonstrate a publish button for an article. Note that we need to include the primary key (id in this case) in the response body to ensure the normalized cache gets updated correctly.

api/Article.ts
import { Entity, createResource } from '@rest-hooks/rest';

export class Article extends Entity {
  readonly id: string | undefined = undefined;
  readonly title: string = '';
  readonly content: string = '';
  readonly published: boolean = false;

  pk() {
    return this.id;
  }
}

const BaseArticleResource = createResource({
  path: '/articles/:id',
  schema: Article,
});
export const ArticleResource = {
  ...BaseArticleResource,
  partialUpdate: BaseArticleResource.partialUpdate.extend({
    getOptimisticResponse(snap, { id }, body) {
      return {
        // we absolutely need the id for primary key here,
        // but won't be a member of body
        id,
        ...body,
      };
    },
  }),
};
PublishButton.tsx
import { useController } from '@rest-hooks/react';
import { ArticleResource } from 'api/Article';

export default function PublishButton({ id }: { id: string }) {
  const controller = useController();

  return (
    <button
      onClick={() =>
        controller.fetch(
          ArticleResource.partialUpdate,
          { id },
          { published: true },
        )
      }
    >
      Publish
    </button>
  );
}

Optimistic create with instant updates​

Optimistic updates can also be combined with atomic mutations, enabling updates to other endpoints instantly. This is most commonly seen when creating new items while viewing a list of them.

Here we demonstrate what could be used in a list of articles with a modal to create a new article. On submission of the form it would instantly add to the list of articles the newly created article - without waiting on a network response.

api/Article.ts
import { Entity, createResource } from '@rest-hooks/rest';
import uuid from 'uuid/v4';

export class Article extends Entity {
  id: string | undefined = undefined;
  title = '';
  content = '';
  published = false;

  pk() {
    return this.id;
  }
}

const BaseArticleResource = createResource({
  path: '/articles/:id',
  schema: Article,
});
export const ArticleResource = {
  ...BaseArticleResource,
  create: BaseArticleResource.create.extend({
    getRequestInit(body) {
      if (body) {
        return this.constructor.prototype.getRequestInit.call(this, {
          id: uuid(),
          ...body,
        });
      }
      return this.constructor.prototype.getRequestInit.call(this, body);
    },
    getOptimisticResponse(snap, params, body) {
      return body;
    },
    update(newResourcePk: string) {
      return {
        [list.key({})]: (resourcePks: string[] = []) => [
          ...resourcePks,
          newResourcePk,
        ],
      };
    },
  }),
};

Since the actual id of the article is created on the server, we will need to fill in a temporary fake id here, so the primary key can be generated. This is needed to properly normalize the article to be looked up in the cache.

Once the network responds, it will have a different id, which will replace the existing data. This is often seamless, but care should be taken if the fake id is used in any renders - like to issue subsequent requests. We recommend disabling edit type features that rely on the primary key until the network fetch completes.

CreateArticle.tsx
import { useController } from '@rest-hooks/react';
import { ArticleResource } from 'api/Article';

export default function CreateArticle() {
  const ctrl = useController();
  const submitHandler = useCallback(
    data => ctrl.fetch(ArticleResource.create, data),
    [create],
  );

  return <Form onSubmit={submitHandler}>{/* rest of form */}</Form>;
}

Optimistic Deletes​

Since deletes automatically update the cache correctly upon fetch success, making your delete endpoint do this optimistically is as easy as adding the getOptimisticResponse function to your options.

We return an empty string because that's the response we expect from the server. Although by default, the server response is ignored.

api/Article.ts
import { Entity, createResource } from '@rest-hooks/rest';

export class Article extends Entity {
  readonly id: string | undefined = undefined;
  readonly title: string = '';
  readonly content: string = '';
  readonly published: boolean = false;

  pk() {
    return this.id;
  }
}

const BaseArticleResource = createResource({
  path: '/articles/:id',
  schema: Article,
});
export const ArticleResource = {
  ...BaseArticleResource,
  delete: BaseArticleResource.delete.extend({
    getOptimisticResponse(snap, params, body) {
      return params;
    },
  }),
};

Optimistic Transforms​

Sometimes user actions should result in data transformations that are dependent on the previous state of data. The simplest examples of this are toggling a boolean, or incrementing a counter; but the same principal applies to more complicated transforms. To make it more obvious we're using a simple counter here.

Fixtures
GET /api/count
{"count":0}
POST /api/count/increment
response(){return{"count":this.count=this.count+1};}
▶api/Count.ts
export class CountEntity extends Entity {

  count = 0;



  pk() {

    return `SINGLETON`;

  }

}

export const getCount = new RestEndpoint({

  path: '/api/count',

  schema: CountEntity,

  name: 'get',

});

export const increment = new RestEndpoint({

  path: '/api/count/increment',

  method: 'POST',

  body: undefined,

  name: 'increment',

  schema: CountEntity,

  getOptimisticResponse(snap) {

    const { data } = snap.getResponse(getCount);

    if (!data) throw new AbortOptimistic();

    return {

      count: data.count + 1,

    };

  },

});
▶CounterPage.tsx
import { useLoading } from '@rest-hooks/hooks';

import { getCount, increment } from './api/Count';



function CounterPage() {

  const ctrl = useController();

  const { count } = useSuspense(getCount);

  const [stateCount, setStateCount] = React.useState(0);

  const [responseCount, setResponseCount] = React.useState(0);

  const [clickHandler, loading, error] = useLoading(async () => {

    setStateCount(stateCount+1);

    const val = await ctrl.fetch(increment);

    setResponseCount(val.count);

    setStateCount(val.count);

  });

  return (

    <div>

      <p>

        Click the button multiple times quickly to trigger the race condition

      </p>

      <div>

        Rest Hooks: {count}

        <br />

        Other Libraries: {responseCount}; with optimistic: {stateCount}

        <br />

        <button onClick={clickHandler}>+</button>

        {loading ? ' ...loading' : ''}

      </div>

    </div>

  );

}

render(<CounterPage />);
Live Preview
Loading...
Store▶

Rest Hooks automatically handles all race conditions due to network timings. Rest Hooks both tracks fetch timings, pairs responses with their respective optimistic update and rollsback in case of resolution or rejection/failure.

You can see how this is problematic for other libraries even without optimistic updates; but optimistic updates make it even worse.

Example race condition​

Here's an example of the race condition. Here we request an increment twice; but the first response comes back to client after the second response.

With other libraries and no optimistic updates this would result in showing 0, then, 2, then 1.

If the other library does have optimistic updates, it should show 0, 1, 2, 2, then 1.

In both cases we end up showing an incorrect state, and along the way see weird janky state updates.

Compensating for Server timing variations​

There are three timings which can vary in an async mutation.

  1. Request timing
  2. Server timing
  3. Response timing

Rest Hooks is able to automatically handling the network timings, aka request and response timing. Typically this is sufficient, as servers tend to process requests received first before others. However, in case persist order varies from request order in the server this could cause another race condition.

This can be be solved by maintaining a total order. Because the servers and clients can potentially has different times, we will need to track time from a consistent perspective. Since we are performing optimistic updates this means we must use the client's clock. This means we will send the request timing to the server in an updatedAt header via getRequestInit(). The server should then ensure processing based on that order, and then store this updatedAt in the entity to return in any request.

Overriding our useIncoming, we can check which data is newer, and disregard old data that resolves out of order.

We use snap.fetchedAt in our getOptimisticResponse. This respresents the moment the fetch is triggered, which will be the same time the updatedAt header is computed.

Fixtures
GET /api/count
{"count":0,"updatedAt":1679342191349}
POST /api/count/increment
fetchResponse(input,init){return{"count":this.count=this.count+1,"updatedAt":JSON.parse(init.body).updatedAt};}
▶api/Count.ts
export class CountEntity extends Entity {

  count = 0;

  updatedAt = 0;



  pk() {

    return `SINGLETON`;

  }



  static useIncoming(existingMeta, incomingMeta, existing, incoming) {

    return existing.updatedAt <= incoming.updatedAt;

  }

}

export const getCount = new RestEndpoint({

  path: '/api/count',

  schema: CountEntity,

  name: 'get',

});

export const increment = new RestEndpoint({

  path: '/api/count/increment',

  method: 'POST',

  body: undefined,

  name: 'increment',

  schema: CountEntity,

  getRequestInit() {

    // this is a substitute for super.getRequestInit() since we aren't in a class context

    return RestEndpoint.prototype.getRequestInit.call(this, {

      updatedAt: Date.now(),

    });

  },

  getOptimisticResponse(snap) {

    const { data } = snap.getResponse(getCount);

    if (!data) throw new AbortOptimistic();

    return {

      count: data.count + 1,

      updatedAt: snap.fetchedAt,

    };

  },

});
▶CounterPage.tsx
import { useLoading } from '@rest-hooks/hooks';

import { getCount, increment } from './api/Count';



function CounterPage() {

  const ctrl = useController();

  const { count } = useSuspense(getCount);

  const [n, setN] = React.useState(count);

  const [clickHandler, loading, error] = useLoading(() => {

    setN(n => n + 1);

    return ctrl.fetch(increment);

  });

  return (

    <div>

      <p>

        Click the button multiple times quickly to trigger the potential race

        condition. This time our vector clock protects us.

      </p>

      <div>

        Network: {count} Should be: {n}

        <br />

        <button onClick={clickHandler}>+</button>

        {loading ? ' ...loading' : ''}

      </div>

    </div>

  );

}

render(<CounterPage />);
Live Preview
Loading...
Store▶