Skip to content

cvara/rxdb-hooks

Repository files navigation

rxdb-hooks

React hooks for integrating with RxDB

npm version downloads

Table of Contents

Click to expand

About

Nothing fancy, just conveniently handles common use cases such as:

  • subscribing to query observables and translating results into React state
  • cleaning up after subscriptions where necessary
  • paginating results
  • maintaining useful state information (i.e. data fetching or data exhaustion during pagination)
  • lazily creating or destroying collections

Installation

# using npm
npm install rxdb-hooks

# using yarn
yarn add rxdb-hooks

Example

Root.jsx:

import React, { useEffect } from 'react';
import { Provider } from 'rxdb-hooks';
import initialize from './initialize';

const Root = () => {
  const [db, setDb] = useState();

  useEffect(() => {
    // RxDB instantiation can be asynchronous
    initialize().then(setDb);
  }, []);

  // Until db becomes available, consumer hooks that
  // depend on it will still work, absorbing the delay
  // by setting their state to isFetching:true
  return (
    <Provider db={db}>
      <App />
    </Provider>
  );
};

Consumer.jsx:

import React from 'react';
import { useRxData } from 'rxdb-hooks';

const Consumer = () => {
  const { result: characters, isFetching } = useRxData(
    // the collection to be queried
    'characters',
    // a function returning the query to be applied
    collection =>
      collection.find({
        selector: {
          affiliation: 'jedi',
        },
      })
  );

  if (isFetching) {
    return 'loading characters...';
  }

  return (
    <ul>
      {characters.map((character, idx) => (
        <li key={idx}>{character.name}</li>
      ))}
    </ul>
  );
};

initialize.js:

const initialize = async () => {
  // create RxDB
  const db = await createRxDatabase({
    name: 'test_database',
  });

  // create a collection
  const collection = await db.addCollections({
    characters: {
      schema: {
        title: 'characters',
        version: 0,
        type: 'object',
        primaryKey: 'id',
        properties: {
          id: {
            type: 'string',
            maxLength: 100,
          },
          name: {
            type: 'string',
          },
        },
      },
    },
  });

  // maybe sync collection to a remote
  // ...

  return db;
};

Compatibility with RxDB

The core API of rxdb-hooks remains largely the same across all major versions beyond 1.x, however some parts of the internal implementation (most notably the plugin) differ based on the version of rxdb we need to target *. Please use the appropriate version of rxdb-hooks as per this table:

rxdb-hooks version targeted RxDB version
5.x 14.x
4.1.x 13.x
4.0.x 10.x, 11.x, 12.x
3.x 9.x
1.x, 2.x 8.x

* Versions 7.x of RxDB and below have not been tested and are not guaranteed to work with rxdb-hooks

Migration Guide

4.x => 5.x

  • useRxDocument has been dropped; for fetching single documents simply use useRxQuery or useRxData
  • observing lazily created collection has become an opt-in feature that, if needed, has to be explicitly enabled by using the provided plugin. For more info see Lazy instantiation of RxDatabase & RxCollections

API

Provider

The <Provider /> makes the RxDatabase instance available to nested components and is required for all subsequent hooks to work.

Props

Property Type Description
db RxDatabase the RxDatabase instance to consume data from

useRxDB

Returns the RxDatabase instance made available by the <Provider />

function useRxDB(): RxDatabase

Example

const db = useRxDB();

useRxCollection

Given a collection name returns an RxCollection instance, if found in RxDatabase.

function useRxCollection<T>(name: string): RxCollection<T> | null

Example

const collection = useRxCollection('characters');

useRxQuery

Subscribes to given RxQuery object providing query results and some helpful extra state variables.

function useRxQuery<T>(query: RxQuery, options?: UseRxQueryOptions): RxQueryResult<T>

options: UseRxQueryOptions

Option Type Description
pageSize number (optional) enables pagination & defines page limit
pagination "Traditional" | "Infinite" (optional) determines pagination mode:
Traditional: results are split into pages, starts by rendering the first page and total pageCount is returned, allowing for requesting results of any specific page.
Infinite: first page of results is rendered, allowing for gradually requesting more.
Default: "Traditional"
json boolean (optional) when true resulting documents will be converted to plain JavaScript objects; equivalent to manually calling .toJSON() on each RxDocument. Default: false

result: RxQueryResult<T>

Property Type Description
result T[] | RxDocument<T>[] the resulting array of objects or RxDocument instances, depending on json option
isFetching boolean fetching state indicator
currentPage number relevant in all pagination modes; holds number of current page
isExhausted boolean relevant in Infinite pagination; flags result list as "exhausted", meaning all documents have been already fetched
fetchMore () => void relevant in Infinite pagination; a function to be called by the consumer to request documents of the next page
resetList () => void relevant in Infinite pagination; a function to be called by the consumer to reset paginated results
pageCount number relevant in Traditional pagination; holds the total number of pages available
fetchPage (page: number) => void relevant in Traditional pagination; a function to be called by the consumer to request results of a specific page

Simple Example

const collection = useRxCollection('characters');

const query = collection.find().where('affiliation').equals('Jedi');

const { result } = useRxQuery(query);

Infinite Scroll Pagination Example

const collection = useRxCollection('characters');

const query = collection.find().where('affiliation').equals('Jedi');

const {
  result: characters,
  isFetching,
  fetchMore,
  isExhausted,
} = useRxQuery(query, {
  pageSize: 5,
  pagination: 'Infinite',
});

if (isFetching) {
  return 'Loading...';
}

return (
  <CharacterList>
    {characters.map((character, index) => (
      <Character character={character} key={index} />
    ))}
    {!isExhausted && <button onClick={fetchMore}>load more</button>}
  </CharacterList>
);

Traditional Pagination Example

const collection = useRxCollection('characters');

const query = collection.find({
  selector: {
    affiliation: 'Jedi',
  },
});

const {
  result: characters,
  isFetching,
  fetchPage,
  pageCount,
} = useRxQuery(query, {
  pageSize: 5,
  pagination: 'Traditional',
});

if (isFetching) {
  return 'Loading...';
}

// render results and leverage pageCount to render page navigation
return (
  <div>
    <CharacterList>
      {characters.map((character, index) => (
        <Character character={character} key={index} />
      ))}
    </CharacterList>
    <div>
      {Array(pageCount)
        .fill()
        .map((x, i) => (
          <button
            onClick={() => {
              fetchPage(i + 1);
            }}
          >
            page {i + 1}
          </button>
        ))}
    </div>
  </div>
);

useRxData

Convenience wrapper around useRxQuery that expects a collection name & a query constructor function

function useRxData<T>(
	collectionName: string,
	queryConstructor: ((collection: RxCollection<T>) => RxQuery<T> | undefined) | undefined,
	options?: UseRxQueryOptions
): RxQueryResult<T>

Example

const { result } = useRxData('characters', collection =>
  collection.find().where('affiliation').equals('Jedi')
);

Recipes

Query and Query Constructor memoization

By design, useRxQuery will re-subscribe to query object whenever it changes, allowing for query criteria to be modified during component updates. For this reason, to avoid unnecessary re-subscriptions, query should be memoized (i.e. via react's useMemo):

const { affiliation } = props;
const collection = useRxCollection('characters');

const query = useMemo(
  () =>
    collection.find({
      selector: {
        affiliation,
      },
    }),
  [collection, affiliation]
);

const { result } = useRxQuery(query);

Same goes for useRxData and the queryConstructor function:

const { affiliation } = props;

const queryConstructor = useCallback(
  collection =>
    collection.find({
      selector: {
        affiliation,
      },
    }),
  [affiliation]
);

const { result } = useRxData('characters', queryConstructor);

Lazy instantiation of RxDatabase & RxCollections

All rxdb-hooks give you the ability to lazily instantiate the database and the collections within it. Initial delay until the above become available is absorbed by indicating the state as fetching (isFetching:true).

Since v5.0.0 of rxdb-hooks, observing newly created collections has become an opt-in feature that, if needed, has to be enabled via the provided observeNewCollections plugin:

import { addRxPlugin } from 'rxdb';
import { observeNewCollections } from 'rxdb-hooks';

addRxPlugin(observeNewCollections);

Adding the plugin makes it possible for all rxdb-hooks to pick up data from collections that are lazily added after the inital db initialization.

Also note that lazily instantiating the rxdb instance itself is supported out-of-the-box, the plugin only affects lazy collection creation.

Mutations

Performing mutations on data is possible through the APIs provided by RxDocument and RxCollection:

Example

const collection = useRxCollection('characters');

collection.upsert({
  name: 'Luke Skywalker',
  affiliation: 'Jedi',
});

LICENSE

MIT