Next-level auto mocking for Relay components. Minimal code required.
Instead of having to specify resolvers for each individual field or going with the default (i.e. <mock-value-1>
), this hook will automatically guess what type/category each field is and will use faker.js to generate fake data for you. Letting you focus on what matters most, which is building great UI/UX.
Install the library and relay-test-utils. Make sure you use the relay-test-utils version corresponding to your react-relay.
yarn add -D use-relay-mock-environment relay-test-utils
OR
npm i --save-dev use-relay-mock-environment relay-test-utils
The hook does three main things:
- Tries to guess which type/category each individual field is, using fuse.js to search through a dictionary of keywords mapping to types. Example:
fieldName: "sirname"
->type: "faker.name.lastName"
. - Using the above guessed type, it generates fake data for that field, using faker.js. Example:
guessedType: "faker.name.lastName"
->fakeData: "Doe"
. - Calls
environment.mock.resolveMostRecentOperation()
, from relay-test-utils, on a interval (300ms by default) to resolve all operations (not just the first Query/Mutation/Subscription) and create mock data from the relay-compiled GraphQL. It is slightly modified to have more contentful mock data (e.g. length of arrays that match with your limit parameters - example:first: 10
-->Array(10)
, or random length arrays).
If the hook guesses the wrong type for a field, you can override it by providing data
in the options of createRelayMockEnvironmentHook(options)
(global level) or useRelayMockEnvironment(options)
(component specific level):
const useRelayMockEnvironment = createRelayMockEnvironmentHook({
data: {
[fieldName]: {
mockType: 'faker.random.word',
},
},
});
// OR
const environment = useRelayMockEnvironment({
data: {
[fieldName]: {
mockType: 'faker.random.word',
},
},
});
You can read more about data overrides
here.
For any environment (Storybook, Cypress, dummy page, etc.), you will first need to create the useRelayMockEnvironment
hook using createRelayMockEnvironmentHook()
and passing in your global options
// useRelayMockEnvironment.(js | jsx | ts | tsx)
import { createRelayMockEnvironmentHook } from 'use-relay-mock-environment';
const useRelayMockEnvironment = createRelayMockEnvironmentHook({
// Add global options here (optional)
});
export default useRelayMockEnvironment;
// MyComponent.stories.(js | jsx | ts | tsx)
import React from 'react';
import { RelayEnvironmentProvider } from 'react-relay';
import useRelayMockEnvironment from './path/to/useRelayMockEnvironment';
import MyComponent from './MyComponentQuery';
export default {
title: 'MyComponent',
component: MyComponent,
};
export const Default = () => {
const environment = useRelayMockEnvironment({
// Add story specific options here (optional)
});
return (
<RelayEnvironmentProvider environment={environment}>
<MyComponent />
</RelayEnvironmentProvider>
);
};
When using a visual testing tool like Chromatic, you can enforce having consistent fake data by providing the seed
, instantInitialLoading
and/or forceInstantInitialLoading
options. We recommend passing these options at the global level, when invoking createRelayMockEnvironmentHook()
:
// useRelayMockEnvironment.(js | jsx | ts | tsx)
import { createRelayMockEnvironmentHook } from 'use-relay-mock-environment';
import isChromatic from 'chromatic/isChromatic';
const useRelayMockEnvironment = createRelayMockEnvironmentHook({
...(isChromatic()
? {
seed: 123, // can be anything you want (string or number)
forceInstantInitialLoading: true,
}
: {}),
// Add any other global options here (optional)
});
export default useRelayMockEnvironment;
With the above you don't have to add any additional options when invoking the useRelayMockEnvironment
hook.
You can specify the following options
in createRelayMockEnvironmentHook(options)
(global level) or useRelayMockEnvironment(options)
(component specific level):
Name | Type | Description |
---|---|---|
customResolvers? |
MockResolvers |
(optional) custom resolvers that are spread after use-relay-mock-environment's ID and String resolvers. const resolvers = { ID() {...}, String() {...}, // your custom resolvers go here } You can of course override the default ID and String resolvers by specifying your own. Read more about custom resolvers here: https://relay.dev/docs/guides/testing-relay-components/#mock-payload-generator-and-the-relay_test_operation-directive |
data? |
RelayMockData |
(optional) an object containing overrides to the types/categories of each field, where each key is the fieldName or parentTypeName (see below). First specify the parentTypeName as the key, and the value is an object containing the fieldName (s) as the key(s). Example: js const mockData = { users: { firstName: { mockType: 'faker.name.firstName' } } } |
debug? |
boolean |
Enables debug logs. Useful to get the parentType of a field to resolve. |
extendStringResolver? |
MockResolver |
(optional) a function to extend use-relay-mock-environment's String resolver. If relay-mock-default is returned, then it will continue with the default mock resolvers using FakerJS. param context is the mock resolver context (read more about it here: https://relay.dev/docs/guides/testing-relay-components/#mock-resolver-context) param generateId is a function to generate a globally unique ID. |
forceLoading? |
boolean |
(optional) Whether to force loading and don't resolve any GraphQL operation. |
generatorOptions? |
MockPayloadGeneratorOptions |
(optional) mock generator options. Please read documentation of type MockPayLoadGeneratorOptions. |
instantInitialLoading? |
boolean |
(optional) Whether to instantly load the GraphQL operation. By default there is a 300ms loading time to mimick real-world network conditions. This only applies to the initial loading. If you would like to change the loading time, set loadTime instead. |
loadTime? |
number |
(optional) Loading time in miliseconds for each GraphQL operations. Default is 300ms as to mimick real-world network conditions. |
searchTypeByName? |
string |
(optional) string input to find the full context (including the parentType) of fields that match that name. Example: Given the following query: graphql query { usersConnection { edges { node { firstName } } } } To find the context of firstName , you simply add the following to your options : json { "searchTypeByName": "firstName", // ... } Use searchTypeByPath for more specific results. |
searchTypeByPath? |
string |
(optional) string input to find the full context (including the parentType) of fields for which the path ends with that string input. Given the following query: graphql query { usersConnection { edges { node { firstName } } } } To find the context of firstName , you simply add the following to your options : json { "searchTypeByName": "usersConnection.edges.node.firstName", // ... } edges.node.firstName , node.firstName and firstName would also be valid inputs. |
seed? |
number | string |
(optional) a number or string used to seed the random generators to get consistent fake data. Useful when doing tests like Jest Snapshots or Visual Regression Tests within Chromatic. If a number is passed in, it directly runs faker.seed(n) with n being the number that you specify. If a string is passed in, it first converts the string into a hashCode number (like Java's String.hashCode()), and then runs faker.seed(n) , where n is the hashCode number. Providing seed will override property generatorOptions.randomLengthArray to false, and will set geneartorOptions.arrayLength to 3, unless specified. |
You can read more about options
here.
This package was bootstrapped using TSDX, and all major functions are exported out of /src/index.ts
/src
index.ts # EDIT THIS
utils.ts # ADD utility functions here
/test
*.test.ts # EDIT THESE
Please see the main tsdx
optimizations docs. In particular, know that you can take advantage of development-only optimizations:
// ./types/index.d.ts
declare var __DEV__: boolean;
// inside your code...
if (__DEV__) {
console.log('foo');
}