markbello-iguazu-rest
v1.9.1
Published
A Redux REST caching library that follows Iguazu patterns
Downloads
3
Maintainers
Readme
Iguazu REST is a plugin for the Iguazu ecosystem that allows for pre-built async calls for REST with smart caching in Redux. If your API uses RESTful patterns, this library will save you time to perform CRUD actions. If your API does not follow REST patterns, check out Iguazu RPC.
👩💻 Hiring 👨💻
Want to get paid for your contributions to iguazu-rest
?
Send your resume to [email protected]
📖 Table of Contents
✨ Features
- Plugs into Iguazu
- Easy dispatchable actions for create, read, update, destroy to call a REST API
- Caching requests based on the RESTful action
- Seamless integration in Redux
🤹 Usage
Installation
npm install --save iguazu-rest
Config
Iguazu REST uses a config object that allows you to register resources and provide default and override behavior.
import { createStore, combineReducers, applyMiddleware } from 'redux';
import { configureIguazuREST, resourcesReducer } from 'iguazu-rest';
import thunk from 'redux-thunk';
configureIguazuREST({
// the resources you want cached
resources: {
// this key will be used in actions
users: {
// returns the url and opts that should be passed to fetch
fetch: () => ({
// the resource id should always use ':id', nested ids should be more specific
url: `${process.env.HOST_URL}/users/:id`,
// opts that will be sent on every request for this resource
opts: {
credentials: 'include',
},
}),
// optionally override the resources id key, defaults to 'id'
// this is only used to extract the id after a create
// or to get the ids of the resources in a collection
idKey: 'userId',
// optionally massage the data to be more RESTful,
// collections need to be lists, resources need to be objects
transformData: (data, { id, actionType, state }) => massageDataToBeRESTful(data),
},
},
// opts that will be sent along with every resource request
defaultOpts: {
headers: {
Accept: 'application/json',
'Content-Type': 'application/json',
},
},
// extend fetch with some added functionality
baseFetch: fetchWith6sTimeout,
// override state location, defaults to state.resources
getToState: (state) => state.data.resources,
});
const store = createStore(
combineReducers({
resources: resourcesReducer,
}),
applyMiddleware(thunk)
);
Advanced Config
You may also supply a custom fetch
client to iguazu-rest using Redux Thunk.
This will override any baseFetch
configuration in Iguazu REST with the
thunk
supplied fetch client. (See Thunk withExtraArgument
docs)
This approach allows for the client and the server to specify different fetch
implementations. For example, the server needs to support cookies inside a
server-side fetch versus the client-side which works with cookies by default.
Thunk-provided fetchClient
overrides baseFetch
as the concerns of the
environment are more important such as providing cookie support on the server or
enforcing request timeouts.
To retain the ability to apply additional fetch functionality with a thunk-provided
fetchClient
you may also extend the behavior by providing a composeFetch
function in the configuration that returns a composed fetch function.
import { createStore, combineReducers, applyMiddleware } from 'redux';
import { configureIguazuREST, resourcesReducer } from 'iguazu-rest';
import thunk from 'redux-thunk';
configureIguazuREST({
// Assume configuration from above...
// Overriden by thunk.withExtraArgument below
baseFetch: fetchWith6sTimeout,
// Restored functionality with composition
composeFetch: (customFetch) => composeFetchWith6sTimeout(customFetch),
});
/* Contrived custom fetch client */
const customFetchClient = (...args) => fetch(...args);
const store = createStore(
combineReducers({
resources: resourcesReducer,
}),
applyMiddleware(thunk.withExtraArgument({
fetchClient: customFetchClient,
}))
);
Resource Fetch Function
In an ideal world, your REST API follows all the right patterns and best practices. If that is the case your fetch function can return a simple url and iguazu-rest will fill in the parameters based on the id you pass to the action. Not all of us live in an ideal world though and sometimes your boss tells you that you need to support some API endpoints that cannot be made RESTful because it's too much effort. Maybe you need to put the resource ID in the header instead of the URL or you need to make a POST to load a resource. In that case you can use some of the parameters passed to the fetch function to get creative.
Arguments
- [
id
] (String, Object, or Undefined): The id that was passed into the action. May be a string if only one id needs to be specified, an object if multiple ids need to be specified (nested urls), or undefined if no ids need to be specified (simple collection). Useful if your unique identifiers are not confined to the url. If you are using an object, the resource id key should always be named 'id' and the nested id keys should be more specific. - [
actionType
] (String): The type of CRUD action. Will be one ofLOAD
,LOAD_COLLECTION
,CREATE
,UPDATE
,DESTROY
. - [
state
] (Object): The redux state. Useful if you need to use some config held in state. Keep in mind that you should not use this to grab any unique identifiers as the id argument is what is used to cache.
Collections
To use the functions that operate on collections, your API and config needs to meet some basic requirements:
- The URL specified on the
fetch
config should return the full collection when the/:id
is omitted. - URLs can have optional path parameters by including
?
, such as${process.env.HOST_URL}/users/:userId?/books
. - The API must return an array of resource objects. If it returns an object (common in paged APIs), you can use the
transformData
config function to return the array field of the object instead. - Each object in the array must have a unique ID field, normally named
id
. If it is not namedid
, then the field name must be set in theidKey
prop on the resource config. Iguazu will throw an error on successful network responses if it cannot find a unique ID for each item in the array.
Reducer
import { resourcesReducer } from 'iguazu-rest';
import { combineReducers, createStore } from 'redux';
const reducer = combineReducers({
resources: resourcesReducer,
// other reducers
});
const store = createStore(reducer);
Actions
All iguazu-rest actions accept the same type of object as their single argument. The object can have the following properties:
- [
resource
] (String): The name of the resource type, must match the key in resources config. - [
id
] (String, Object, or Undefined): The id or ids to be populated in the REST call. May be a string if only one id needs to be specified, an object if multiple ids need to be specified (nested urls), or undefined if no ids need to be specified (simple collection). If you are using an object, the resource id key should always be named 'id' and the nested id keys should be more specific. - [
opts
] (Object or Undefined): fetch opts to be sent along with the REST call - [
forceFetch
] (Boolean or Undefined): forces the REST call to be made regardless of whether the resource or collection is loaded or not. Only applies to queryResource, queryCollection, loadCollection, and queryCollection.
🎛️ API
Iguazu Actions
queryResource({ resource, id, opts, forceFetch })
queryCollection({ resource, id, opts, forceFetch })
These actions return an object following the iguazu pattern { data, status, promise}
.
Example:
// Author.jsx
import { connectAsync } from 'iguazu';
import { queryResource } from 'iguazu-rest';
import BookList from './BookList';
const Author = ({ id, author: { name, bio } }) => (
<div>
<div>{name}</div>
<div>{bio}</div>
<BookList authorId={id} />
</div>
);
function loadDataAsProps({ store: { dispatch }, ownProps: { id } }) {
return {
author: () => dispatch(queryResource({ resource: 'author', id })),
};
}
export default connectAsync({ loadDataAsProps })(Author);
// BookList.jsx
import { connectAsync } from 'iguazu';
import { queryCollection } from 'iguazu-rest';
const BookList = ({ books }) => (
<div>
{books.map((book) => <Book key={book.id} book={book} />)}
</div>
);
function loadDataAsProps({ store: { dispatch }, ownProps: { authorId } }) {
return {
books: () => dispatch(queryCollection({ resource: 'book', id: { authorId } })),
};
}
export default connectAsync({ loadDataAsProps })(BookList);
CRUD Actions
loadResource({ resource, id, opts, forceFetch })
loadCollection({ resource, id, opts, forceFetch })
createResource({ resource, id, opts })
updateResource({ resource, id, opts })
destroyResource({ resource, id, opts })
patchResource({ resource, id, opts })
These actions return a promise that resolves with the fetched data on successful requests and reject with an error on unsuccessful requests. The error also contains the status and the body if you need to inspect those.
Selectors
getResource({ resource, id })(state)
getCollection({ resource, id, opts })(state)
Cleaning Actions
clearResource({ resource, id, opts })
clearCollection({ resource, id, opts })
These actions allow you to remove the associated data to a resource or collection from the state tree without performing any operation on the remote resource itself.
- resources associated to a collection will only be removed if no other collection is using that same resource
- individual resources will only be removed if they are not associated to any collection
Query opts
iguazu-rest allows you to specify your query parameters as part of the opts passed to fetch. If they are used to filter a collection, make sure they are passed in this way instead of adding them directly to the url because it is necessary for proper caching. All of the resources get normalized, so iguazu-rest needs to a way to cache which resources came back for a set of query parameters.
// Articles.jsx
import { connectAsync } from 'iguazu';
import { queryCollection } from 'iguazu-rest';
const Articles = ({ articles }) => (
<div>
{articles.map((article) => <Article key={article.id} article={article} />)}
</div>
);
function loadDataAsProps({ store: { dispatch } }) {
return {
articles: () => dispatch(queryCollection({
resource: 'articles',
opts: {
query: { limit: 10, sortBy: 'title' },
},
})),
};
}
export default connectAsync({ loadDataAsProps })(Articles);
🏆 Contributing
We welcome Your interest in the American Express Open Source Community on Github. Any Contributor to any Open Source Project managed by the American Express Open Source Community must accept and sign an Agreement indicating agreement to the terms below. Except for the rights granted in this Agreement to American Express and to recipients of software distributed by American Express, You reserve all right, title, and interest, if any, in and to Your Contributions. Please fill out the Agreement.
Please feel free to open pull requests and see CONTRIBUTING.md to learn how to get started contributing.
🗝️ License
Any contributions made under this project will be governed by the Apache License 2.0.
🗣️ Code of Conduct
This project adheres to the American Express Community Guidelines. By participating, you are expected to honor these guidelines.