Utility to normalize JSON API data for redux applications
json-api-normalizer helps awesome JSON API and redux work together. Unlike normalizr json-api-normalizer supports JSON API specification, which means that you don't have to care about schemes. It also converts collections into maps, which is a lot more suitable for redux.
Demo - https://yury-dymov.github.io/json-api-react-redux-example/
Demo sources - https://github.com/yury-dymov/json-api-react-redux-example
Works great together with redux-object, which helps to fetch and denormalize data from the store.
json-api-normalizer was recently featured in SmashingMagazine: https://www.smashingmagazine.com/2017/05/json-api-normalizer-redux/
import normalize from 'json-api-normalizer';
const json = {
data: [{
"type": "post-block",
"relationships": {
"question": {
"data": {
"type": "question",
"id": "295"
}
}
},
"id": "2620",
"attributes": {
"text": "I am great!",
"id": 2620
}
}],
included: [{
"type": "question",
"id": "295",
"attributes": {
"text": "How are you?",
id: 295
}
}]
};
console.log(normalize(json));
/* Output:
{
question: {
"295": {
attributes: {
id: 295,
text: "How are you?"
}
}
},
postBlock: {
"2620": {
attributes: {
id: 2620,
text: "I am great!"
},
relationships: {
question: ["295"]
}
}
}
}
*/
While using redux, it is supposed that cache is incrementally updated during the application lifecycle. However, you might face an issue if two different requests are working with the same data objects, and after normalization, it is not clear how to distinguish, which data objects are related to which request. json-api-normalizer can handle such situations by saving the API response structure as metadata, so you can easily get only data corresponding to the certain request.
console.log(normalize(json, { endpoint: '/post-block/2620' }));
/* Output:
{
question: {
...
},
postBlock: {
...
},
meta: {
"/post-block/2620": {
data: [{
type: "postBlock",
id: "2620",
relationships: {
"question": ["295"]
}]
}
}
}
*/
By default request query options are ignored as it is supposed that data is incrementally updated. You can override this behavior by setting filterEndpoint
option value to false
.
const d1 = normalize(json, { endpoint: '/post-block/2620?page[cursor]=0' });
const d2 = normalize(json, { endpoint: '/post-block/2620?page[cursor]=20' });
console.log(Object.assign({}, d1, d2));
/* Output:
{
question: {
...
},
postBlock: {
...
},
meta: {
"/post-block/2620": {
...
}
}
}
*/
const d1 = normalize(json, { endpoint: '/post-block/2620?page[cursor]=0', filterEndpoint: false });
const d2 = normalize(json, { endpoint: '/post-block/2620?page[cursor]=20', filterEndpoint: false });
console.log(Object.assign({}, d1, d2));
/* Output:
{
question: {
...
},
postBlock: {
...
},
meta: {
"/post-block/2620: {
"?page[cursor]=0": {
...
},
"?page[cursor]=20": {
...
}
}
}
}
*/
If JSON API returns links section and you define the endpoint, then links are also stored in metadata.
const json = {
data: [{
...
}],
included: [{
...
}],
links: {
first: "http://example.com/api/v1/post-block/2620?page[cursor]=0",
next: "http://example.com/api/v1/post-block/2620?page[cursor]=20"
}
};
console.log(normalize(json, { endpoint: '/post-block/2620?page[cursor]=0'}));
/* Output:
{
question: {
...
},
postBlock: {
...
},
meta: {
"/post-block/2620": {
data: [{
...
}],
links: {
first: "http://example.com/api/v1/post-block/2620?page[cursor]=0",
next: "http://example.com/api/v1/post-block/2620?page[cursor]=20"
}
}
}
}
*/
By default all object keys and type names are camelized, however, you can disable this with camelizeKeys
option.
const json = {
data: [{
type: 'post-block',
id: '1',
attributes: {
'camel-me': 1,
id: 1
}
}]
}
console.log(normalize(json));
/* Output:
{
postBlock: {
"1": {
attributes: {
id: 1,
camelMe: 1
}
}
}
}
*/
console.log(normalize(json, { camelizeKeys: false }));
/* Output:
{
"post-block": {
"1": {
attributes: {
id: 1,
"camel-me": 1
}
}
}
}
*/
MIT (c) Yury Dymov