This is a sample website written in Typescript utilizing the Kontent.ai Delivery API to retrieve content from Kontent.ai. You can register your developer account at https://app.kontent.ai. For a brief walkthrough, check out Running the React sample app at Kontent.ai Learn.
- Install the latest version of NodeJS and npm. You can download both at https://nodejs.org/en/download/.
- Clone the sample application repository.
- Navigate to the root folder of the application in the command line.
- Type
npm install
to install required npm packages. - Type
npm start
to start a development server. - The application opens in your browser at http://localhost:3000.
At the first run of the app, you'll be presented with a configuration page. It will allow you to connect the app to your Kontent.ai project or create a new one. You'll also be able to start a trial and convert to a free plan when the trial expires.
Alternatively, you can connect your project manually as per the chapter below.
If you want to change the source Kontent.ai project, follow these steps:
- In Kontent.ai, choose Project settings from the app menu.
- Under Development, choose API keys.
- Copy your Environemnt ID.
- Open
.env.example
in the root directory. - Replace
your_environment_id
with your Environment ID and removeREACT_APP_PREVIEW_API_KEY
entry. - Save and rename the file
.env
.
When you now run the sample application, the application retrieves content from your project.
Deploy, explore and change the app directly in the browser.
If you already have a Kontent.ai account and you want to connect the sample application to a project of your own, you need to provide your Environment ID and your Preview API key to authorize requests to the Delivery Preview API. For example, you can connect the application to your modified version of the sample project.
To preview content in the sample application, follow these steps:
- In Kontent.ai, choose Project settings from the app menu.
- Under Development, choose API keys.
- Copy your Environemnt ID and Preview API key.
- Open
.env.example
in the root directory . - Replace
your_environment_id
andyour_api_key
with your Environment ID and Preview API key. - Save and rename the file
.env
.
When you now run the application, you will see all project content including the unpublished version of content items.
- Navigate to https://app.kontent.ai in your browser.
- Sign in with your credentials.
- Manage content in the content administration interface of your sample project.
You can learn more about content editing in our tutorials at Kontent.ai Learn.
You can retrieve content either through the Kontent.ai Delivery SDKs or the Kontent.ai Delivery API:
- For published content, use
https://deliver.kontent.ai/ENVIRONMENT_ID/items
. - For unpublished content, use
https://preview-deliver.kontent.ai/ENVIRONMENT_ID/items
.
For more info about the API, see the API reference.
You can find the Delivery and other SDKs at https://github.com/kontent-ai.
This application is based on the Create React App using the following template --template typescript
.
There are two types of model mapping in this application:
Content type definitions are being generated from content types via Kontent.ai model generator tool. All generated types can be found in src/Models
folder. The _project.ts
contains information about the project structure such as project languages as well as other structure information like codenames about content types.
Some models displayed in views might require an adjustment from content types. For example, the Cafe
content type contains fields for city
and street
and we would like to have a model containing an address in the format city, street
. An example of such a view model is in CafeModel.tsx
that can be found in the src/ViewModels
folder. To convert Cafe
into CafeModel
the function located in src/Utilities/CafeListing.ts
can be used.
This solution fetches data using the Delivery client. For more implementation detail to set up the client see src/Client.ts
. The data are fetched and stored in a container
component directly in its state. Then they are passed to the presentation
component. For a better understanding see the code example below. However, depending on your needs, you can use other technologies for managing application states such as:
const Component: React.FC = () => {
const [data, setData] = useState<GeneratedDTO[]>([]);
useEffect(() => {
const query = Client.items<GeneratedDTO>()
.type(projectModel.contentTypes.generatedDTO.codename)
...
query.ToPromise()
.then(data => setData(data.items));
}, []);
return (
{data.map(item => <DisplayItem dto={item}/>)}
);
...
}
Filters in Kontent.ai are implemented using taxonomies. Filtering examples can be found in src/Components/BrewerStoreContainer.tsx
or src/Components/CoffeeStoreContainer.tsx
. Firstly, the taxonomies groups that contain possible values for filters are loaded in useEffect
blocks. Selected values for filtering are stored in the filter
variable. Items to be displayed are then selected with the functional filter
function checking whether the item matches the filter.
interface FilterType {
[index: string]: string[];
processings: string[];
productStatuses: string[];
}
const Container: React.FC = () => {
const [processings, setProcessings] = useState<ITaxonomyTerms[]>([]);
const [productStatuses, setProductStatuses] = useState<ITaxonomyTerms[]>([]);
const [filter, setFilter] = useState<FilterType>({
processings: [],
productStatuses: [],
});
useEffect(() => {
Client.taxonomy('processings')
.toPromise()
.then((response) => {setProcessings(response.data.taxonomy.terms);});
}, []);
useEffect(() => {
Client.taxonomy('product_status')
.toPromise()
.then((response) => {setProductStatuses(response.data.taxonomy.terms);});
}, []);
const matches = (coffee: Coffee): boolean =>
matchesTaxonomy(coffee, filter.processings, 'processings') &&
matchesTaxonomy(coffee, filter.productStatuses, 'productStatuses');
// To see how matchesTaxonomy can work see src/Utilities/CheckboxFilter
const toggleFilter = (filterName: string, filterValue: string): void => {
setFilter((filter) => ({
...filter,
[filterName]: filter[filterName].includes(filterValue)
? filter[filterName].filter((x: string) => x !== filterValue)
: [...filter[filterName], filterValue],
}));
};.
return (
<div>
...
<CheckboxFilter
...
onChange: (event) => toggleFilter('processings', event.target.id),
/>
...
<ItemListing
items={ items[language].filter((item: ItemDTO) =>matches(coffee)) }
/>
...
</div>
);
}
In Kontent.ai each language is identified by codename, in case of this project, it is en-US
and es-ES
.
Not every text of the application must be stored in Kontent.ai. Some strings, such as button texts, navigation texts, and so on, can be stored directly in the application. For those texts React-intl is used. For every language, there is a JSON file in src/Localization
folder.
React-intl
can not parse nested JSON objects and therefore the format of files iskey:value
. To load all files fromsrc/Localization
folder there is asrc/utilities/LocalizationLoader.ts
script.
// en-US.json
{
"LatestArticles.title": "Latest articles",
"LatestArticles.noTitleValue": "(Article has no title)",
"LatestArticles.noTeaserValue": "(Article has no teaser image)",
"LatestArticles.noSummaryValue": "No summary filled"
// ...
}
The language prefix is obtained from the URL in the LocalizedApp.tsx
and then it is propagated via IntlProvider to the whole application. Content language is then adjusted by modifying Client
with languageParameter()
method to obtain items in a specific language. By default it uses language fallbacks set up in the project.
const Component: React.FC = () => {
const { locale: language } = useIntl();
useEffect(() => {
const query = Client.items<ItemDTO>()
.type(projectModel.contentTypes.itemDTO.codename);
if (language) {
query.languageParameter(language);
}
...
You might want to request items based on the URL slugs. For more information check out Kontent.ai/learn tutorial. An example in this application for this is provided in src/Pages/About.tsx
page.
The showcase is not ideal, because it is using a combination of the language prefixes and localizable routes. You should try to stick with one of the approaches. Because it is hard to define the behavior (priority) for language setting clashes like `//articles/.
To deal with content that is not available in current language, this project uses method called language fallbacks. It will fetch the content in the language which set as fallback language in the Kontent.ai project and redirect the website to the URL with prefix of the given language. However, it is possible to disable language fallbacks by adding a filter of system.language
to your query. For more information about getting localized content check this link.
var query = Client.items<AboutUs>().type(contentTypes.about_us.codename);
if (this.language) {
query
.languageParameter(this.language)
.equalsFilter('system.language', 'es-ES');
}
For the not found resources, prefixed 404 pages are used for both languages. As the content on one page should be in one language, this approach might help you to optimize SEO. If language is not set in the URL the application uses the last used language, which is set in cookies.
You can use eg. surge to deploy your app live. Check out the step-by-step guide on our blog.
We would like to express our thanks to the following people who contributed and made the project possible:
Would you like to become a hero too? Pick an issue and send us a pull request!