Highly configurable and flexible translations loader for @ngx-translate/core. Fetch multiple translations (http only) and configure them to your needs. Each translation file has it's own namespace out of the box so the key/value pairs do not conflict with each other.
@larscom/ngx-translate-module-loader
depends on @ngx-translate/core and Angular.
npm i @larscom/ngx-translate-module-loader
Import the provideModuleTranslateLoader
function and provide the options.
import { provideHttpClient } from '@angular/common/http'
import { bootstrapApplication } from '@angular/platform-browser'
import { provideModuleTranslateLoader } from '@larscom/ngx-translate-module-loader'
import { provideTranslateService } from '@ngx-translate/core'
import { AppComponent } from './app/app.component'
const baseTranslateUrl = './assets/i18n'
bootstrapApplication(AppComponent, {
providers: [
provideTranslateService({
// use loader and provide options
loader: provideModuleTranslateLoader({
modules: [
// final url: ./assets/i18n/en.json
{ baseTranslateUrl },
// final url: ./assets/i18n/feature1/en.json
{ moduleName: 'feature1', baseTranslateUrl },
// final url: ./assets/i18n/feature2/en.json
{ moduleName: 'feature2', baseTranslateUrl }
]
})
}),
// http client is mandatory
provideHttpClient()
]
}).catch((err) => console.error(err))
By default, each translation file gets it's own namespace based on the moduleName
, what does it mean?
For example with these options:
const baseTranslateUrl = './assets/i18n'
const options: IModuleTranslationOptions = {
modules: [
// no moduleName/namespace
{ baseTranslateUrl },
// namespace: FEATURE1
{ baseTranslateUrl, moduleName: 'feature1' },
// namespace: FEATURE2
{ baseTranslateUrl, moduleName: 'feature2' }
]
}
Lets say each module in the above example resolves to the following JSON:
{
"KEY": "VALUE"
}
The final translation you are working with would be:
{
"KEY": "VALUE",
"FEATURE1": {
"KEY": "VALUE"
},
"FEATURE2": {
"KEY": "VALUE"
}
}
Even though all JSON files from those modules are the same, they don't conflict because they are not on the same level after they get merged.
The configuration is very flexible, you can even define custom templates for fetching translations.
export interface IModuleTranslationOptions {
/**
* The translation module configurations
*/
modules: IModuleTranslation[]
/**
* Custom parser after retrieving a response from the server in 'text' format.
* By using this property you can parse raw text into the required TranslationObject.
*
* This parser will be used for every 'module' unless you specificy a fileParser at 'module' level.
*/
fileParser?: {
/**
* This property will append the file extension to the url when fetching from the server.
* For example: ./assets/i18n/feature1/en.json5 when you set it as json5
*/
fileExtension: 'json5' | 'xml' | string
/**
* The parser function that can parse a string to a TranslationObject
*
* For example: for a json file you would typically use: JSON.parse()
*
* @param translation the raw translation file as text
*/
parseFn: (translation: string) => TranslationObject
}
/**
* By default, each module gets its own namespace so it doesn't conflict with other modules
*/
disableNamespace?: boolean
/**
* By default, namespaces are uppercase
*/
lowercaseNamespace?: boolean
/**
* By default, it'll perform a deepmerge when merging translation files
*/
deepMerge?: boolean
/**
* Set a version to prevent the browser from caching the translation files.
* Each translation will get a query parameter with the version number
* @example 'en.json?v=123'
*/
version?: string | number
/**
* Function that gets executed if an error occurred while retrieving a translation file
* @param error the error that occurred
* @param path the path to the location file
*/
translateError?: (error: any, path: string) => void
/**
* Custom translate merge function after retrieving all translation files
* @param translations the resolved translation files
*/
translateMerger?: (translations: TranslationObject[]) => TranslationObject
/**
* Provide custom headers at 'root' level, which means this headers gets added to every request
* unless you specify headers at 'module' level.
* @see modules
*/
headers?: HttpHeaders
}
export interface IModuleTranslation {
/**
* The module name
*
* For example: shared
* @description omit moduleName if you have a translate file at baseTranslateUrl level
* @see baseTranslateUrl
*/
moduleName?: string
/**
* The base translate URL
*
* For example: ./assets/i18n
* @description the final url will then be: ./assets/i18n/shared if the moduleName is shared
* @see moduleName
*/
baseTranslateUrl: string
/**
* By default, it uses the moduleName as namespace
* @see moduleName
*
* Use this property if you want to override the default namespace
*/
namespace?: string
/**
* Custom translation map function after retrieving a translation file
* @param translation the resolved translation file
*/
translateMap?: (translation: TranslationObject) => TranslationObject
/**
* Custom parser after retrieving a response from the server in 'text' format.
* By using this property you can parse raw text into the required TranslationObject.
*/
fileParser?: {
/**
* This property will append the file extension to the url when fetching from the server.
* For example: ./assets/i18n/feature1/en.json5 when you set it as json5
*/
fileExtension: 'json5' | 'xml' | string
/**
* The parser function that can parse a string to a TranslationObject
*
* For example: for a json file you would typically use: JSON.parse()
*
* @param translation the raw translation file as text
*/
parseFn: (translation: string) => TranslationObject
}
/**
* Custom path template for fetching translations
* @example
* '{baseTranslateUrl}/{moduleName}/{language}'
* or
* @example
* '{baseTranslateUrl}/{language}'
*
* It depends whether you have a moduleName defined
* @see moduleName
*/
pathTemplate?: string
/**
* Provide custom headers at 'module' level. These headers only apply to this module.
*/
headers?: HttpHeaders
}
By default, translations gets fetched by using the following template:
'{baseTranslateUrl}/{moduleName}/{language}'
e.g. ./assets/feature1/en.json
You can override this option if you wish to do so:
const options: IModuleTranslationOptions = {
modules: [
// resolves to: ./assets/my-path/en.json
{ baseTranslateUrl, pathTemplate: '{baseTranslateUrl}/my-path/{language}' },
// resolves to: ./assets/my-path/en/feature1.json
{ baseTranslateUrl, moduleName: 'feature1', pathTemplate: '{baseTranslateUrl}/my-path/{language}/{moduleName}' },
// resolves to: ./assets/my-path/en/feature2.json
{ baseTranslateUrl, moduleName: 'feature2', pathTemplate: '{baseTranslateUrl}/my-path/{language}/{moduleName}' }
]
}
const options: IModuleTranslationOptions = {
// global headers, applied to every request, unless you specify headers at 'module' level
headers: new HttpHeaders().set('Header-Name', 'Header value')
modules: [
{ baseTranslateUrl },
// headers only applied to this module
{ baseTranslateUrl, moduleName: 'feature1', headers: new HttpHeaders().set('Header-Name', 'Header value') },
{ baseTranslateUrl, moduleName: 'feature2' }
]
};
You can provide a custom file parser, this parser kicks in after retrieving the file from the server in text format. This allows you to use JSON5 or any other format instead of the default JSON format.
const options: IModuleTranslationOptions = {
// global fileParser, applied to every module, unless you specify fileParser at 'module' level
fileParser: {
fileExtension: 'json5',
parseFn: (text) => JSON5.parse(text)
},
modules: [
{ baseTranslateUrl },
// fileParser only applied to this module
{
baseTranslateUrl,
moduleName: 'feature1',
fileParser: {
fileExtension: 'xml',
parseFn: (text) => XML.parse(text)
}
},
{ baseTranslateUrl, moduleName: 'feature2' }
]
}