@predictive-text-studio/models-types
v14.0.0-alpha.0
Published
Type definitions in used in the modeling (lexical model/predictive text) component of Keyman.
Downloads
9
Maintainers
Readme
@keymanapp/models-types
Declares TypeScript types and interfaces required to define Lexical Models, Word breaking functions, and all things in between.
Install
(see below if you are working in the keymanapp/keyman repo!)
npm install --save-dev @keymanapp/models-types
Usage
Say you are creating a custom lexical model. Then import this module in your lexical model file!
// my-nifty-model.ts
import "@keymanapp/models-types";
export class MyNiftyClass implements LexicalModel {
configure(capabilities: Capabilities): Configuration {
// TODO: implement me!
throw new Error("Method not implemented.");
}
predict(transform: Transform, context: Context): ProbabilityMass<Suggestion>[] {
// TODO: implement me!
throw new Error("Method not implemented.");
}
wordbreak(context: Context): string {
// TODO: implement me!
throw new Error("Method not implemented.");
}
punctuation?: LexicalModelPunctuation;
}
If your are creating a custom word breaker, you would do the same:
import "@keymanapp/models-types";
export let myShinyWordBreaker: WordBreakingFunction;
myShinyWordBreaker = function (phrase: string): Span[] {
// TODO: implement me!
throw new Error("Function not implemented")
}
Look at index.d.ts for documentation on each and every type!
Usage within keymanapp/keyman repo
To use it in other subprojects within keymanapp/keyman, ensure that this package is linked using Lerna. From the root of the repository, run:
npx lerna bootstrap
Add this dependency to your subproject like so:
npx lerna add @keymanapp/models-types path/to/your/subproject