Keyboard Support

Contact and Search

Keyman.com Homepage

Header bottom

Keyman.com

Other versions
Version 18.0Version 17.0 (current version)

On this page

LexicalModelSource interface

Home > @keymanapp/kmc-model > LexicalModelSource

LexicalModelSource interface

Base interface for a lexical model source definition

Signature:

export interface LexicalModelSource extends LexicalModelDeclaration 

Extends: LexicalModelDeclaration

Properties

Property Modifiers Type Description
applyCasing? readonly CasingFunction

(Optional) Specifies the casing rules for a language. Should implement three casing forms: - 'lower' -- a fully-lowercased version of the text appropriate for the language's use of the writing system. - 'upper' -- a fully-uppercased version of the text - 'initial' -- a version preserving the input casing aside from the initial character, which is uppercased (like with proper nouns and sentence-initial words in English sentences.)

This is only utilized if languageUsesCasing is defined and set to true. 14.0

languageUsesCasing? readonly boolean (Optional) When set to true, suggestions will attempt to match the case of the input text even if the lexicon entries use a different casing scheme due to search term keying effects. 14.0
punctuation? readonly LexicalModelPunctuation (Optional) Punctuation and spacing suggested by the model.
rootClass? readonly string (Optional) The name of the type to instantiate (without parameters) as the base object for a custom predictive model.
searchTermToKey? readonly WordformToKeySpec (Optional) How to simplify words, to convert them into simplified search keys This often involves removing accents, lowercasing, etc.
sources readonly Array<string>
wordBreaker? readonly WordBreakerSpec | SimpleWordBreakerSpec

(Optional) Which word breaker to use. Choose from:

- 'default' -- breaks according to Unicode UAX #29 ยง4.1 Default Word Boundary Specification, which works well for *most* languages. - 'ascii' -- a very simple word breaker, for demonstration purposes only. - word breaking function -- provide your own function that breaks words. - class-based word-breaker - may be supported in the future.