Can you give me the correct version of Lexical model source documents
My request is differ tonation but undiffer the upper and lower cases as I have mentioned before.
Please don’t send me several rows and hope me can add them inside and make it work. Because I can not.
Please test whether it can function well them send the whole code to me. So please compile the code in latest keyman developper lexical model then send it to me.
Because the code is so delicate, a punctional fault can make the code totally useless and I probably can never find the fault.
If you have made the code once then I can copy the code again and again to make new keyboard lexical model
like that, However, I don’t know how to change word breaker, I want to change the hyphen and the space to not wordbreaker because in Taiwanese the hyphen can be part of the word
How can we do that?
Have you sent the wordlist file to us anywhere that we might’ve forgotten to pick it up? If not, please send the tsv file here so I can use your wordlist to try out the code.
Please allow us some time to look into the code for you.
const source: LexicalModelSource = {
format: ‘trie-1.0’,
wordBreaker: {
use: ‘default’, // we want to use the default word breaker, BUT!
// CUSTOMIZE THIS:
joinWordsAt: [‘-’, ’ ', ‘1’, ‘2’, ‘3’, ‘4’, ‘5’, ‘6’, ‘7’, ‘8’, ‘9’, ‘0’, ‘a’, 'a '], // join words that contain hyphens
}
sources: [‘wordlist.tsv’],
languageUsesCasing: true,
searchTermToKey: function(term, applyCasing) {
return Array.from(term)
.map(function(c) { return applyCasing(‘lower’, c) };
}
};
export default source;
here is my new code, it seems work more
However it still shows a issue, the keyman developper can not identify words started with space for example I want to include a " ê" in the model, it will automatically remove the space for me.
By the way, can keyman lexical model have this feature:
When I add a punctuation after the automatically added space, the space will be deleted.
I know iPhone keyboard have this feature. Can keyman keyboard lexical model also get this feature.
Thanks
This one is on our radar, but unfortunately is not likely to be addressed for the next version. We’re aware of the desire for this and do have some ideas in mind for implementing it well in a manner generalizable for all languages; it’s just that it involves some tweaks to the model format, and we want to get those right “the first time.”