taipa/lib/dparser/tagger.js

Taiwanese morphological parsing library

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.tag = void 0;
const rules_1 = require("./rules");
const symbols_1 = require("./symbols");
const dictionary_1 = require("./dictionary");
const lemmatizer_1 = require("../unchange/lemmatizer");
function tag(features) {
    // const pairs: Pairs<string, string> = [];
    const pairs = [];
    let expecting = '';
    for (let i = 0; i < features.length; i++) {
        if (rules_1.proceedingAdverbialParticles.long === features[i].token &&
            dictionary_1.baseVerbs.includes(features[i].nextToken) &&
            dictionary_1.subsidiariesA.includes(features[i].nextToken2)) {
            pairs.push([features[i].token, symbols_1.Tagset.padv]);
            continue;
        }
        if (rules_1.inflectedAdverbialParticles.includes(features[i].token)) {
            pairs.push([features[i].token, symbols_1.Tagset.padv]);
            continue;
        }
        if (rules_1.inflectedPhrasalVerbParticles.includes(features[i].token)) {
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (dictionary_1.baseVerbs.includes(features[i].token) &&
            pairs.length == 1 &&
            rules_1.inflectedAdverbialParticles.includes(pairs[pairs.length - 1][0]) &&
            dictionary_1.subsidiariesA.includes(features[i].nextToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (dictionary_1.baseVerbs.includes(features[i].token) &&
            dictionary_1.auxiliaries.includes(features[i].prevToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (dictionary_1.baseVerbs.includes(features[i].token) &&
            dictionary_1.subsidiariesA.includes(features[i].nextToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (dictionary_1.baseVerbs.includes(features[i].token) &&
            dictionary_1.basePhrasalVerbParticles.includes(features[i].nextToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (dictionary_1.baseVerbs.includes(features[i].token) &&
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].prevToken) &&
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].prevToken2)) {
            // a verb after a preceding phrasal verb
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if ((0, lemmatizer_1.lemmatize)(features[i].token).getLemmas().length == 3 &&
            dictionary_1.baseVerbs.includes((0, lemmatizer_1.lemmatize)(features[i].token).getLemmas()[0].literal) &&
            !dictionary_1.basePhrasalVerbParticles.includes(features[i].nextToken) && // object of the verb
            (dictionary_1.basePhrasalVerbParticles.includes(features[i].nextToken2) ||
                rules_1.inflectedPhrasalVerbParticles.includes(features[i].nextToken2))) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if ((dictionary_1.basePhrasalVerbParticles.includes(features[i].token) ||
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].token)) &&
            features[i].prevToken &&
            pairs[pairs.length - 1][1] === symbols_1.Tagset.nn &&
            features[i].prevToken2 &&
            pairs[pairs.length - 2][1] === symbols_1.Tagset.vb) {
            // we may also check if the next token is a phrasal verb particle
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (dictionary_1.basePhrasalVerbParticles.includes(features[i].token) &&
            features[i].prevToken &&
            pairs[pairs.length - 1][1] === symbols_1.Tagset.ppv &&
            features[i].prevToken2 &&
            pairs[pairs.length - 2][1] === symbols_1.Tagset.nn) {
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (features[i].nextToken &&
            (0, rules_1.isPhrasalVerbVp)(features[i].token, features[i].nextToken)) {
            // is main verb of phrasal verb
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (rules_1.inflectedPhrasalVerbParticles.includes(features[i].token) &&
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].prevToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (dictionary_1.subsidiariesA.includes(features[i].token) &&
            dictionary_1.baseVerbs.includes(features[i].prevToken)) {
            // to check the tone pattern. to check if last word
            pairs.push([features[i].token, symbols_1.Tagset.psub]);
            continue;
        }
        if (dictionary_1.subsidiariesA.includes(features[i].token) &&
            dictionary_1.basePhrasalVerbParticles.includes(features[i].prevToken) &&
            dictionary_1.baseVerbs.includes(features[i].prevToken2)) {
            // to check the tone pattern. to check if last word
            pairs.push([features[i].token, symbols_1.Tagset.psub]);
            continue;
        }
        if (rules_1.inflectedPersonalPronouns.includes(features[i].token)) {
            pairs.push([features[i].token, symbols_1.Tagset.npr]);
            continue;
        }
        if (features[i].prevToken &&
            (0, rules_1.isPhrasalVerbVp)(features[i].prevToken, features[i].token)) {
            // is a particle of phrasal verb
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (features[i].prevToken &&
            features[i].prevToken2 &&
            (0, rules_1.isPhrasalVerbVpp)(features[i].prevToken2, features[i].prevToken, features[i].token)) {
            // is the 2nd particle of phrasal verb
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (rules_1.inflectedVerbs.includes(features[i].token) &&
            features[i].nextToken &&
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].nextToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (rules_1.inflectedPhrasalVerbParticles.includes(features[i].token) &&
            features[i].nextToken &&
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].nextToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (rules_1.inflectedVerbs.includes(features[i].token) &&
            features[i].nextToken &&
            rules_1.inflectedPhrasalVerbParticles.includes(features[i].nextToken2)) {
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (rules_1.inflectedPhrasalVerbParticles.includes(features[i].token) &&
            features[i].prevToken &&
            rules_1.inflectedVerbs.includes(features[i].prevToken)) {
            pairs.push([features[i].token, symbols_1.Tagset.ppv]);
            continue;
        }
        if (dictionary_1.demonstrativePronouns.includes(features[i].token)) {
            pairs.push([features[i].token, symbols_1.Tagset.npr]);
            continue;
        }
        if (dictionary_1.auxiliaries.includes(features[i].token)) {
            pairs.push([features[i].token, symbols_1.Tagset.aux]);
            continue;
        }
        if (dictionary_1.seperateVVCompounds.map(it => it[0]).includes(features[i].token)) {
            // the first word of a VV compound
            expecting = dictionary_1.seperateVVCompounds.filter(it => it[0] === features[i].token)[0][1];
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        if (dictionary_1.seperateVVCompounds.map(it => it[1]).includes(expecting) &&
            features[i].token === expecting) {
            // the second word of a VV compound
            expecting = '';
            pairs.push([features[i].token, symbols_1.Tagset.vb]);
            continue;
        }
        pairs.push([features[i].token, symbols_1.Tagset.nn]);
    }
    return pairs;
}
exports.tag = tag;
//# sourceMappingURL=tagger.js.map