wehub-resource-sync
119 lines
3.6 KiB
JavaScript
119 lines
3.6 KiB
JavaScript
import test from 'tape'
|
|
import nlp from '../_lib.js'
|
|
const here = '[two/json] '
|
|
|
|
const hasTag = function (term, want) {
|
|
if (!term || !term.tags) {
|
|
return false
|
|
}
|
|
return term.tags.some(tag => tag === want)
|
|
}
|
|
|
|
test('json out default', function (t) {
|
|
const doc = nlp('who are you? what is this?')
|
|
const json = doc.json({ terms: true })
|
|
t.equal(json.length, 2, here + 'json-len')
|
|
t.equal(json[1].text, 'what is this?', here + 'json-text')
|
|
t.equal(json[1].terms.length, 3, here + 'json-three-terms')
|
|
t.equal(hasTag(json[1].terms[1], 'Copula'), true, here + 'json-has-tag')
|
|
t.end()
|
|
})
|
|
|
|
test('json out trim', function (t) {
|
|
const doc = nlp('who are you? what is this?')
|
|
const json = doc.json({ trim: false, terms: false })
|
|
t.equal(json.length, 2, here + 'json-len')
|
|
t.equal(json[0].text, 'who are you?', here + 'json-text')
|
|
t.equal(json[1].text, 'what is this?', here + 'json-text')
|
|
t.equal(json[1].terms, undefined, here + 'json-no-terms')
|
|
t.end()
|
|
})
|
|
|
|
test('json out implicit', function (t) {
|
|
const str = `he isn't`
|
|
const doc = nlp(str)
|
|
const json = doc.json()
|
|
t.equal(json.length, 1, here + 'json-len')
|
|
t.equal(json[0].text, str, here + 'json-text')
|
|
t.equal(json[0].terms.length, 3, here + 'json-three-terms')
|
|
const t0 = json[0].terms[0] || {}
|
|
t.equal(t0.implicit, undefined, here + 'has-no-implicit')
|
|
const t1 = json[0].terms[1] || {}
|
|
t.equal(hasTag(t1, 'Copula'), true, here + 'json-has-tag')
|
|
t.equal(t1.machine, 'is', here + 'has-implicit1')
|
|
t.equal(hasTag(json[0].terms[2], 'Negative'), true, here + 'json-has-tag2')
|
|
const t2 = json[0].terms[2] || {}
|
|
t.equal(t2.machine, 'not', here + 'has-implicit2')
|
|
t.end()
|
|
})
|
|
|
|
test('json terms out', function (t) {
|
|
const doc = nlp(`she was not`)
|
|
const json = doc.json({ text: false, terms: { clean: true, id: true, whitespace: true } })
|
|
t.equal(json.length, 1, 'json-len')
|
|
t.equal(json[0].text, undefined, here + 'json-text')
|
|
t.equal(json[0].terms.length, 3, here + 'json-terms-length')
|
|
// let t0 = json[0].terms[0] || {}
|
|
// t.equal(t0.bestTag, 'Pronoun', 'json-terms-bestTag')
|
|
const t1 = json[0].terms[1] || {}
|
|
// t.equal(t1.bestTag, 'Copula', 'json-terms-bestTag1')
|
|
// let t2 = json[0].terms[2] || {}
|
|
// t.equal(t2.bestTag, 'Negative', 'json-terms-bestTag2')
|
|
|
|
t.equal(t1.pre, '', here + 'json-terms-whitespace-pre')
|
|
t.equal(t1.post, ' ', here + 'json-terms-whitespace-post')
|
|
t.end()
|
|
})
|
|
|
|
// test('json-index:', function (t) {
|
|
// let doc = nlp(`john is not really walking`)
|
|
// let obj = doc.match('really').json({ index: true })[0]
|
|
// t.equal(obj.terms[0].index[1], 3, 'index:3')
|
|
// t.equal(obj.index, 3, 'phrase-index:3')
|
|
// t.end()
|
|
// })
|
|
|
|
// test('json-unique:', function (t) {
|
|
// let doc = nlp(`a b c b a`)
|
|
// let arr = doc.terms().json({ unique: true, terms: false, count: true })
|
|
// t.equal(arr.length, 3, 'no duplicates')
|
|
// t.end()
|
|
// })
|
|
|
|
|
|
|
|
test('out-custom:', function (t) {
|
|
const doc = nlp('The competent drum work of Don Brewer?')
|
|
const arr = doc.json({
|
|
terms: {
|
|
text: true,
|
|
normal: false,
|
|
tags: true,
|
|
sdf: true,
|
|
root: true,
|
|
},
|
|
})[0].terms
|
|
t.equal(arr[0].text, 'The', here + 'has text')
|
|
// t.equal(arr[5].root, 'don', 'has root')
|
|
t.equal(arr[5].sdf, undefined, 'has no sdf')
|
|
t.equal(
|
|
arr[0].tags.some(tag => tag === 'Determiner'),
|
|
true,
|
|
here + 'has tags'
|
|
)
|
|
t.end()
|
|
})
|
|
|
|
test('out:', function (t) {
|
|
|
|
const doc = nlp(`john is not really walking`)
|
|
let json = doc.out('json')
|
|
t.equal(json.length, 1, here + 'one-json')
|
|
json = doc.out('offset')
|
|
t.equal(json.length, 1, here + 'offset-out')
|
|
|
|
// let terms = doc.out('terms')
|
|
// t.equal(terms[0], 'john', here + 'terms-out')
|
|
t.end()
|
|
})
|