spencermountain--compromise
wehub-resource-sync
109 行
2.7 KiB
JavaScript
109 行
2.7 KiB
JavaScript
import test from 'tape'
|
|
import nlp from '../_lib.js'
|
|
const here = '[two/match-contraction] '
|
|
|
|
test('match-contractions', function (t) {
|
|
const doc = nlp(`i haven't done it`)
|
|
let m = doc.match(`have not done`)
|
|
t.equal(m.text(), `haven't done`, here + 'full-text')
|
|
|
|
m = doc.match(`have not`)
|
|
t.equal(m.text(), `haven't`, here + 'first-half-found')
|
|
|
|
m = doc.match(`not done`)
|
|
t.equal(m.text(), `done`, here + 'second-half-found')
|
|
|
|
m = doc.match(`haven't`)
|
|
t.equal(m.text(), `haven't`, here + 'match-contraction')
|
|
|
|
m = doc.match(`haven't done`)
|
|
t.equal(m.text(), `haven't done`, here + 'match-contraction-full')
|
|
|
|
m = doc.match(`have done`)
|
|
t.equal(m.text(), ``, here + 'only-outsides')
|
|
|
|
t.end()
|
|
})
|
|
|
|
test('false-positive-contractions', function (t) {
|
|
const doc = nlp(`i have done it`)
|
|
let m = doc.match(`have not done`)
|
|
t.equal(m.text(), ``, here + 'have not done')
|
|
|
|
m = doc.match(`have not`)
|
|
t.equal(m.text(), ``, here + 'have not')
|
|
|
|
m = doc.match(`not`)
|
|
t.equal(m.text(), ``, here + 'not')
|
|
|
|
m = doc.match(`not done`)
|
|
t.equal(m.text(), ``, here + 'not done')
|
|
|
|
m = doc.match(`haven't`)
|
|
t.equal(m.text(), ``, here + `haven't`)
|
|
|
|
m = doc.match(`haven't done`)
|
|
t.equal(m.text(), ``, here + `haven't done`)
|
|
|
|
m = doc.match(`have not done`)
|
|
t.equal(m.text(), ``, here + `have not done`)
|
|
|
|
m = doc.match(`have done`)
|
|
t.equal(m.text(), `have done`, here + `have done`)
|
|
t.end()
|
|
})
|
|
|
|
test('i am contraction', function (t) {
|
|
const doc = nlp(`so i'm glad`)
|
|
let m = doc.match(`i am`)
|
|
t.equal(m.text(), `i'm`, here + 'i am')
|
|
|
|
m = doc.match(`i`)
|
|
t.equal(m.text(), `i'm`, here + `i`)
|
|
|
|
m = doc.match(`am`)
|
|
t.equal(m.text(), ``, here + `am`)
|
|
|
|
m = doc.match(`i am glad`)
|
|
t.equal(m.text(), `i'm glad`, here + `i'm glad`)
|
|
|
|
m = doc.match(`i glad`)
|
|
t.equal(m.text(), ``, here + 'i glad')
|
|
t.end()
|
|
})
|
|
|
|
test('contraction-optional', function (t) {
|
|
const doc = nlp(`so i'm glad`)
|
|
let m = doc.match(`i am?`)
|
|
t.equal(m.text(), `i'm`, here + 'i am?')
|
|
|
|
m = doc.match(`i am?`)
|
|
t.equal(m.text(), `i'm`, here + `i am?`)
|
|
|
|
m = doc.match(`am glad?`)
|
|
t.equal(m.text(), `glad`, here + `am glad?`)
|
|
|
|
m = doc.match(`i am? glad`)
|
|
t.equal(m.text(), `i'm glad`, here + `i am? glad`)
|
|
|
|
m = doc.match(`i glad?`)
|
|
t.equal(m.text(), `i'm`, here + 'i glad?')
|
|
t.end()
|
|
})
|
|
|
|
|
|
|
|
test('lookup contraction', function (t) {
|
|
const arr = [
|
|
'foobar',
|
|
'marines',
|
|
'afghanistan',
|
|
'foo',
|
|
]
|
|
const trie = nlp.buildTrie(arr)
|
|
const res = nlp(`so we're adding 3201 Marines to our forces in Afghanistan.`).lookup(trie)
|
|
t.equal(res.has('marines'), true, 'post-contraction found first one')
|
|
t.equal(res.has('afghanistan'), true, 'post-contraction found second one')
|
|
t.end()
|
|
})
|