wehub-resource-sync
73 lines
2.3 KiB
JavaScript
73 lines
2.3 KiB
JavaScript
import test from 'tape'
|
|
import nlp from './_lib.js'
|
|
const here = '[three/topics] '
|
|
|
|
test('topics:', function (t) {
|
|
const list = [
|
|
['Tony Hawk lives in Toronto. Tony Hawk is cool.', 'tony hawk'],
|
|
['I live Toronto. I think Toronto is cool.', 'toronto'],
|
|
['The EACD united in 1972. EACD must follow regulations.', 'eacd'],
|
|
// ['The Elkjsdflkjsdf sells hamburgers. I think the Elkjsdflkjsdf eats turky.', 'elkjsdflkjsdf'],
|
|
// ["Toronto's citizens love toronto!", 'toronto'],
|
|
["Toronto's citizens love toronto", 'toronto'],
|
|
]
|
|
list.forEach(function (a) {
|
|
const arr = nlp(a[0]).topics().out('freq')
|
|
t.equal(arr[0].normal, a[1], here + a[0])
|
|
})
|
|
t.end()
|
|
})
|
|
|
|
test('topics-false-positives:', function (t) {
|
|
const arr = [
|
|
'somone ate her lunch',
|
|
'everybody is dancing all night',
|
|
"a man and a woman ate her son's breakfast",
|
|
'my brother walks to school',
|
|
`She's coming by`,
|
|
`if she doesn't like something about us she can keep us off`,
|
|
` She's it! She could be a soap opera.`,
|
|
`she's a little dare-devil!`,
|
|
]
|
|
arr.forEach(function (str, i) {
|
|
const doc = nlp(str).topics()
|
|
t.equal(doc.length, 0, here + ' #' + i + ' -> ' + doc.out())
|
|
})
|
|
t.end()
|
|
})
|
|
|
|
test('topics-basic', function (t) {
|
|
const doc = nlp('i went to Gloop University in Paris, France, with John H. Smith')
|
|
const arr = doc.topics().out('array')
|
|
// t.deepEqual(arr, ['Gloop University', 'Paris, France,', 'John H. Smith'], here + 'found all three')
|
|
t.equal(arr.length, 3, here + 'found all three')
|
|
t.end()
|
|
})
|
|
|
|
test('misc entities', function (t) {
|
|
const doc = nlp('The Children are right to laugh at you, Ralph')
|
|
let m = doc.people()
|
|
t.equal(m.length, 1, here + 'one person')
|
|
|
|
m = doc.places()
|
|
t.equal(m.length, 0, here + 'no places')
|
|
|
|
m = doc.organizations()
|
|
t.equal(m.length, 0, here + 'no organizations')
|
|
|
|
m = doc.topics()
|
|
t.equal(m.length, 1, here + 'one entity')
|
|
t.end()
|
|
})
|
|
|
|
test('topics concat:', function (t) {
|
|
const things = nlp('spencer and danny are in Paris France and germany for Google Inc and IBM')
|
|
.topics()
|
|
.json({ normal: true, trim: true })
|
|
.map(o => o.normal)
|
|
const want = ['spencer', 'danny', 'paris france', 'germany', 'google inc', 'ibm']
|
|
// t.equal(things.join(', '), want.join(', '), here + 'found right things')
|
|
t.equal(things.length, want.length, here + 'found right things')
|
|
t.end()
|
|
})
|