Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

feat(nlp): expand aliases to improve natural language understanding and dedupe days & months #644

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

feat(nlp): expand aliases to improve natural language understanding and dedupe days & months #644

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
52 changes: 26 additions & 26 deletions src/nlp/i18n.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,14 +40,14 @@ const ENGLISH: Language = {
numberAsText: /^(one|two|three)/i,
every: /^every/i,
'day(s)': /^days?/i,
'weekday(s)': /^weekdays?/i,
'week(s)': /^weeks?/i,
'hour(s)': /^hours?/i,
'minute(s)': /^minutes?/i,
'month(s)': /^months?/i,
'year(s)': /^years?/i,
'weekday(s)': /^(weekdays?|business\s+days?|work\s+days?)/i,
'week(s)': /^(weeks?|wks?|weekly)/i,
'hour(s)': /^(hours?|hrs?|hourly)/i,
'minute(s)': /^(minutes?|mins?)/i,
'month(s)': /^(months?|monthly)/i,
'year(s)': /^(years?|yearly|annually)/i,
on: /^(on|in)/i,
at: /^(at)/i,
at: /^(at|@)/i,
the: /^the/i,
first: /^first/i,
second: /^second/i,
Expand All @@ -57,26 +57,26 @@ const ENGLISH: Language = {
for: /^for/i,
'time(s)': /^times?/i,
until: /^(un)?til/i,
monday: /^mo(n(day)?)?/i,
tuesday: /^tu(e(s(day)?)?)?/i,
wednesday: /^we(d(n(esday)?)?)?/i,
thursday: /^th(u(r(sday)?)?)?/i,
friday: /^fr(i(day)?)?/i,
saturday: /^sa(t(urday)?)?/i,
sunday: /^su(n(day)?)?/i,
january: /^jan(uary)?/i,
february: /^feb(ruary)?/i,
march: /^mar(ch)?/i,
april: /^apr(il)?/i,
monday: /^(mo\.?(n\.?(days?)?)?)/i,
tuesday: /^(tu\.?(e\.?(s\.?(days?)?)?)?)/i,
wednesday: /^(we\.?(d\.?(n\.?(esdays?)?)?)?)/i,
thursday: /^(th\.?(u\.?(r\.?(s\.?(days?)?)?)?)?)/i,
friday: /^(fr\.?(i\.?(days?)?)?)/i,
saturday: /^(sa\.?(t\.?(urdays?)?)?)/i,
sunday: /^(su\.?(n\.?(days?)?)?)/i,
january: /^(jan\.?(uary)?)/i,
february: /^(feb\.?(ruary)?)/i,
march: /^(mar\.?(ch)?)/i,
april: /^(apr\.?(il)?)/i,
may: /^may/i,
june: /^june?/i,
july: /^july?/i,
august: /^aug(ust)?/i,
september: /^sep(t(ember)?)?/i,
october: /^oct(ober)?/i,
november: /^nov(ember)?/i,
december: /^dec(ember)?/i,
comma: /^(,\s*|(and|or)\s*)+/i,
june: /^(june?|jun\.)/i,
july: /^(july?|jul\.)/i,
august: /^(aug(ust)?|aug\.)/i,
september: /^(sep(t(ember)?)?|sept\.|sep\.)/i,
october: /^(oct(ober)?|oct\.)/i,
november: /^(nov(ember)?|nov\.)/i,
december: /^(dec(ember)?|dec\.)/i,
comma: /^(,\s*|(and|or|&)\s*)+/i,
},
}

Expand Down
22 changes: 18 additions & 4 deletions src/nlp/parsetext.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,6 +102,14 @@ export default function parseText(text: string, language: Language = ENGLISH) {
if (!ttr.start(text)) return null

S()
// deduplicate arrays
if (options.byweekday && Array.isArray(options.byweekday)) {
options.byweekday = [...new Set(options.byweekday)]
}
if (options.bymonth && Array.isArray(options.bymonth)) {
options.bymonth = [...new Set(options.bymonth)]
}

return options

function S() {
Expand All @@ -114,7 +122,15 @@ export default function parseText(text: string, language: Language = ENGLISH) {
switch (ttr.symbol) {
case 'day(s)':
options.freq = RRule.DAILY
if (ttr.nextSymbol()) {
ttr.nextSymbol()
if (!ttr.isDone()) {
if (ttr.accept('on')) {
const m = decodeM()
if (m) {
options.bymonth = [m]
ttr.nextSymbol()
}
}
AT()
F()
}
Expand Down Expand Up @@ -186,7 +202,6 @@ export default function parseText(text: string, language: Language = ENGLISH) {

if (!ttr.nextSymbol()) return

// TODO check for duplicates
while (ttr.accept('comma')) {
if (ttr.isDone()) throw new Error('Unexpected end')

Expand Down Expand Up @@ -222,7 +237,6 @@ export default function parseText(text: string, language: Language = ENGLISH) {

if (!ttr.nextSymbol()) return

// TODO check for duplicates
while (ttr.accept('comma')) {
if (ttr.isDone()) throw new Error('Unexpected end')

Expand All @@ -242,7 +256,7 @@ export default function parseText(text: string, language: Language = ENGLISH) {
break

default:
throw new Error('Unknown symbol')
throw new Error(`Unknown symbol: ${ttr.symbol}`)
}
}

Expand Down
67 changes: 47 additions & 20 deletions test/nlp.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,45 @@ const texts = [
['Every week for 20 times', 'RRULE:FREQ=WEEKLY;COUNT=20'],
]

const fromTexts = [
...texts,
// Testing more flexible weekday formats
['Every Mon & Wed', 'RRULE:FREQ=WEEKLY;BYDAY=MO,WE'],
['Every Mon and Wed', 'RRULE:FREQ=WEEKLY;BYDAY=MO,WE'],
['Every Tues & Thurs', 'RRULE:FREQ=WEEKLY;BYDAY=TU,TH'],
['Every tu. and thurs. & wed.', 'RRULE:FREQ=WEEKLY;BYDAY=TU,TH,WE'],

// Testing business/work day variations
['Every business day', 'RRULE:FREQ=WEEKLY;BYDAY=MO,TU,WE,TH,FR'],
['Every work day', 'RRULE:FREQ=WEEKLY;BYDAY=MO,TU,WE,TH,FR'],

// Testing month abbreviations
['Every month in Jan', 'RRULE:FREQ=MONTHLY;BYMONTH=1'],
['Every week in Sept.', 'RRULE:FREQ=WEEKLY;BYMONTH=9'],
['Every day in Dec.', 'RRULE:FREQ=DAILY;BYMONTH=12'],

// Testing time unit variations
['Every hr', 'RRULE:FREQ=HOURLY'],
['Every 2 hrs', 'RRULE:INTERVAL=2;FREQ=HOURLY'],
['Every min', 'RRULE:FREQ=MINUTELY'],
['Every 5 mins', 'RRULE:INTERVAL=5;FREQ=MINUTELY'],

// Testing @ symbol for at
['Every day @ 10', 'RRULE:FREQ=DAILY;BYHOUR=10'],
['Every week @ 9', 'RRULE:FREQ=WEEKLY;BYHOUR=9'],

// Testing plural variations of days
['Every Monday and Wednesdays', 'RRULE:FREQ=WEEKLY;BYDAY=MO,WE'],
['Every Tuesdays & Thursdays', 'RRULE:FREQ=WEEKLY;BYDAY=TU,TH'],

// Deduplicates
[
'Every Monday, Tuesday, Wednesday, and Monday',
'RRULE:FREQ=WEEKLY;BYDAY=MO,TU,WE',
],
['Every month in Jan & January', 'RRULE:FREQ=MONTHLY;BYMONTH=1'],
]

const toTexts = [
...texts,
[
Expand All @@ -41,30 +80,18 @@ const toTexts = [
]

describe('NLP', () => {
it('fromText()', function () {
texts.forEach(function (item) {
const text = item[0]
const str = item[1]
expect(RRule.fromText(text).toString()).toBe(str)
})
it.each(fromTexts)('fromText() %s', (text, str) => {
expect(RRule.fromText(text).toString()).toBe(str)
})

it('toText()', function () {
toTexts.forEach(function (item) {
const text = item[0]
const str = item[1]
expect(RRule.fromString(str).toText().toLowerCase()).toBe(
text.toLowerCase()
)
})
it.each(toTexts)('toText() %s', (text, str) => {
expect(RRule.fromString(str).toText().toLowerCase()).toBe(
text.toLowerCase()
)
})

it('parseText()', function () {
texts.forEach(function (item) {
const text = item[0]
const str = item[1]
expect(optionsToString(RRule.parseText(text))).toBe(str)
})
it.each(texts)('parseText() %s', (text, str) => {
expect(optionsToString(RRule.parseText(text))).toBe(str)
})

it('permits integers in byweekday (#153)', () => {
Expand Down