Maven/internal/router/timementions.go

package router

import (
	"strconv"
	"strings"

	"github.com/kami/maven/internal/lexicon"
	"github.com/kami/maven/internal/morph"
)

// MentionsTime reports whether the sentence names a time at all, whether or not
// a parser could read it.
//
// The caller is the multi-turn seam. A reminder whose time slot is empty used to
// inherit the previous reminder's hour, so "напомни без четверти восемь
// выходить" landed at 07:30 because the turn before it had (V-543). Inheriting
// is right when the sentence names no time and wrong when it names one the
// parser missed, and this is the test that tells those apart. Missing the time
// he said means asking; inheriting means a wrong alarm he stops thinking about.
//
// Every signal here is a closed lexicon class or a digit, so this reads data and
// decides nothing about meaning.
func MentionsTime(text string) bool {
	toks := strings.Fields(strings.ToLower(text))
	for i, raw := range toks {
		tok := cleanWord(raw)
		if timeMarkers[tok] {
			return true
		}
		if isDigitClock(tok) {
			return true
		}
		if _, ok := numeralDigit(tok); ok && hasTimeNeighbour(toks, i) {
			return true
		}
		if _, _, ok := halfPastAt(toks, i); ok {
			return true
		}
		if _, _, _, ok := quarterToAt(toks, i); ok {
			return true
		}
		if isWeekday(tok) {
			return true
		}
	}
	return false
}

// isDigitClock reports whether the token is a written clock, "19:30". The
// minutes must be written as two digits, because a clock is and a score is not:
// "счёт 3:2" names no time.
func isDigitClock(tok string) bool {
	h, m, found := strings.Cut(tok, ":")
	if !found || len(m) != 2 {
		return false
	}
	hn, err := strconv.Atoi(h)
	if err != nil || hn < 0 || hn > 23 {
		return false
	}
	mn, err := strconv.Atoi(m)
	return err == nil && mn >= 0 && mn <= 59
}

// NamesADay reports whether the sentence names a calendar day: a weekday, a
// relative day word, or a month beside a date. A bare clock names none of them,
// which is what lets a parser roll it forward to the next occurrence.
//
// "напомни сегодня в 12:30" names the day, so it stays on it even when 12:30 has
// passed. Rolling that one forward would move a reminder he placed deliberately.
func NamesADay(text string) bool {
	for _, raw := range strings.Fields(strings.ToLower(text)) {
		tok := cleanWord(raw)
		if _, ok := lexicon.DayOffset(tok); ok {
			return true
		}
		if isWeekday(tok) || isMonth(tok) {
			return true
		}
	}
	return false
}

// isMonth reports whether the token is a month name. The lexicon holds the
// genitive, which is the form a spoken date uses: "10 июля".
func isMonth(tok string) bool {
	for m := 1; m <= 12; m++ {
		if tok == lexicon.MonthGenitive(m) {
			return true
		}
	}
	return false
}

// isWeekday reports whether the token is a day of the week in any case. The
// lexicon lists the nominative, and "в пятницу" is what a reminder says, so the
// match is by lemma — grammar is morph's job, not a second word list.
func isWeekday(tok string) bool {
	for i := 0; i < 7; i++ {
		if morph.SameWord(tok, lexicon.Weekday(i)) {
			return true
		}
	}
	return false
}

// timeMarkers — the words that name a time on their own: the qualifiers that
// turn an hour into a part of the day, the relative day words, the weekdays and
// the two relative openers. Built from the lexicon at init, so a word added
// there is a word this reads.
var timeMarkers = buildTimeMarkers()

func buildTimeMarkers() map[string]bool {
	m := map[string]bool{
		"утра": true, "вечера": true, "дня": true, "ночи": true,
		"часа": true, "часов": true, "час": true, "часу": true,
		"минут": true, "минуты": true, "минуту": true,
		"через": true, "полчаса": true, "сейчас": true,
		"am": true, "pm": true, "noon": true, "midnight": true,
	}
	for _, w := range lexicon.PartsOfDay() {
		m[w] = true
	}
	for _, w := range lexicon.DayOffsetWords() {
		m[w] = true
	}
	for i := 0; i < 7; i++ {
		if w := lexicon.Weekday(i); w != "" {
			m[w] = true
		}
	}
	for w := range halfWords {
		m[w] = true
	}
	for w := range minutesTo {
		m[w] = true
	}
	return m
}