claude
997f92f5c4
ruNumerals was a second copy of the number words. It stopped at fifty, had no oblique forms, and disagreed with lexicon_ru_v1.json about its own members, so "к семи" was not the hour "в семь" was. The lexicon now carries the oblique cardinals and numwords.go asks lexicon.Cardinal. "час" and "часу" stay local: they are the hour noun as often as the number one, and nobody counts "час яблок". reminderbody.go built its markers from three inline word lists. Two of them are new lexicon sets, reminder_verbs and parts_of_day, and the day offsets were already there. The alternation helper sorts by length so a longer form wins the regex. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
94 lines
3.4 KiB
Go
94 lines
3.4 KiB
Go
package router
|
|
|
|
import (
|
|
"strconv"
|
|
"strings"
|
|
|
|
"github.com/kami/maven/internal/lexicon"
|
|
)
|
|
|
|
// hourNouns — the two words that are the hour noun as often as they are the
|
|
// number one. "в час дня" means one o'clock, so rewriting it to "в 1 дня" is
|
|
// right either way.
|
|
//
|
|
// They are not cardinals and do not belong in the lexicon's number set: nobody
|
|
// counts "час яблок". Everything else this file reads comes from
|
|
// lexicon.Cardinal, which is where the number words live complete, oblique forms
|
|
// included (Vikunja #530). The table here used to be a second copy that stopped
|
|
// at fifty and disagreed with the lexicon about its own members.
|
|
var hourNouns = map[string]string{"час": "1", "часу": "1"}
|
|
|
|
// numeralDigit reports the digits a spoken number is written as, for a clock
|
|
// hour or the minutes after it.
|
|
func numeralDigit(word string) (string, bool) {
|
|
if d, ok := hourNouns[word]; ok {
|
|
return d, true
|
|
}
|
|
n, ok := lexicon.Cardinal(word)
|
|
if !ok {
|
|
return "", false
|
|
}
|
|
return strconv.Itoa(n), true
|
|
}
|
|
|
|
// numeralContext — the words that make a numeral a time. A numeral is only
|
|
// rewritten when one of these sits next to it, so "три яблока" in a note is
|
|
// left alone and "в три часа" is not.
|
|
var numeralContext = map[string]bool{
|
|
"в": true, "во": true, "к": true, "около": true, "на": true,
|
|
"часа": true, "часов": true, "час": true, "часу": true,
|
|
"утра": true, "вечера": true, "дня": true, "ночи": true,
|
|
"минут": true, "минуты": true, "минуту": true,
|
|
"at": true, "by": true,
|
|
}
|
|
|
|
// SpellOutDigits rewrites spoken numbers as digits so the date parsers see the
|
|
// shape they know.
|
|
//
|
|
// "напомни мне позвонить маме в семь вечера" parsed to nothing, while "в 19:00"
|
|
// parsed fine (Vikunja #469). Speech is where reminders come from, and speech
|
|
// says the hour in words, so this is not a long-tail case — it is the ordinary
|
|
// one. dateparser reads "в 7 вечера" through the qualifier rewrite the python
|
|
// script already does; it does not read "в семь вечера".
|
|
//
|
|
// Conservative by construction: a numeral is only rewritten when a time word
|
|
// stands beside it. "три часа" becomes "3 часа"; "три яблока" stays as it is,
|
|
// and a note or a fact carrying a spoken number is untouched.
|
|
func SpellOutDigits(text string) string {
|
|
toks := strings.Fields(text)
|
|
if len(toks) == 0 {
|
|
return text
|
|
}
|
|
out := make([]string, len(toks))
|
|
copy(out, toks)
|
|
for i, tok := range toks {
|
|
key := strings.ToLower(strings.Trim(tok, ".,!?;:«»\"'"))
|
|
digit, ok := numeralDigit(key)
|
|
if !ok {
|
|
continue
|
|
}
|
|
// "час" and "часу" are the hour noun as often as they are the number
|
|
// one, and rewriting "в час дня" to "в 1 дня" is right either way. What
|
|
// must not happen is rewriting the noun that gives another numeral its
|
|
// context: "в семь часов" must keep "часов".
|
|
if !hasTimeNeighbour(toks, i) {
|
|
continue
|
|
}
|
|
out[i] = digit
|
|
}
|
|
return strings.Join(out, " ")
|
|
}
|
|
|
|
// hasTimeNeighbour reports whether the token before or after i is a time word.
|
|
func hasTimeNeighbour(toks []string, i int) bool {
|
|
for _, j := range []int{i - 1, i + 1} {
|
|
if j < 0 || j >= len(toks) {
|
|
continue
|
|
}
|
|
if numeralContext[strings.ToLower(strings.Trim(toks[j], ".,!?;:«»\"'"))] {
|
|
return true
|
|
}
|
|
}
|
|
return false
|
|
}
|