lrctransliterate_kagome

main.go

package main

import (
	"fmt"
	"strings"
	"unicode"
	"bufio"
	"os"
	
	"github.com/gojp/kana"
	"github.com/ikawaha/kagome-dict/ipa"
	"github.com/ikawaha/kagome/v2/tokenizer"
)

func main() {
	//input := `こぼれたままの流星群  一秒  一秒`
	input := ""
	input_kana := ""
	input_kana2 := ""
	
	t, err := tokenizer.New(ipa.Dict(), tokenizer.OmitBosEos())
	if err != nil {
		panic(err)
	}
	
	scanner := bufio.NewScanner(os.Stdin)
    for {
        //fmt.Print("Enter Text: ")
        // Scans a line from Stdin(Console)
        scanner.Scan()
        // Holds the string that scanned
        input = scanner.Text()
        if len(input) == 0 {
			break
		}

		lines := strings.Split(input, "\n")
		//fmt.Println(lines)
		for _, line := range lines {
			line = strings.TrimSpace(line)
			
			if ! strings.Contains(line, "]") {
				continue
			}
			
			s := strings.SplitN(line, "]", 2)  // TODO: handle multiple tags on the same line
			ts, line_text := s[0], s[1]
			//fmt.Println(line_text)
			
			// transliterate https://github.com/ikawaha/kagome/issues/308

			// tokenize
			input_kana = ""
			input_kana2 = ""
			tokens := t.Tokenize(line_text)
			tmpChunk := ""
			for _, token := range tokens {
				//features := strings.Join(token.Features(), ",")
				//fmt.Println(token.ID)  // general id
				//fmt.Println(token.Index)  // position
				//fmt.Println(token.Class)
				/*
				if p, ok := token.Pronunciation(); ok {
					tmpChunk = p
				} else if r, ok := token.Reading(); ok {
					tmpChunk = r // fallback to reading if pronunciation is not available
				} else {
					tmpChunk = strings.ToUpper(token.Surface)
				}
				input_kana2 = input_kana2 + tmpChunk + " "
        */
			
				if token.Class == tokenizer.UNKNOWN {
					input_kana = input_kana + strings.ToUpper(token.Surface) + " "
				} else {
					token_pronunciation, _ := token.Pronunciation()
					input_kana = input_kana + token_pronunciation + " "
				}
				
				
			}
			//fmt.Println(input_kana)
			//fmt.Println(input_kana2)
			
			yomi := strings.Map(func(r rune) rune {
				if unicode.IsLetter(r) {
					return r
				}
				if unicode.IsGraphic(r) {
				//if unicode.IsSpace(r) {
					return r
				}
				return -1

			}, kana.KanaToRomaji(input_kana2))

			if yomi == "" {
				continue
			}
			
			// remove extra spaces
			yomi = strings.TrimSpace(yomi)
			yomi = strings.Replace(yomi, "   ", "  ", -1)
			yomi = strings.Replace(yomi, "  ", " ", -1)

			fmt.Println(ts + "]" + yomi)
		}
	}
}