epwing.go

package yomichan

import (
	"fmt"
	"regexp"
	"strconv"
	"strings"

	zig "foosoft.net/projects/zero-epwing-go"
)

type epwingExtractor interface {
	extractTerms(entry zig.BookEntry, sequence int) []dbTerm
	extractKanji(entry zig.BookEntry) []dbKanji
	getFontNarrow() map[int]string
	getFontWide() map[int]string
	getRevision() string
}

func epwingExportDb(inputPath, outputPath, language, title string, stride int, pretty bool) error {
	book, err := zig.Load(inputPath)
	if err != nil {
		return err
	}

	translateExp := regexp.MustCompile(`{{([nw])_(\d+)}}`)
	epwingExtractors := map[string]epwingExtractor{
		"三省堂　スーパー大辞林":    makeDaijirinExtractor(),
		"大辞泉":            makeDaijisenExtractor(),
		"明鏡国語辞典":         makeMeikyouExtractor(),
		"故事ことわざの辞典":      makeKotowazaExtractor(),
		"研究社　新和英大辞典　第５版": makeWadaiExtractor(),
		"広辞苑第六版":         makeKoujienExtractor(),
		"付属資料":           makeKoujienExtractor(),
		"学研国語大辞典":        makeGakkenExtractor(),
		"古語辞典":           makeGakkenExtractor(),
		"故事ことわざ辞典":       makeGakkenExtractor(),
		"学研漢和大字典":        makeGakkenExtractor(),
		"小学館２":           makeShougakukan2Extractor(),
	}

	var (
		terms     dbTermList
		kanji     dbKanjiList
		revisions []string
		titles    []string
		sequence  int
	)

	for _, subbook := range book.Subbooks {
		if extractor, ok := epwingExtractors[subbook.Title]; ok {
			fontNarrow := extractor.getFontNarrow()
			fontWide := extractor.getFontWide()

			translate := func(str string) string {
				for _, matches := range translateExp.FindAllStringSubmatch(str, -1) {
					var font map[int]string
					if matches[1] == "n" {
						font = fontNarrow
					} else {
						font = fontWide
					}

					code, _ := strconv.Atoi(matches[2])
					replacement, ok := font[code]
					if !ok {
						replacement = "�"
					}

					str = strings.Replace(str, matches[0], replacement, -1)
				}
				pattern := regexp.MustCompile("\n+")
				str = pattern.ReplaceAllLiteralString(str, "\n")

				return str
			}

			for _, entry := range subbook.Entries {
				entry.Heading = translate(entry.Heading)
				entry.Text = translate(entry.Text)

				terms = append(terms, extractor.extractTerms(entry, sequence)...)
				kanji = append(kanji, extractor.extractKanji(entry)...)

				sequence++
			}

			revisions = append(revisions, extractor.getRevision())
			titles = append(titles, subbook.Title)
		} else {
			return fmt.Errorf("failed to find compatible extractor for '%s'", subbook.Title)
		}
	}

	if title == "" {
		title = strings.Join(titles, ", ")
	}

	recordData := map[string]dbRecordList{
		"kanji": kanji.crush(),
		"term":  terms.crush(),
	}

	index := dbIndex{
		Title:     title,
		Revision:  strings.Join(revisions, ";"),
		Sequenced: true,
	}

	return writeDb(
		outputPath,
		index,
		recordData,
		stride,
		pretty,
	)
}