1
yomichan-import/epwing.go

118 lines
2.9 KiB
Go
Raw Normal View History

2021-01-01 22:31:58 +00:00
package yomichan
2016-11-16 04:53:21 +00:00
import (
2016-12-19 03:14:21 +00:00
"fmt"
2016-12-06 05:32:14 +00:00
"regexp"
2016-12-14 02:00:36 +00:00
"strconv"
"strings"
2016-11-16 04:53:21 +00:00
2022-07-04 03:59:33 +00:00
zig "foosoft.net/projects/zero-epwing-go"
2021-01-01 05:53:10 +00:00
)
2016-11-16 04:53:21 +00:00
2016-12-13 21:29:05 +00:00
type epwingExtractor interface {
2021-01-01 05:53:10 +00:00
extractTerms(entry zig.BookEntry, sequence int) []dbTerm
extractKanji(entry zig.BookEntry) []dbKanji
2016-12-14 02:00:36 +00:00
getFontNarrow() map[int]string
getFontWide() map[int]string
2016-12-24 05:52:49 +00:00
getRevision() string
2016-12-13 21:29:05 +00:00
}
2017-06-26 00:22:17 +00:00
func epwingExportDb(inputPath, outputPath, language, title string, stride int, pretty bool) error {
2021-01-01 05:53:10 +00:00
book, err := zig.Load(inputPath)
2016-12-19 02:47:07 +00:00
if err != nil {
return err
}
2016-11-16 04:53:21 +00:00
2016-12-14 02:00:36 +00:00
translateExp := regexp.MustCompile(`{{([nw])_(\d+)}}`)
2016-12-13 21:29:05 +00:00
epwingExtractors := map[string]epwingExtractor{
2017-02-19 23:31:37 +00:00
"三省堂 スーパー大辞林": makeDaijirinExtractor(),
"大辞泉": makeDaijisenExtractor(),
"明鏡国語辞典": makeMeikyouExtractor(),
2017-03-21 06:08:12 +00:00
"故事ことわざの辞典": makeKotowazaExtractor(),
2017-02-19 23:31:37 +00:00
"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(),
2022-05-16 15:13:55 +00:00
"小学館2": makeShougakukan2Extractor(),
2016-11-16 06:21:00 +00:00
}
2016-12-29 01:45:33 +00:00
var (
terms dbTermList
kanji dbKanjiList
revisions []string
titles []string
2021-01-01 05:53:10 +00:00
sequence int
2016-12-29 01:45:33 +00:00
)
2016-12-13 21:29:05 +00:00
2016-12-06 05:32:14 +00:00
for _, subbook := range book.Subbooks {
2016-12-13 21:29:05 +00:00
if extractor, ok := epwingExtractors[subbook.Title]; ok {
2016-12-14 02:00:36 +00:00
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 = "<22>"
}
str = strings.Replace(str, matches[0], replacement, -1)
}
pattern := regexp.MustCompile("\n+")
str = pattern.ReplaceAllLiteralString(str, "\n")
2016-12-14 02:00:36 +00:00
return str
}
2016-12-06 05:32:14 +00:00
for _, entry := range subbook.Entries {
2016-12-14 02:00:36 +00:00
entry.Heading = translate(entry.Heading)
entry.Text = translate(entry.Text)
terms = append(terms, extractor.extractTerms(entry, sequence)...)
2016-12-13 21:29:05 +00:00
kanji = append(kanji, extractor.extractKanji(entry)...)
sequence++
2016-11-16 06:21:00 +00:00
}
2016-12-24 05:52:49 +00:00
revisions = append(revisions, extractor.getRevision())
2016-12-29 01:45:33 +00:00
titles = append(titles, subbook.Title)
2016-12-19 04:01:26 +00:00
} else {
return fmt.Errorf("failed to find compatible extractor for '%s'", subbook.Title)
2016-11-16 04:53:21 +00:00
}
}
2016-12-29 01:45:33 +00:00
if title == "" {
title = strings.Join(titles, ", ")
}
2017-09-10 20:25:11 +00:00
recordData := map[string]dbRecordList{
"kanji": kanji.crush(),
2017-09-10 20:45:06 +00:00
"term": terms.crush(),
2017-09-10 20:25:11 +00:00
}
index := dbIndex{
2023-01-29 00:39:08 +00:00
Title: title,
Revision: strings.Join(revisions, ";"),
Sequenced: true,
}
2016-11-16 04:53:21 +00:00
return writeDb(
2017-06-26 00:22:17 +00:00
outputPath,
index,
2017-09-10 20:25:11 +00:00
recordData,
2016-12-29 01:45:33 +00:00
stride,
2016-12-19 01:31:27 +00:00
pretty,
2016-11-16 04:53:21 +00:00
)
}