317 lines
8.5 KiB
Go
317 lines
8.5 KiB
Go
package rime
|
||
|
||
import (
|
||
"bufio"
|
||
"fmt"
|
||
mapset "github.com/deckarep/golang-set/v2"
|
||
"log"
|
||
"os"
|
||
"path"
|
||
"strconv"
|
||
"strings"
|
||
"sync"
|
||
"time"
|
||
"unicode"
|
||
"unicode/utf8"
|
||
)
|
||
|
||
var (
|
||
specialWords = mapset.NewSet[string]() // 特殊词汇列表，不进行任何检查
|
||
|
||
polyphoneWords = mapset.NewSet[string]() // 需要注音的字词
|
||
|
||
wrongWords = mapset.NewSet[string]() // 异形词、错别字
|
||
wrongWordsFilter = mapset.NewSet[string]() // 过滤一部分，包含错别字但不是错别字，如「作爱」是错的，但「叫作爱」是对的
|
||
|
||
hanPinyin = make(map[string][]string) // 汉字拼音映射，用于检查注音是否正确
|
||
hanPinyinFilter = mapset.NewSet[string]() // 过滤一部分，比如「深厉浅揭qi」只在这个词中念qi，并不是错误。
|
||
)
|
||
|
||
// 初始化特殊词汇列表、需要注音列表、错别字列表、拼音列表
|
||
func init() {
|
||
// 特殊词汇列表，不进行任何检查
|
||
specialWords.Add("狄尔斯–阿尔德反应")
|
||
specialWords.Add("特里斯坦–达库尼亚")
|
||
specialWords.Add("特里斯坦–达库尼亚群岛")
|
||
specialWords.Add("茱莉亚·路易斯-德瑞弗斯")
|
||
specialWords.Add("科科斯（基林）群岛")
|
||
specialWords.Add("刚果（金）")
|
||
specialWords.Add("刚果（布）")
|
||
specialWords.Add("赛博朋克：边缘行者")
|
||
specialWords.Add("赛博朋克：边缘跑手")
|
||
specialWords.Add("赛博朋克：命运之轮")
|
||
specialWords.Add("哈勃–勒梅特定律")
|
||
|
||
// 需要注音的列表
|
||
file1, err := os.Open(需要注音TXT)
|
||
if err != nil {
|
||
log.Fatalln(err)
|
||
}
|
||
defer file1.Close()
|
||
sc := bufio.NewScanner(file1)
|
||
for sc.Scan() {
|
||
line := sc.Text()
|
||
if strings.HasPrefix(line, "#") || line == "" {
|
||
continue
|
||
}
|
||
polyphoneWords.Add(line)
|
||
}
|
||
|
||
// 错别字的两个列表： wrongWords wrongWordsFilter
|
||
file2, err := os.Open(错别字TXT)
|
||
if err != nil {
|
||
log.Fatalln(err)
|
||
}
|
||
defer file2.Close()
|
||
sc = bufio.NewScanner(file2)
|
||
isMark := false
|
||
for sc.Scan() {
|
||
line := sc.Text()
|
||
if strings.HasPrefix(line, "# -_-") {
|
||
isMark = true
|
||
continue
|
||
}
|
||
if strings.HasPrefix(line, "#") || line == "" {
|
||
continue
|
||
}
|
||
if !isMark {
|
||
wrongWords.Add(line)
|
||
} else {
|
||
wrongWordsFilter.Add(line)
|
||
}
|
||
}
|
||
|
||
// 汉字拼音映射 hanPinyin hanPinyinFilter
|
||
// 将所有读音读入 hanPinyin
|
||
file3, err := os.Open(HanziPath)
|
||
if err != nil {
|
||
log.Fatalln(err)
|
||
}
|
||
defer file3.Close()
|
||
isMark = false
|
||
sc = bufio.NewScanner(file3)
|
||
for sc.Scan() {
|
||
line := sc.Text()
|
||
if !isMark {
|
||
if strings.Contains(line, mark) {
|
||
isMark = true
|
||
}
|
||
continue
|
||
}
|
||
parts := strings.Split(line, "\t")
|
||
text, code := parts[0], parts[1]
|
||
hanPinyin[text] = append(hanPinyin[text], code)
|
||
}
|
||
// 给 hanPinyin 补充不再字表的读音，和过滤列表 hanPinyinFilter
|
||
file4, err := os.Open(汉字拼音映射TXT)
|
||
if err != nil {
|
||
log.Fatalln(err)
|
||
}
|
||
defer file4.Close()
|
||
sc = bufio.NewScanner(file4)
|
||
isMark = false
|
||
for sc.Scan() {
|
||
line := sc.Text()
|
||
if strings.HasPrefix(line, "#") && !strings.HasPrefix(line, "# -_-") || line == "" {
|
||
continue
|
||
}
|
||
if strings.HasPrefix(line, "# -_-") {
|
||
isMark = true
|
||
continue
|
||
}
|
||
if !isMark {
|
||
parts := strings.Split(line, " ")
|
||
key := parts[0]
|
||
values := parts[1:]
|
||
hanPinyin[key] = append(hanPinyin[key], values...)
|
||
} else {
|
||
hanPinyinFilter.Add(line)
|
||
}
|
||
}
|
||
}
|
||
|
||
// Check 对传入的词库文件进行检查
|
||
// dictPath: 词库文件路径
|
||
// _type: 词库类型 1 只有汉字 2 汉字+注音 3 汉字+注音+权重 4 汉字+权重
|
||
func Check(dictPath string, _type int) {
|
||
// 控制台输出
|
||
defer printlnTimeCost("检查 "+path.Base(dictPath), time.Now())
|
||
|
||
// 打开文件
|
||
file, err := os.Open(dictPath)
|
||
if err != nil {
|
||
log.Fatalln(err)
|
||
}
|
||
defer file.Close()
|
||
|
||
// 开始检查！
|
||
lineNumber := 0
|
||
isMark := false
|
||
sc := bufio.NewScanner(file)
|
||
var wg sync.WaitGroup
|
||
for sc.Scan() {
|
||
lineNumber++
|
||
line := sc.Text()
|
||
if !isMark {
|
||
if strings.HasPrefix(line, mark) {
|
||
isMark = true
|
||
}
|
||
continue
|
||
}
|
||
wg.Add(1)
|
||
go func(line string, lineNumber int) {
|
||
defer wg.Done()
|
||
checkLine(dictPath, _type, line, lineNumber)
|
||
}(line, lineNumber)
|
||
}
|
||
|
||
wg.Wait()
|
||
|
||
if err := sc.Err(); err != nil {
|
||
log.Fatalln(err)
|
||
}
|
||
}
|
||
|
||
// 检查一行
|
||
func checkLine(dictPath string, _type int, line string, lineNumber int) {
|
||
// 忽略注释，base 中有很多被注释了的词汇，暂时没有删除
|
||
if strings.HasPrefix(line, "#") {
|
||
// 注释以 '#' 开头，但不是以 '# '开头（强迫症晚期）
|
||
if !strings.HasPrefix(line, "# ") {
|
||
fmt.Println("has # but not #␣", line)
|
||
}
|
||
return
|
||
}
|
||
|
||
// 有空行
|
||
if strings.TrimSpace(line) == "" {
|
||
fmt.Println("empty line", line)
|
||
}
|
||
|
||
// +---------------------------------------------------------------
|
||
// | 开始检查分割后的内容，分割为： 词汇text 编码code 权重weight
|
||
// +---------------------------------------------------------------
|
||
parts := strings.Split(line, "\t")
|
||
var text, code, weight string
|
||
switch {
|
||
case _type == 1 && len(parts) == 1: // 一列，【汉字】
|
||
text = parts[0]
|
||
case _type == 2 && len(parts) == 2: // 两列，【汉字+注音】
|
||
text, code = parts[0], parts[1]
|
||
case _type == 3 && len(parts) == 3: // 三列，【汉字+注音+权重】
|
||
text, code, weight = parts[0], parts[1], parts[2]
|
||
case _type == 4 && len(parts) == 2: // 两列，【汉字+权重】
|
||
text, weight = parts[0], parts[1]
|
||
default:
|
||
log.Fatalln("❌ 分割错误：", lineNumber, line)
|
||
}
|
||
|
||
// weight 应该是纯数字
|
||
if weight != "" {
|
||
_, err := strconv.Atoi(weight)
|
||
if err != nil {
|
||
fmt.Println("❌ weight 非数字：", line)
|
||
}
|
||
}
|
||
|
||
// text 和 weight 不应该含有空格
|
||
if strings.Contains(text, " ") || strings.Contains(weight, " ") {
|
||
fmt.Println("❌ text 和 weight 含有空格：", line)
|
||
}
|
||
|
||
// code 前后不应该有空格
|
||
if strings.HasPrefix(code, " ") || strings.HasSuffix(code, " ") {
|
||
fmt.Println("❌ code 前后有空格：", line)
|
||
}
|
||
|
||
// code 不应该有非小写字母
|
||
for _, r := range code {
|
||
if string(r) != " " && !unicode.IsLower(r) {
|
||
fmt.Println("❌ code 含有非小写字母：", line)
|
||
break
|
||
}
|
||
}
|
||
|
||
// 过滤特殊词条
|
||
if specialWords.Contains(text) {
|
||
return
|
||
}
|
||
|
||
// text 不应该有非汉字内容，除了间隔号 ·
|
||
for _, c := range text {
|
||
if string(c) != "·" && !unicode.Is(unicode.Han, c) {
|
||
fmt.Println("❌ text 含有非汉字内容：", line)
|
||
break
|
||
}
|
||
}
|
||
|
||
// 除了字表，其他词库不应该含有单个的汉字
|
||
if dictPath != HanziPath && utf8.RuneCountInString(text) == 1 {
|
||
fmt.Println("❌ 意外的单个汉字：", line)
|
||
}
|
||
|
||
// 除了 base，其他词库不应该含有两个字的词汇
|
||
if dictPath != BasePath && utf8.RuneCountInString(text) == 2 {
|
||
fmt.Println("❌ 意外的两字词：", line)
|
||
}
|
||
|
||
// 汉字个数应该和拼音个数相等
|
||
if code != "" {
|
||
codeCount := len(strings.Split(code, " "))
|
||
textCount := utf8.RuneCountInString(text)
|
||
if strings.Contains(text, "·") {
|
||
textCount -= strings.Count(text, "·")
|
||
}
|
||
if strings.HasPrefix(text, "# ") {
|
||
textCount -= 2
|
||
}
|
||
if textCount != codeCount {
|
||
fmt.Println("❌ 汉字个数 != 拼音个数：", line)
|
||
}
|
||
}
|
||
|
||
// +---------------------------------------------------------------
|
||
// | 其他检查
|
||
// +---------------------------------------------------------------
|
||
|
||
// 需要注音但没有注音的字
|
||
if dictPath == TencentPath {
|
||
for _, word := range polyphoneWords.ToSlice() {
|
||
if strings.Contains(text, word) {
|
||
fmt.Println("❌ 需要注音：", line)
|
||
}
|
||
}
|
||
}
|
||
|
||
// 检查拼写错误，如「赞zan」写成了zna；顺便检查是否存在字表中没有注音的字
|
||
if dictPath != HanziPath && (_type == 2 || _type == 3) && !hanPinyinFilter.Contains(text) {
|
||
// 把汉字和拼音弄成一一对应关系，「拼音:pin yin」→「拼:pin」「音:yin」
|
||
textWithoutDian := strings.ReplaceAll(text, "·", "") // 去掉间隔号
|
||
pinyins := strings.Split(code, " ")
|
||
i := 0
|
||
for _, zi := range textWithoutDian {
|
||
if !contains(hanPinyin[string(zi)], pinyins[i]) {
|
||
fmt.Printf("❌ 注音错误 or 字表未包含的汉字及注音: %s - %s.+%s\n", line, string(zi), pinyins[i])
|
||
}
|
||
i++
|
||
}
|
||
}
|
||
|
||
// 错别字检查，最耗时的检查
|
||
if dictPath != HanziPath && !wrongWordsFilter.Contains(text) {
|
||
wrongWords.Each(func(wrongWord string) bool {
|
||
if strings.HasPrefix(wrongWord, "=") {
|
||
wrongWord = strings.TrimLeft(wrongWord, "= ")
|
||
if text == wrongWord {
|
||
fmt.Printf("❌ 错别字: %s = %s\n", text, wrongWord)
|
||
return true
|
||
}
|
||
} else if strings.Contains(text, wrongWord) {
|
||
fmt.Printf("❌ 错别字: %s - %s\n", text, wrongWord)
|
||
return true
|
||
}
|
||
return false
|
||
})
|
||
}
|
||
}
|