我正在使用string.Replace替換子字符串Afunc removeIP(text string) string { text = strings.Replace(text, "someWord", "**NewWord**", -1) return text}func removeIPUsingRegex(text string) string { var re = regexp.MustCompile(`\b` + "someWord" + `\b`) // I want to replace whole word only text = re.ReplaceAllString(text, "**NewWord**")}我在這里面臨的問題是，只有在字符串替換不支持的情況下，我才想替換整個(gè)單詞。因?yàn)槲冶仨毺鎿Q非常非常大的字符串，所以可能以 GB 為單位。與字符串替換相比，正則表達(dá)式非常慢。eg: text: "abcdef defgh /def/ .def/ =def= def xxxy"-> Replace defwith DEFoutput: "abcdef defgh /DEF/ .DEF/ =DEF= DEF xxxy"//注意只有整個(gè)單詞被替換了。正則表達(dá)式將時(shí)間縮短了近 100 倍（https://medium.com/codezillas/golang-replace-vs-regexp-de4e48482f53）。任何想法將不勝感激。
1 回答
鴻蒙傳說
TA貢獻(xiàn)1865條經(jīng)驗(yàn) 獲得超7個(gè)贊
使用的 KMP 算法
// ReplaceWholeWord ...
func ReplaceWholeWord(text string, oldWord string, newWord string) string {
var patternLength = len(oldWord)
var textLength = len(text)
var copyIndex = 0
var textIndex = 0
var patternIndex = 0
var newString strings.Builder
var lps = computeLPSArray(oldWord)
for textIndex < textLength {
if oldWord[patternIndex] == text[textIndex] {
patternIndex++
textIndex++
}
if patternIndex == patternLength {
startIndex := textIndex - patternIndex
endIndex := textIndex - patternIndex + patternLength - 1
if checkIfWholeWord(text, startIndex, endIndex) {
if copyIndex != startIndex {
newString.WriteString(text[copyIndex:startIndex])
}
newString.WriteString(newWord)
copyIndex = endIndex + 1
}
patternIndex = 0
textIndex = endIndex + 1
} else if textIndex < textLength && oldWord[patternIndex] != text[textIndex] {
if patternIndex != 0 {
patternIndex = lps[patternIndex-1]
} else {
textIndex = textIndex + 1
}
}
}
newString.WriteString(text[copyIndex:])
return newString.String()
}
func computeLPSArray(pattern string) []int {
var length = 0
var i = 1
var patternLength = len(pattern)
var lps = make([]int, patternLength)
lps[0] = 0
for i = 1; i < patternLength; {
if pattern[i] == pattern[length] {
length++
lps[i] = length
i++
} else {
if length != 0 {
length = lps[length-1]
} else {
lps[i] = length
i++
}
}
}
return lps
}
func checkIfWholeWord(text string, startIndex int, endIndex int) bool {
startIndex = startIndex - 1
endIndex = endIndex + 1
if (startIndex < 0 && endIndex >= len(text)) ||
(startIndex < 0 && endIndex < len(text) && isNonWord(text[endIndex])) ||
(startIndex >= 0 && endIndex >= len(text) && isNonWord(text[startIndex])) ||
(startIndex >= 0 && endIndex < len(text) && isNonWord(text[startIndex]) && isNonWord(text[endIndex])) {
return true
}
return false
}
func isNonWord(c byte) bool {
return !((c >= '0' && c <= '9') || (c >= 'a' && c <= 'z') || (c >= 'A' && c <= 'Z') || (c == '_'))
}
- 1 回答
- 0 關(guān)注
- 99 瀏覽
添加回答
舉報(bào)
0/150
提交
取消