本文转自:http://www.cnblogs.com/KissKnife/archive/2008/03/23/1118423.html
有一段时间,正则表达式学习很火热很潮流,当时在CSDN一天就能看到好几个正则表达式的帖子,那段时间借助论坛以及Wrox Press出版的《C#字符串和正则表达式参考手册》学习了一些基础的知识,同时也为我在CSDN大概赚了1000分,今天想起来,去找《C#字符串和正则表达式参考手册》时,已经不知所踪了。现在用到正则的时候也比较少,把以前的笔记等整理一下,以志不忘。
(1)“@”符号 符下两ows表研究室的火热,当晨在“@”虽然并非C#正则表达式的“成员”,但是它经常与C#正则表达式出双入对。“@”表示,跟在它后面的字符串是个“逐字字符串”,不是很好理解,举个例子,以下两个声明是等效的: string x="D:\\My Huang\\My Doc"; string y = @"D:\My Huang\My Doc"; 事实上,如果按如下声明,C#将会报错,因为“\”在C#中用于实现转义,如“\n”换行: string x = "D:\My Huang\My Doc"; (2)基本的语法字符。 \d 0-9的数字 \D \d的补集(以所以字符为全集,下同),即所有非数字的字符 \w 单词字符,指大小写字母、0-9的数字、下划线 \W \w的补集 \s 空白字符,包括换行符\n、回车符\r、制表符\t、垂直制表符\v、换页符\f \S \s的补集 . 除换行符\n外的任意字符 […] 匹配[]内所列出的所有字符 [^…] 匹配非[]内所列出的字符 下面提供一些简单的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string i ="\n";
string m ="3";
Regex r =new Regex(@"\D");
//同Regex r = new Regex("\\D");
//r.IsMatch(i)结果:true
//r.IsMatch(m)结果:false
string i ="%";
string m ="3";
Regex r =new Regex("[a-z0-9]");
//匹配小写字母或数字字符
//r.IsMatch(i)结果:false
//r.IsMatch(m)结果:true ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(3)定位字符 “定位字符”所代表的是一个虚的字符,它代表一个位置,你也可以直观地认为“定位字符”所代表的是某个字符与字符间的那个微小间隙。 ^ 表示其后的字符必须位于字符串的开始处 $ 表示其前面的字符必须位于字符串的结束处 \b 匹配一个单词的边界 \B 匹配一个非单词的边界 另外,还包括:\A 前面的字符必须位于字符处的开始处,\z 前面的字符必须位于字符串的结束处,\Z 前面的字符必须位于字符串的结束处,或者位于换行符前 下面提供一些简单的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string i ="Live for nothing,die for something";
Regex r1 =new Regex("^Live for nothing,die for something$");
//r1.IsMatch(i) true
Regex r2 =new Regex("^Live for nothing,die for some$");
//r2.IsMatch(i) false
Regex r3 =new Regex("^Live for nothing,die for some");
//r3.IsMatch(i) true
string i =@"Live for nothing,
die for something";//多行
Regex r1 =new Regex("^Live for nothing,die for something$");
Console.WriteLine("r1 match count:"+ r1.Matches(i).Count);//0
Regex r2 =new Regex("^Live for nothing,die for something$", RegexOptions.Multiline);
Console.WriteLine("r2 match count:"+ r2.Matches(i).Count);//0
Regex r3 =new Regex("^Live for nothing,\r\ndie for something$");
Console.WriteLine("r3 match count:"+ r3.Matches(i).Count);//1
Regex r4 =new Regex("^Live for nothing,$");
Console.WriteLine("r4 match count:"+ r4.Matches(i).Count);//0
Regex r5 =new Regex("^Live for nothing,$", RegexOptions.Multiline);
Console.WriteLine("r5 match count:"+ r5.Matches(i).Count);//0
Regex r6 =new Regex("^Live for nothing,\r\n$");
Console.WriteLine("r6 match count:"+ r6.Matches(i).Count);//0
Regex r7 =new Regex("^Live for nothing,\r\n$", RegexOptions.Multiline);
Console.WriteLine("r7 match count:"+ r7.Matches(i).Count);//0
Regex r8 =new Regex("^Live for nothing,\r$");
Console.WriteLine("r8 match count:"+ r8.Matches(i).Count);//0
Regex r9 =new Regex("^Live for nothing,\r$", RegexOptions.Multiline);
Console.WriteLine("r9 match count:"+ r9.Matches(i).Count);//1
Regex r10 =new Regex("^die for something$");
Console.WriteLine("r10 match count:"+ r10.Matches(i).Count);//0
Regex r11 =new Regex("^die for something$", RegexOptions.Multiline);
Console.WriteLine("r11 match count:"+ r11.Matches(i).Count);//1
Regex r12 =new Regex("^");
Console.WriteLine("r12 match count:"+ r12.Matches(i).Count);//1
Regex r13 =new Regex("$");
Console.WriteLine("r13 match count:"+ r13.Matches(i).Count);//1
Regex r14 =new Regex("^", RegexOptions.Multiline);
Console.WriteLine("r14 match count:"+ r14.Matches(i).Count);//2
Regex r15 =new Regex("$", RegexOptions.Multiline);
Console.WriteLine("r15 match count:"+ r15.Matches(i).Count);//2
Regex r16 =new Regex("^Live for nothing,\r$\n^die for something$", RegexOptions.Multiline);
Console.WriteLine("r16 match count:"+ r16.Matches(i).Count);//1
//对于一个多行字符串,在设置了Multiline选项之后,^和$将出现多次匹配。
string i ="Live for nothing,die for something";
string m ="Live for nothing,die for some thing";
Regex r1 =new Regex(@"\bthing\b");
Console.WriteLine("r1 match count:"+ r1.Matches(i).Count);//0
Regex r2 =new Regex(@"thing\b");
Console.WriteLine("r2 match count:"+ r2.Matches(i).Count);//2
Regex r3 =new Regex(@"\bthing\b");
Console.WriteLine("r3 match count:"+ r3.Matches(m).Count);//1
Regex r4 =new Regex(@"\bfor something\b");
Console.WriteLine("r4 match count:"+ r4.Matches(i).Count);//1
//\b通常用于约束一个完整的单词
![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(4)重复描述字符 “重复描述字符”是体现C#正则表达式“很好很强大”的地方之一: {n} 匹配前面的字符n次 {n,} 匹配前面的字符n次或多于n次 {n,m} 匹配前面的字符n到m次 ? 匹配前面的字符0或1次 + 匹配前面的字符1次或多于1次 * 匹配前面的字符0次或式于0次 以下提供一些简单的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="1024";
string y ="+1024";
string z ="1,024";
string a ="1";
string b="-1024";
string c ="10000";
Regex r =new Regex(@"^\+?[1-9],?\d{3}$");
Console.WriteLine("x match count:"+ r.Matches(x).Count);//1
Console.WriteLine("y match count:"+ r.Matches(y).Count);//1
Console.WriteLine("z match count:"+ r.Matches(z).Count);//1
Console.WriteLine("a match count:"+ r.Matches(a).Count);//0
Console.WriteLine("b match count:"+ r.Matches(b).Count);//0
Console.WriteLine("c match count:"+ r.Matches(c).Count);//0
//匹配1000到9999的整数。
![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(5)择一匹配 C#正则表达式中的 (|) 符号似乎没有一个专门的称谓,姑且称之为“择一匹配”吧。事实上,像[a-z]也是一种择一匹配,只不过它只能匹配单个字符,而(|)则提供了更大的范围,(ab|xy)表示匹配ab或匹配xy。注意“|”与“()”在此是一个整体。下面提供一些简单的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="0";
string y ="0.23";
string z ="100";
string a ="100.01";
string b ="9.9";
string c ="99.9";
string d ="99.";
string e ="00.1";
Regex r =new Regex(@"^\+?((100(.0+)*)|([1-9]?[0-9])(\.\d+)*)$");
Console.WriteLine("x match count:"+ r.Matches(x).Count);//1
Console.WriteLine("y match count:"+ r.Matches(y).Count);//1
Console.WriteLine("z match count:"+ r.Matches(z).Count);//1
Console.WriteLine("a match count:"+ r.Matches(a).Count);//0
Console.WriteLine("b match count:"+ r.Matches(b).Count);//1
Console.WriteLine("c match count:"+ r.Matches(c).Count);//1
Console.WriteLine("d match count:"+ r.Matches(d).Count);//0
Console.WriteLine("e match count:"+ r.Matches(e).Count);//0
//匹配0到100的数。最外层的括号内包含两部分“(100(.0+)*)”,“([1-9]?[0-9])(\.\d+)*”,这两部分是“OR”的关系,即正则表达式引擎会先尝试匹配100,如果失败,则尝试匹配后一个表达式(表示[0,100)范围中的数字)。
![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(6)特殊字符的匹配 下面提供一些简单的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="\\";
Regex r1 =new Regex("^\\\\$");
Console.WriteLine("r1 match count:"+ r1.Matches(x).Count);//1
Regex r2 =new Regex(@"^\\$");
Console.WriteLine("r2 match count:"+ r2.Matches(x).Count);//1
Regex r3 =new Regex("^\\$");
Console.WriteLine("r3 match count:"+ r3.Matches(x).Count);//0
//匹配“\”
string x ="\"";
Regex r1 =new Regex("^\"$");
Console.WriteLine("r1 match count:"+ r1.Matches(x).Count);//1
Regex r2 =new Regex(@"^""$");
Console.WriteLine("r2 match count:"+ r2.Matches(x).Count);//1
//匹配双引号 ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(7)组与非捕获组 以下提供一些简单的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="Live for nothing,die for something";
string y ="Live for nothing,die for somebody";
Regex r =new Regex(@"^Live ([a-z]{3}) no([a-z]{5}),die \1 some\2$");
Console.WriteLine("x match count:"+ r.Matches(x).Count);//1
Console.WriteLine("y match count:"+ r.Matches(y).Count);//0
//正则表达式引擎会记忆“()”中匹配到的内容,作为一个“组”,并且可以通过索引的方式进行引用。表达式中的“\1”,用于反向引用表达式中出现的第一个组,即粗体标识的第一个括号内容,“\2”则依此类推。
string x ="Live for nothing,die for something";
Regex r =new Regex(@"^Live for no([a-z]{5}),die for some\1$");
if (r.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("group1 value:"+ r.Match(x).Groups[1].Value);//输出:thing
}
//获取组中的内容。注意,此处是Groups[1],因为Groups[0]是整个匹配的字符串,即整个变量x的内容。
string x ="Live for nothing,die for something";
Regex r =new Regex(@"^Live for no(?<g1>[a-z]{5}),die for some\1$");
if (r.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("group1 value:"+ r.Match(x).Groups["g1"].Value);//输出:thing
}
//可根据组名进行索引。使用以下格式为标识一个组的名称(?<groupname>…)。
string x ="Live for nothing nothing";
Regex r =new Regex(@"([a-z]+) \1");
if (r.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
x = r.Replace(x, "$1");
Console.WriteLine("var x:"+ x);//输出:Live for nothing
}
//删除原字符串中重复出现的“nothing”。在表达式之外,使用“$1”来引用第一个组,下面则是通过组名来引用:
string x ="Live for nothing nothing";
Regex r =new Regex(@"(?<g1>[a-z]+) \1");
if (r.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
x = r.Replace(x, "${g1}");
Console.WriteLine("var x:"+ x);//输出:Live for nothing
}
string x ="Live for nothing";
Regex r =new Regex(@"^Live for no(?:[a-z]{5})$");
if (r.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("group1 value:"+ r.Match(x).Groups[1].Value);//输出:(空)
}
//在组前加上“?:”表示这是个“非捕获组”,即引擎将不保存该组的内容。 ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(8)贪婪与非贪婪 正则表达式的引擎是贪婪,只要模式允许,它将匹配尽可能多的字符。通过在“重复描述字符”(*,+)后面添加“?”,可以将匹配模式改成非贪婪。请看以下示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="Live for nothing,die for something";
Regex r1 =new Regex(@".*thing");
if (r1.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("match:"+ r1.Match(x).Value);//输出:Live for nothing,die for something
}
Regex r2 =new Regex(@".*?thing");
if (r2.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("match:"+ r2.Match(x).Value);//输出:Live for nothing
} ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(9)回溯与非回溯 使用“(?>…)”方式进行非回溯声明。由于正则表达式引擎的贪婪特性,导致它在某些情况下,将进行回溯以获得匹配,请看下面的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="Live for nothing,die for something";
Regex r1 =new Regex(@".*thing,");
if (r1.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("match:"+ r1.Match(x).Value);//输出:Live for nothing,
}
Regex r2 =new Regex(@"(?>.*)thing,");
if (r2.IsMatch(x))//不匹配 ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("match:"+ r2.Match(x).Value);
}
//在r1中,“.*”由于其贪婪特性,将一直匹配到字符串的最后,随后匹配“thing”,但在匹配“,”时失败,此时引擎将回溯,并在“thing,”处匹配成功。
在r2中,由于强制非回溯,所以整个表达式匹配失败。 ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(10)正向预搜索、反向预搜索 正向预搜索声明格式:正声明 “(?=…)”,负声明 “(?!...)” ,声明本身不作为最终匹配结果的一部分,请看下面的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="1024 used 2048 free";
Regex r1 =new Regex(@"\d{4}(?= used)");
if (r1.Matches(x).Count==1) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("r1 match:"+ r1.Match(x).Value);//输出:1024
}
Regex r2 =new Regex(@"\d{4}(?! used)");
if (r2.Matches(x).Count==1) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("r2 match:"+ r2.Match(x).Value); //输出:2048
}
//r1中的正声明表示必须保证在四位数字的后面必须紧跟着“ used”,r2中的负声明表示四位数字之后不能跟有“ used”。 ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
反向预搜索声明格式:正声明“(?<=)”,负声明“(?<!)”,声明本身不作为最终匹配结果的一部分,请看下面的示例:
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
string x ="used:1024 free:2048";
Regex r1 =new Regex(@"(?<=used:)\d{4}");
if (r1.Matches(x).Count==1) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("r1 match:"+ r1.Match(x).Value);//输出:1024
}
Regex r2 =new Regex(@"(?<!used:)\d{4}");
if (r2.Matches(x).Count==1) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
Console.WriteLine("r2 match:"+ r2.Match(x).Value);//输出:2048
}
//r1中的反向正声明表示在4位数字之前必须紧跟着“used:”,r2中的反向负声明表示在4位数字之前必须紧跟着除“used:”之外的字符串。 ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(11)十六进制字符范围 正则表达式中,可以使用 "\xXX" 和 "\uXXXX" 表示一个字符("X" 表示一个十六进制数)形式字符范围: \xXX 编号在 0到255 范围的字符,比如:空格可以使用 "\x20" 表示。 \uXXXX 任何字符可以使用 "\u" 再加上其编号的4位十六进制数表示,比如:汉字可以使用“[\u4e00-\u9fa5]”表示。
(12)对[0,100]的比较完备的匹配 下面是一个比较综合的示例,对于匹配[0,100],需要特殊考虑的地方包括 *00合法,00.合法,00.00合法,001.100合法 *空字符串不合法,仅小数点不合法,大于100不合法 *数值是可带后缀的,如“1.07f”表示该值为一个float类型(未考虑)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
Code
Regex r =new Regex(@"^\+?0*(?:100(\.0*)?|(\d{0,2}(?=\.\d)|\d{1,2}(?=($|\.$)))(\.\d*)?)$");
string x ="";
while (true) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedBlockStart.gif)
![](https://www.cnblogs.com/Images/OutliningIndicators/ContractedBlock.gif)
{
x = Console.ReadLine();
if (x !="exit") ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedSubBlockStart.gif)
{
if (r.IsMatch(x)) ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedSubBlockStart.gif)
{
Console.WriteLine(x +" succeed!");
}
else ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedSubBlockStart.gif)
{
Console.WriteLine(x +" failed!");
}
}
else ![](https://www.cnblogs.com/Images/OutliningIndicators/ExpandedSubBlockStart.gif)
{
break;
}
} ![](https://www.cnblogs.com/Images/OutliningIndicators/None.gif)
(13)精确匹配有时候是困难的 有些需求要做到精确匹配比较困难,例如:日期、Url、Email地址等,其中一些你甚至需要研究一些专门的文档写出精确完备的表达式,对于这种情况,只能退而求其次,保证比较精确的匹配。例如对于日期,可以基于应用系统的实际情况考虑一段较短的时间,或者对于像Email的匹配,可以只考虑最常见的形式。
【推荐】编程新体验,更懂你的AI,立即体验豆包MarsCode编程助手
【推荐】凌霞软件回馈社区,博客园 & 1Panel & Halo 联合会员上线
【推荐】抖音旗下AI助手豆包,你的智能百科全书,全免费不限次数
【推荐】博客园社区专享云产品让利特惠,阿里云新客6.5折上折
【推荐】轻量又高性能的 SSH 工具 IShell:AI 加持,快人一步