asp.net 安全的截取指定长度的html或者ubb字符串

在截取字符串时需要记录每一个标签是否关闭，如果截取到指定长度还有没有关闭的标签，那么我们需要将标签关闭，或者删除掉未关闭的标签。不考虑某些不需要关闭标签的情况，html开始和结束标签总是成对出现的，我们可以遍历输入的字符串，并在标签开始时放入堆栈中，遇到结束标签时从堆栈中弹出一个元素，这样遍历到指定长度，堆栈中留下的标签就是需要补全或者删除掉的标签。

下面是代码实现，如果大家有更好的方法请给出来：

static char end_slash = '/';

/// <summary>
/// 安全的截断字符串
/// </summary>
/// <param name="input">输入串</param>
/// <param name="length">截断长度</param>
/// <param name="trimhalftag">true：截断半截标签；false：补全半截标签</param>
/// <param name="tagstartchar">标签开始字符</param>
/// <param name="tagendchar">标签结束字符</param>
/// <param name="mustclosetags">需要关闭的标签数组</param>
/// <returns>length长度的字符串</returns>
public static string safetrim(string input, int length, bool trimhalftag, char tagstartchar, char tagendchar, string[] mustclosetags)
{
if (length <= 0) throw new argumentexception("length 必须是正数");
if (mustclosetags == null) throw new argumentnullexception("mustclosetags");

int inputlen = input.length;
if (string.isnullorempty(input) || inputlen <= length) return input;

string result = string.empty;

//声明堆栈用来放标签
stack<string> tags = new stack<string>();

for (int i = 0; i < length; i++)
{
char c = input[i];

if (c == tagstartchar)
{
string tag = string.empty;
int tagindex = i + 1;
bool istagend = false;
bool istagnameend = false;
result += c;
bool hasmarktaginstack = false;
while (tagindex < inputlen)
{
char tagc = input[tagindex];
result += tagc;
tagindex++;
if (tag == string.empty && tagc == end_slash)
{
istagend = true;
continue;
}
if (!istagnameend)
{
if (char.isletter(tagc) || char.isnumber(tagc))
{
tag += tagc;
}
else
{
istagnameend = true;
}
}

if (!string.isnullorempty(tag))
{
if (istagnameend && !hasmarktaginstack)
{
if (istagend)
{
tags.pop();
}
else
{
tags.push(tag);
}
hasmarktaginstack = true;
}
}

if (istagnameend)
{
if (tagc == tagendchar)
{
i = tagindex - 1;
break;
}
}

}
}
else
{
result += c;
}
}

while (tags.count > 0)
{
string tag = tags.pop();

bool ismustclosetag = false;
foreach (string mustclosetag in mustclosetags)
{
if (string.compare(mustclosetag, tag, true) == 0)
{
ismustclosetag = true;
break;
}
}
if (ismustclosetag)
{
if (trimhalftag)
{
int lasttagindex = result.lastindexof(tagstartchar.tostring() + tag, stringcomparison.currentcultureignorecase);

result = result.substring(0, lasttagindex);
}
else
{
result += (tagstartchar.tostring() + end_slash + tag + tagendchar);
}
}
}

return result;
}

转载请保留链接