最大匹配算法： 分词入门级算法 C#实现

StreamReader sr =
new StreamReader(@wordLibPath,
System.Text.Encoding.GetEncoding("gb2312"))
;

private string skipBlanks( string strInput )
{
char[] strArray = strInput.ToCharArray();
int wordLen = strInput.Length;
strInput = "";

for (int i = 0; i < wordLen; i++)
{
if (strArray[i] == ' ' || strArray[i] == '\n'
|| strArray[i] == '\t')
continue;
else
strInput += strArray[i];
}

return strInput;
}

private string maxBackfowardPatch(string strInput)
{
strInput = skipBlanks(strInput);

string strWord = "";
string strOutput = "";
int leftIdx = strInput.Length;
char[] strArray = strInput.ToCharArray();
int wordLen = leftIdx;
List<int> segPos = new List<int>(); //分段点
int subLen = 0;

/*最大后向匹配算法*/
{
while (!strInput.Equals(""))
{
bool flag = true;

while (leftIdx > 0 && subLen < maxLen)
{
strWord = strInput.Substring(--leftIdx, ++subLen);

if (wordLib.Contains(strWord))
{
segPos.Add(leftIdx);
segPos.Add(leftIdx + subLen);
flag = false;
break;
}
}
if (flag)
leftIdx = strInput.Length - 1;
strInput = strInput.Remove(leftIdx);
leftIdx = strInput.Length;
subLen = 0;
}
}

//输出窜
for (int i = 0; i < wordLen; i++) {
if (segPos.Contains(i)) strOutput += '/';
strOutput += strArray[i];
}
segPos.Clear();

return strOutput;
}