如何在一串文本中提取短语然后提取单词?

时间:2009-09-09 15:35:35

标签: c# regex search phrase

我有一个搜索方法,它接收用户输入的字符串,在每个空格字符处拆分它,然后根据分隔的字词列表继续查找匹配项:

string[] terms = searchTerms.ToLower().Trim().Split( ' ' );

现在我已经获得了进一步的要求:能够通过双引号分隔符来搜索短语。因此,如果提供的搜索字词是:

  

“一行”文字

搜索将匹配“一行”和“文本”的出现,而不是四个单独的术语[在搜索之前还需要删除打开和关闭的双引号]。

如何在C#中实现这一目标?我会假设正则表达式是要走的路,但是没有涉足它们,所以不知道它们是否是最好的解决方案。

如果您需要更多信息,请询问。在此先感谢您的帮助。

6 个答案:

答案 0 :(得分:2)

这是一个正则表达式模式,它将返回名为“term”的组中的匹配项:

("(?<term>[^"]+)"\s*|(?<term>[^ ]+)\s*)+

所以对于输入:

"a line" of text

term”群组标识的输出项目为:

a line
of
text

答案 1 :(得分:1)

正则表达式肯定是要走的路......

您应该检查此MSDN链接以获取有关Regex类的一些信息: http://msdn.microsoft.com/en-us/library/system.text.regularexpressions.regex.aspx

这是学习一些正则表达式语法的优秀链接: http://www.radsoftware.com.au/articles/regexlearnsyntax.aspx

然后,为了添加一些代码示例,您可以按照以下方式进行操作:

string searchString = "a line of";

Match m = Regex.Match(textToSearch, searchString);

或者如果您只是想知道字符串是否包含匹配项:

bool success = Regex.Match(textToSearch, searchString).Success;

答案 2 :(得分:1)

使用此处的正则表达式构建器

http://gskinner.com/RegExr/

并且您将能够将正则表达式操作为您需要它的显示方式

答案 3 :(得分:1)

使用Regexs ....

string textToSearchIn =“”一行“text”;
string result = Regex.Match(textToSearchIn,“(?&lt; =”)。*?(?=“)”)。值;

或者如果多于一个,请将其放入匹配集合中......

MatchCollection allPhrases = Regex.Matches(textToSearchIn,“(?&lt; =”)。*?(?=“)”);

答案 4 :(得分:0)

Knuth-Morris-Pratt(KMP算法)被认为是在字符串中查找子串的最快算法(好吧,技术上不是字符串而是字节数组)。

using System.Collections.Generic;

namespace KMPSearch
{
    public class KMPSearch
    {
        public static int NORESULT = -1;

        private string _needle;
        private string _haystack;
        private int[] _jumpTable;

        public KMPSearch(string haystack, string needle)
        {
            Haystack = haystack;
            Needle = needle;
        }

        public void ComputeJumpTable()
        {
            //Fix if we are looking for just one character...
            if (Needle.Length == 1)
            {
                JumpTable = new int[1] { -1 };
            }
            else
            {
                int needleLength = Needle.Length;
                int i = 2;
                int k = 0;

                JumpTable = new int[needleLength];
                JumpTable[0] = -1;
                JumpTable[1] = 0;

                while (i <= needleLength)
                {
                    if (i == needleLength)
                    {
                        JumpTable[needleLength - 1] = k;
                    }
                    else if (Needle[k] == Needle[i])
                    {
                        k++;
                        JumpTable[i] = k;
                    }
                    else if (k > 0)
                    {
                        JumpTable[i - 1] = k;
                        k = 0;
                    }

                    i++;
                }
            }
        }

        public int[] MatchAll()
        {
            List<int> matches = new List<int>();
            int offset = 0;
            int needleLength = Needle.Length;
            int m = Match(offset);

            while (m != NORESULT)
            {
                matches.Add(m);
                offset = m + needleLength;
                m = Match(offset);
            }

            return matches.ToArray();
        }

        public int Match()
        {
            return Match(0);
        }

        public int Match(int offset)
        {
            ComputeJumpTable();

            int haystackLength = Haystack.Length;
            int needleLength = Needle.Length;

            if ((offset >= haystackLength) || (needleLength > ( haystackLength - offset))) 
                return NORESULT;

            int haystackIndex = offset;
            int needleIndex = 0;

            while (haystackIndex < haystackLength)
            {
                if (needleIndex >= needleLength)
                    return haystackIndex;

                if (haystackIndex + needleIndex >= haystackLength)
                    return NORESULT;

                if (Haystack[haystackIndex + needleIndex] == Needle[needleIndex])
                {
                    needleIndex++;
                } 
                    else
                {
                    //Naive solution
                    haystackIndex += needleIndex;

                    //Go back
                    if (needleIndex > 1)
                    {
                        //Index of the last matching character is needleIndex - 1!
                        haystackIndex -= JumpTable[needleIndex - 1];
                        needleIndex = JumpTable[needleIndex - 1];
                    }
                    else
                        haystackIndex -= JumpTable[needleIndex];


                }
            }

            return NORESULT;
        }

        public string Needle
        {
            get { return _needle; }
            set { _needle = value; }
        }

        public string Haystack
        {
            get { return _haystack; }
            set { _haystack = value; }
        }

        public int[] JumpTable
        {
            get { return _jumpTable; }
            set { _jumpTable = value; }
        }
    }
}

用法: -

using System;
using System.Collections.Generic;
using System.Text;
using System.Reflection;
namespace KMPSearch
{
    class Program
    {
        static void Main(string[] args)
        {
            if (args.Length != 2)
            {
                Console.WriteLine("Usage: " + Environment.GetCommandLineArgs()[0] + " haystack needle");
            }
            else
            {
                KMPSearch search = new KMPSearch(args[0], args[1]);
                int[] matches = search.MatchAll();
                foreach (int i in matches)
                    Console.WriteLine("Match found at position " + i+1);
            }
        }

    }
}

答案 5 :(得分:0)

试试这个,它会返回一个文本数组。例如:{“一行”文字“记事本”}:

string textToSearch = "\"a line of\" text \" notepad\"";

MatchCollection allPhrases = Regex.Matches(textToSearch, "(?<=\").*?(?=\")");

var RegArray = allPhrases.Cast<Match>().ToArray();

输出:{“一行”,“文字”,“记事本”}