用正则表达式解析文本

Question

I need to parse a text and split it very precisly. I chose to use the regular expressions to do the job but I encounter a problem with an advanced manipulation of it in C#. I would appreciate any help to find the perfect solution, even if I need to take something else than regexes.

Here are my criteria :

• The text need to be splited when there is a : ; ! ? \\r
• We can also split it if there are dots "." followed by a white-space
• If there are white-spaces behind a separator, they need to be added.
• If there is an URL we do not split the ":"
• If there suspension dots "...", they need to be added behind

And here is a sample text to understand better :

---Lorem ipsum dolor sit amet, consectetur adipiscing elit. Mauris euismod : tristiquetellus non egestas; Pellentesque fermentum lectus orci ! A dictum nunc placerat sed ? Quisque eget felis in lacus \\rcursus posuere\\r\\r Aliquam venenatis\\r nisi vitae dictum pharetra. ---Vivamus semper dolor quam, pellent.esque hendrerit sapien blandit ut. \\r\\r\\r\\rCras sem massa, tempor sit amet nunc id, condimentum facilisis augue... \\rhttps://www.google.com dictum nunc placerat sed

And finally the result wanted :

 ---Lorem ipsum dolor sit amet, consectetur adipiscing elit. 
 Mauris euismod  : 
 tristiquetellus non egestas; 
 Pellentesque fermentum lectus orci ! 
 A dictum nunc placerat sed ? 
 Quisque eget felis in lacus \r
 cursus posuere\r\r 
 Aliquam venenatis\r
 nisi vitae dictum pharetra.     \r
 ---Vivamus semper dolor quam, pellent.esque hendrerit sapien blandit ut.  \r\r\r\r
 Cras sem massa, tempor sit amet nunc id, condimentum facilisis augue...  \r
 https://www.google.com dictum nunc placerat sed

I am really far away of the result that is why I am posting here. I try at this moment to success the 1) step. Here is my actual code :

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Text.RegularExpressions;
using System.Threading.Tasks;

namespace ConsoleApplication58
{
    class Program
    {
        static void Main(string[] args)
        {            
            Regex r = new Regex(@"(\S*\w+\s*\p{P}*)+[:;!?]+\s*");

            string lorem = "---Lorem ipsum dolor sit amet, consectetur adipiscing elit. " +
                "Mauris euismod : " +
                "tristiquetellus non egestas; " +
                "Pellentesque fermentum lectus orci ! " +
                "A dictum nunc placerat sed ? " +
                "Quisque eget felis in lacus \r" +
                "cursus posuere\r\r " +
                "Aliquam venenatis\r " +
                "nisi vitae dictum pharetra. " +
                "---Vivamus semper dolor quam, pellent.esque hendrerit sapien blandit ut. \r\r\r\r" +
                "Cras sem massa, tempor sit amet nunc id, condimentum facilisis augue... \r" +
                "https://www.google.com dictum nunc placerat sed";

            MatchCollection m2 = r.Matches(lorem);

            foreach (Match match in m2)
            {
                string txt = match.Value;
                Console.WriteLine("*{0}*", txt);
            }
        }
    }
}

Thank you very much for reading this and trying to help me. This is kinda urgent and I can not figure out the good combination with the Matches() method from the regexes. Do not hesitate to ask me for more details if necessary.

Answer 1

Since you still haven't been really clear about whether \\r is supposed to be a carriage return or a literal \\r , I'll put both:

Literal:

(.+?)((?:\.{3} |[:;!?](?!/)|\. )(?:\\r)*\s*|(?:\\r)+\s*|$)

ideone demo .

Carriage return:

(.+?)((?:\.{3} |[:;!?](?!/)|\. )(?:\r)*\s*|(?:\r)+\s*|$)

ideone demo .

Answer 2

I see you have the Regex, just split the string using the Regex instance like this: ... string[] splitStringValues = r.split(lorem); or

char u = ':'; //just initializing

        switch (u) 
        {
            case ':':
                //do split work here
                break;
            default:
                //do split work here
                break;
        }

Answer 3

签出(((http(s*))\\://){1}\\S+)|((\\S*\\w+\\s*\\p{P}*)+[:;!?]+\\s*)|(\\...)