CODEWITHSUNDEEP
c#regexstringreplacesplit
Alika87
Alika87

Reputation: 311

C#: Deleting all characters between two Delimiters

got a little Stringcutting problem. I got a string like this:

@VAR;Variable=Deteministic;Value=mostly;Note=Unless Slave is already in use;Op==;@ENDVAR;

the Note is not neccessary for me, so I want to cut out everything that begins with Note until the next Semicolon. The Replace-Method would be nice, but I dont know how to get the Chars after Note away.

I tried something like this:

 int index1 = rs.IndexOf("Note=");
 int index2 = rs.IndexOf(';', index1+1);
 rs = rs.Remove(index1, index2);

I thought this should be sufficient, but it fails if there is no note and my program pops out an error. Regex would be an option, but I cant think of one the fits in here.

Please note, the example was just a sample. I dont know how the inputstring looks like. One line can contain two Notes and another line can have none. Please consider this in your answers.

Upvotes: 1

Views: 1087

Answers (6)

Ovidiu
Ovidiu

Reputation: 1407

Your code should look like this

int index1 = rs.IndexOf(";Note=");
while(index1 > -1)
{
    int index2 = rs.IndexOf(';', index1 + 1);
    rs = rs.Remove(index1, index2);
    index1 = rs.IndexOf(";Note=");
}

Upvotes: 0

Isak Savo
Isak Savo

Reputation: 35944

Most of the other regex answers to this post will indeed work, but it might be appropriate with a non-regex answer just to show that there are other tools available than the hammer :)

I'm assuming you just want to ignore all entries starting with Note=

string input = "@VAR;Variable=Deteministic;Value=mostly;Note=Unless Slave is already in use;Op==;@ENDVAR;";

// Entries will contain all entries except those starting with 'Note='
string[] entries = input.Split(';').Where(s => !s.StartsWith("Note=")).ToArray();
// If you want to, you can put it all back together without the Note entries
string output = String.Join(";", entries);

// Ouput: @VAR;Variable=Deteministic;Value=mostly;Op==;@ENDVAR;
Console.WriteLine(output);

// Output:
// @VAR
// Variable=Deteministic
// Value=mostly
// Op==
// @ENDVAR
foreach (var entry in entries)
    Console.WriteLine(entry);

Upvotes: 2

Heinzi
Heinzi

Reputation: 172468

A regex can indeed help here:

rs = Regex.Replace(s, "(?<=;)Note=.*?;", "");

Let me explain the more obscure parts of it:

  • (?<=;) makes sure Note is preceded by a semicolon. That semicolon is, however, not part of the replacement. (That's a positive look-behind assertion).

  • .*?; matches all characters until the semicolon, but non-greedy. This ensures that Note=A;x=B; is only matched until the first semicolon and x=B is retained.

Upvotes: 7

Jerry
Jerry

Reputation: 71598

You can use the regex: 

Note[^;]+;?

Something a bit like:

rs = Regex.Replace(rs, @"\bNote[^;]+;?", "");

\b matches a word boundary.

[^;] matches any character which is not a semicolon.

Upvotes: 1

Alireza
Alireza

Reputation: 10486

First use StringBuilder. It is much more efficient that string. Then if you want to preserve you original idea: 

 int index1 = -1;

while( (index1 = sb.IndexOf("Note=")) >=0)
{
   int index2 = rs.IndexOf(';', index1+1);
   rs = rs.Remove(index1, index2);
}

Upvotes: 1

Sriram Sakthivel
Sriram Sakthivel

Reputation: 73502

You mean this ?

var regex = new Regex(@"Note.*?;");      
var ouput = regex.Replace(input, "");

Upvotes: 1

Related Questions