Question about the split function

  • Thread starter Thread starter needin4mation
  • Start date Start date
N

needin4mation

Hi, I am trying to parse a colon delimited text file.

Assume these as examples:

Book:mybook
Video:myvideo
Name:myname


When I use this:

ReadFromFile("c:\\Inetpub\\wwwroot\\myfile.txt");
}
public void ReadFromFile(string filename)
{
StreamReader SR;
string S;
string[] myStr;
SR=File.OpenText(filename);
S=SR.ReadLine();
while(S!=null)
{
myStr = S.Split(new char[] { ':' });
Response.Write(myStr[0] + "<BR>");
S=SR.ReadLine();
}
SR.Close();
}

On my web page I get this:

Book
Video

What I am trying to get is the stuff on the other side of the colon.
How can I get either the data on the other side of the colon or just
the entire line?

Thank you for any help.
 
It is working correctly you are just printing out the first value only.

Split returns an array of strings (myStr[0], myStr[1], etc.) that each
contain a value.

So in your example:
myStr[0] = Book
myStr[1] = mybook

myStr[0] = Video
myStr[1] = myvideo
 
Let me fix your code...
First of all if you are trying to parse a file which is less than 10 MB u
may use File.OpenText otherwise u should use StreamReader

public void ReadFromFile(string pFilename)
{
StreamReader sr = null;
String sLine;
String[] stringData;
try
{
sr = File.OpenText(pFilename);
while (sr.Peek()>=0)
{
sLine= sr.ReadLine();
stringData= sLine.Split(':'); // Only available with 2.0
if (stringData.Length > 1)
Console.WriteLine("{0}\t{1}",stringData);
}
}
finally
{
if (sr != null)
sr.Close();
}
}

Also, u can access "mybook" using stringData[1]
--
HTH

Thanks,
Yunus Emre ALPÖZEN
BSc, MCSD.NET
Microsoft .NET & Security MVP
 
Thanks! Do you know if there is a way to start back over at the
beginning of the file? Sometimes there are error codes in a given text
file. I only want to parse the ones that have errors. So I thought
that if I searched once for "ERROR:" and if it had it I could go back
to the top of the file and start actually getting data. I thought this
might be better than getting data only to find out that I really don't
need to have captured from it in the first place. Doesn't sound like
much, but there could be a thousand text files to search through.
 
Let me fix your code...

;)
stringData= sLine.Split(':'); // Only available with 2.0
Click to expand...

What are you talking about? String.Split(char) has been with us for a long
time now. ;)

Anyway...I like mine better :)

using (StreamReader reader = File.OpenText("myfile.txt"))
{
while(reader.Peek() != -1)
{
string line = reader.ReadLine();

string[] parts = line.Split(':');

Console.WriteLine("{0} : {1}", parts[0], parts[1]);
}
}

Chris
 
If you only want to process lines that contain the string "ERROR:" before
the delimiter, you have no need to go through the file twice.

Simply test for the presence of "ERROR:" in myStr[0] (example):

sLine= sr.ReadLine();
stringData= sLine.Split(':');
if(stringData[0]=="ERROR")
{

// your cool processing code here
}

-- That's all.
Peter
 
As you are creating a file stream you should be able to Seek to the
beginning.
A definition of SR like:

public void ReadFromFile(string filename) {
StreamReader SR;
string S;
string[] myStr;
FileStream fs = new FileStream(filename,FileMode.Open);
SR= new StreamReader(fs);
S=SR.ReadLine();
while(S!=null) {
myStr = S.Split(new char[] { ':' });
Response.Write(myStr[0] + "<BR>");
S=SR.ReadLine();
}
SR.Close();
}

You would get a FileStream to file called fs.
You could then use :
fs.Seek(0,SeekOrigin.Begin);

To jump back to the beginning of the file.

HTH

Ciaran
 
Ciaran said:
As you are creating a file stream you should be able to Seek to the
beginning.
Click to expand...

You would get a FileStream to file called fs.
You could then use :
fs.Seek(0,SeekOrigin.Begin);

To jump back to the beginning of the file.
Click to expand...

Just to mention that a somewhat more readable (IMO) way of doing a Seek
is to use the Position property:

fs.Position = 0;
 
Hi Chris,
I really like your joke. The overload i had used with my code available
since 2.0. Original one was

"stringData = sLine.Split(new char[] { ':' },
StringSplitOptions.RemoveEmptyEntries);"
But then update my code for clarity. Thanks for your notice..

Anyway, using "using" keyword is same as using "try_finally" blocks. I used
to implement my codes using four languages (C#, VB.NET, J#, C++.NET) as
writing articles, writing posts.. And unfortunately using is not available
for J# and C++.NET yet :( Also it is available for VB.NET since 2.0... So I
prefer using "try_finally" blocks force of habit :)

--
HTH

Thanks,
Yunus Emre ALPÖZEN
BSc, MCSD.NET
Microsoft .NET & Security MVP

chris martin said:
Let me fix your code...
;)

stringData= sLine.Split(':'); // Only available with 2.0
Click to expand...

What are you talking about? String.Split(char) has been with us for a long
time now. ;)

Anyway...I like mine better :)

using (StreamReader reader = File.OpenText("myfile.txt"))
{
while(reader.Peek() != -1)
{
string line = reader.ReadLine();

string[] parts = line.Split(':');

Console.WriteLine("{0} : {1}", parts[0], parts[1]);
}
}

Chris
Click to expand...
 
<"Yunus Emre ALPÖZEN [MVP]" <yemre>> wrote:

Anyway, using "using" keyword is same as using "try_finally" blocks. I used
to implement my codes using four languages (C#, VB.NET, J#, C++.NET) as
writing articles, writing posts.. And unfortunately using is not available
for J# and C++.NET yet :( Also it is available for VB.NET since 2.0... SoI
prefer using "try_finally" blocks force of habit :)
Click to expand...

Even if you're writing articles or posts though, I'd use idiomatic
language where possible - and the preferred idiom for disposing of
objects in C# *is* to use the using statement. Might as well use it in
languages where it's available...

(I would personally not use Peek, by the way - keep reading lines until
reader.ReadLine returns null.)
 
Back
Top