substitution 1

[altaratz]

I know this is so simple, but I'm still a real newbie at Perl. I'm trying to search through a space delimited text file, and replace the spaces with comma delimits. The lines of data is as such:

AAA BBB CCC DDD EEE FFF
XXX YYY ZZZ RRR GGG YYY

and I want

AAA,BBB,CCC,DDD,EEE,FFF
XXX,YYY,ZZZ,RRR,GGG,YYY


so some elements are separated by 3 spaces, some by 2 spaces, and some by 1 space.

My code looks like this


$counter = 1;

#############open test file#####################

open (TESTFILE, ">>test.txt"

@element=<TESTFILE>;

$lengtharray = @element;

if ($counter<$lengtharray) {

$element[$counter]=$a;

$a=~s/ /,/;
$a=~s/ /,/;
$a=~s/ /,/;



$counter++;



}

close (TESTFILE);


nothing is happending - can someone kindly help. Thanks!

Ethan

 

[luckydexte]

altaraz,

Try this:

$element[$cnt] =~ s/\s+/,/g;

This will replace any whitespace with a ','. The + sign will match all of the whitespace and the g substitutes globally. If you have a comma on the end of each line you could use the chop command to take off the comma.

Brandon

 

[Rieekan]

You could also try the following line of code. It's not as fast as luckydexte's, but it's useful with other things (like removing spaces from filenames and such)

$a = join ',', (split / /, $element[$cnt]);

 

[goBoating]

some of your problems are structural and some are in your syntax.
- Using >> in your open statement opens the file for appending, not reading
- Your 'if' needs to be a 'while'. As it is, it will only run once.
I think you want to run once for each line.
- you can do the space replacement with on regex.

$counter = 1;
open (TESTFILE, &quot;[red]>>[/red]test.txt&quot;
@element=<TESTFILE>;
$lengtharray = @element;
[red]if[/red] ($counter<$lengtharray) {
$element[$counter]=$a;
$a=~s/ /,/;
$a=~s/ /,/;
$a=~s/ /,/;
$counter++;
}
close (TESTFILE);

a slightly different approach, do the entire file in one shot

# open test.txt for reading
open(TESTFILE, &quot;<test.txt&quot;) or die &quot;open test.txt, $!\n&quot;;
while (<TESTFILE>) { $buffer .= $_; }
close TESTFILE;

# replace one or more spaces with a comma
$buffer =~ s/ +/,/gs;

# open test.txt for writing
open(OPF,&quot;>text.txt&quot;) or die &quot;open test.txt, $!\n&quot;;
print OPF &quot;$buffer&quot;;
close OPF;

HTH If you are new to Tek-Tips, please use descriptive titles, check the FAQs,
and beware the evil typo.

 

[esserc]

I like doing things in a more straight forward way. Like this:

#!/usr/local/bin/perl
open(IN, &quot;input.dat&quot or die &quot;Cannot open datafile. $!&quot;;
open(OUT, &quot;>output.dat&quot or die &quot;Cannot open output file. $!&quot;;

while (<IN>) { # Read a line from the input file
chomp; # Remove CR/LF
s/^\s+//; # Remove any whitespace at the beginning
s/\s+$//; # Remove ant whitespace at the end
s/\s+/,/g; # Replace space(s) with a comma
print OUT &quot;$_\n&quot;; # Print the line to the output file
}
close(IN);
close(OUT);

 

[hmerrill]

I have to say I agree with esserc. In my book, straightforward is the best way, even if it makes for longer code - it's the easiest to understand and maintain.
Hardy Merrill
Mission Critical Linux, Inc.

http://www.missioncriticallinux.com

 

[altaratz]

Thanks to all very much for the help - I ended up donig it esserc's way - thanks!

Ethan