I kind of figured splitting on whitespace was the original objective, but I suggested unpack as what I thought might be a better alternative. However, split is pretty simple to use, so here's an example that pulls out the 2nd and 6th fields:

{
    local $, = "\t";
    print +(split)[1,5], $/ while <DATA>;
}
[download]

HTH!

great thanks alot! You rock! I asked the email list at beginners@perl.org and I received no responses. Anyway I have two questions related to my desired output. 1) from my code I am getting these errors Filehandle FOO opened only for output at derek line 23, <FOO> line 39. 08/09/04 11:02:43 E01220 Filehandle FOO opened only for output at derek line 23, Here is my code:

## Define ENV

$ENV{"PATH"} = qq(/usr/epoch/bin:/usr/epoch/EB/bin:/usr/bin:/usr/sbin:
+/bin:/sbin);
delete $ENV{"IFS"};
        
## Declare scalers for client templates and client tape logs

        my $scratches="/usr/local/log/scratch_tapes.log";
        my $logf="/usr/epoch/EB/log/ebcat.log";

## Gather data for scratch tapes; open file, create array list, then p
+rint it.

        open (FOO, ">$scratches") || die "could not open file:$!";
        my $scratch_list = `grep 9840S $logf |grep "ebexpire, ebexpire
+"`;
        print FOO "$scratch_list";
                while (<FOO>) {
                        local $, = "\t";
                        print +(split)[0,1,5], $/;
                }
        close (FOO);
[download]

can anyone shed some light? Also to use the Perl grep I have code as follows but it is not working.

## Define ENV

 $ENV{"PATH"} = qq(/usr/epoch/bin:/usr/epoch/EB/bin:/usr/bin:/usr/sbin
+:/bin:/sbin);
 delete $ENV{"IFS"};
        
## Declare scalers for client templates and client tape logs

        my $scratches="/usr/local/log/scratch_tapes.log2";
        my $logf="/usr/epoch/EB/log/ebcat.log";

        open (FOO, ">$scratches") || die "could not open file:$!";
        open (D, "$logf") || die "could not open file:$!";
        while ( <D> ) {
                if (( /9840S/ ) && ( /"ebexpire, ebexpire"/ )) {
                        print $_ $scratches
                }
        }
        close (FOO);
        close (D);
[download]

2 example lines from the ebcat.log looks like 04/29/04 11:01:17 6687:ebexpire, ebexpire.c@7475 E01012 9840S 820 6E01C1874C2E12CA 04/29/04 11:01:17 6687:ebexpire, ebexpire.c@7475 E01093 9840S 898 9601C7263CDC2B34 thank you!

The problem is that you've opened the filehandle FOO for output, but then you try to read from it with the <FOO> operator. You can't do that. Opening the file for output is going to clobber whatever its contents were, and you can't read from a file opened with ">".

See perlopentut and perlopen for details.




Dave

so... is the code

print +(split)[1,5], $/ while <DATA>;
[download]

saying "split <DATA> on the 1st and 5th fields as IRS's for output? But what does the +(split) mean? thanks!

print +(split)[1,5], $/ while <DATA>;

Could also be written like this:

while ( my $line = <DATA> ) {
    my ( $second, $sixth ) = ( split //, $line )[1,5];
    print $second, $sixth, "\n";
}
[download]

The unary + sign is used to signal to Perl that you mean print( (split)[1,5], "\n" ); and not print(split[1,5]),"\n"; (which doesn't make sense, but is confusing to Perl nevertheless.

For a description of split, see split. For a description of using slices, see perldata. For a description of what $/ is, see perlvar. And for an explanation of the "do something while condition" syntax, see perlsyn.




Dave