The following should have the makings of what you need (untested).
I'm worried, though, that you are just plugging in stuff from books without trying to understand. For example, the following opens the file and then your 'die' message should be reporting which file could not be opened:
open(FILE, "2.txt")
or die("can't open myfile.txt: $!\n");
Instead your code reports that some mystical 'myfile.txt' file could not be opened. The example below shows how the filename of the moment will appear in the error message if it has a problem opening.
I haven't looked through the rest of your program, I assume it does what you want. The key element from my sample code is the opendir/readdir combination which lets you work with directories.
Good luck with your quest of learning to use Perl!
P.S. You will also want to learn a bit about how the 'markup' system works at perlmonks. I believe your [ and ] symbols are being interpreted as html links. Putting <code> in front of, and </code> at the end of your code should keep it from being mangled by the perlmonks input system. (UPDATE: I notice that your code has since been cleaned up, but still, understanding the perlmonks markup system can be challenging but worthwhile. :-)
P.P.S. UPDATE: It occurred to me that since you are reusing the same code over many files, your use of global variables may get you in trouble. Each new file is going to be adding its contents into the variables %hash, @index, and @a2. You'd probably like them to start from scratch for each new file and so you should declare them anew with each pass through 'while'. Also, it appears that you will be blowing away the output with each
open FILEHANDLE, "> output.txt"
and so you should probably declare a new variable to hold the filename or plan on appending new info to the file with open FILEHANDLE, ">> output.txt"
A way to force yourself to think about where variables should be declared and how long they should be in force ("in scope") is to put use strict; near the start of your code. You will find it imposes rules that keep you from doing things that get you into some forms of trouble. It will also be troubling, at first, to accomodate use strict;, but the effort will be worth it. I have attached a second version of my untested quickie code to illustrate the differences.
$folder = '/tmp';
opendir TMPDIR, $folder
or die "can't opendir $folder: $!\n";
while ( $filename = readdir TMPDIR ) {
next unless $filename =~ /\.txt$/i;
open FILE, "$filename"
or die "can't open $filename: $!\n";
while(<FILE>) {
if ($_ =~ /\s+(\S+)\s+(\S+)/) {
$hash{$1} = $2;
push(@index,$1);
push (@a2, $2);
}
}
close(FILE);
# now do whatever
}
closedir TMPDIR;
Second version:
#!/usr/bin/perl -w
# Even though W32 doesn't process the #! portion, I believe it
# is smart enough to know that your inclusion of '-w' at the end
# means that you want warnings reported.
use strict;
my $folder = '/tmp';
opendir TMPDIR, $folder
or die "can't opendir $folder: $!\n";
while ( my $filename = readdir TMPDIR ) {
next unless $filename =~ /\.txt$/i;
my ( %hash, @index, @a2 );
open FILE, "$filename"
or die "can't open '$folder/$filename': $!\n";
while(<FILE>) {
if ($_ =~ /\s+(\S+)\s+(\S+)/) {
$hash{$1} = $2;
push(@index,$1);
push (@a2, $2);
}
}
close(FILE);
# now do whatever
}
closedir TMPDIR;
|