Process files by extension instead of individually - perl

Question

I have multiple files that have the extension .tdx.
Currently my program works on individual files using $ARGV[0], however the number of files are growing and I would like to use a wildcard based upon the file extension.
After much research I am at a loss.
I would like to read each file individually so the extract from the file is identified by the user.
#!C:\Perl\bin\perl.exe
use warnings;
use FileHandle;
open my $F_IN, '<', $ARGV[0] or die "Unable to open file: $!\n";
open my $F_OUT, '>', 'output.txt' or die "Unable to open file: $!\n";
while (my $line = $F_IN->getline) {
if ($line =~ /^User/) {
$F_OUT->print($line);
}
if ($line =~ /--FTP/) {
$F_OUT->print($line);
}
if ($line =~ /^ftp:/) {
$F_OUT->print($line);
}
}
close $F_IN;
close $F_OUT;
All the files are in one directory, so I assume I will need to open the directory.
I am just not sure how if I need to build an array of files or build a list and chomp it.

Answer 1

You have many options --
Loop over #ARGV, allowing the user to pass in a list of files
Use glob to pass in a pattern that perl will expand into a list of files (and then loop over that list, as in #1). This can be messy as they have to make sure to quote it so the shell doesn't interpolate it first.
Write some wrapper to call your existing script over and over again.
There's also a variant of the first one, which is to read from <>. This is set to either STDIN, or it'll automatically open the files named in #ARGV. See eof for an example of how to use it.
As an variant of #2, you can pass in a directory name, and use either opendir and readdir to loop over the list (making sure to grab only files with your extension, or at the very least ignore . and ..) or append /* or /*.tdx to it and use glob again.

Answer 2

The glob function can help you. Just try
my #files = glob '*.tdx';
for my $file (#files) {
# Process $file...
}
In list context, glob expands its argument to the list of file names that match the pattern. For details, see glob in perlfunc.

Answer 3

I never got glob to work. What I ended up doing was building an array based on the file extension .tdx. from there I copied the array to a filelist and read from that. What I ended up with is:
#!C:\Perl\bin\perl.exe
use warnings;
use FileHandle;
open my $F_OUT, '>', 'output.txt' or die "Unable to open file: $!\n";
open(FILELIST, "dir /b /s \"%USERPROFILE%\\Documents\\holding\\*.tdx\" |");
#filelist=<FILELIST>;
close(FILELIST);
foreach $file (#filelist)
{
chomp($file);
open my $F_IN, '<', $file or die "Unable to open file: $!\n";
while (my $line = $F_IN->getline)
{
Doing Something
}
close $F_IN;
}
close $F_OUT;
Thank you for your answers they helped in the learning experaince.

Answer 4

If you're on a Windows machine, putting in *.tdx on the command line might not work, nor may glob which historically used the shell's globbing abilities. (It now appears that the built in glob function now uses File::Glob, so that may no longer be an issue).
One thing you can do is not use globs, but allow the user to input the directories and suffixes they want. Then use opendir and readdir to go through the directories yourself.
use strict;
use warnings;
use feature qw(say);
use autodie;
use Getopt::Long; # Why not do it right?
use Pod::Usage; # It's about time to learn about POD documentation
my #suffixes; # Hey, why not let people put in more than one suffix?
my #directories; # Let people put in the directories they want to check
my $help;
GetOptions (
"suffix=s" => \#suffixes,
"directory=s" => \#directories,
"help" => \$help,
) or pod2usage ( -message => "Invalid usage" );
if ( not #suffixes ) {
#suffixes = qw(tdx);
}
if ( not #directories ) {
#directories = qw(.);
}
if ( $help ) {
pod2usage;
}
my $regex = join, "|", #suffixes;
$regex = "\.($regex)$"; # Will equal /\.(foo|bar|txt)$/ if Suffixes are foo, bar, txt
for my $directory ( #directories ) {
opendir my ($dir_fh), $directory; # Autodie will take care of this:
while ( my $file = readdir $dir_fh ) {
next unless -f $file;
next unless $file =~ /$regex/;
... Here be dragons ...
}
}
This will go through all of the directories your user input and then examines each entry. It uses the suffixes your user inputs (With .tdx being the default) to create a regular expression to check against the file name. If the file name matches the regular expression, do whatever you wanted to do with that file.