perlmv

#!/usr/bin/perl

use strict;
use warnings;
use FindBin '$Bin';
use lib "$Bin/../lib";
use App::perlmv;

my $pmv = App::perlmv->new;
$pmv->{ 'mode' } = 'rename';
$pmv->run;

1;
# ABSTRACT: Rename files using Perl code
# PODNAME: perlmv

=head1 SYNOPSIS

Usage:

 # Show help
 perlmv -h

 # Execute a single scriptlet
 perlmv [options] <scriptlet> <file...>

 # Execute code from command line
 perlmv [options] -e <code> <file...>

 # Execute multiple scriptlets/command-line codes
 perlmv [options] [ -x <scriptlet> | -e <code> ]+ <file...>

 # Create a new scriptlet
 perlmv -e <code> -w <name>

 # List available scriptlets
 perlmv -l

 # Show source code of a scriptlet
 perlmv -s <name>

 # Delete scriptlet
 perlmv -d <name>

=head2 Usage examples

 $ ls -1
 A.txt
 B1
 c2.txt
 D3.pl
 D4.pl

Rename files with prewritten scriptlet (B<remove-common-prefix>) and show (B<-v>)
each file as it is being renamed.

 $ perlmv -v remove-common-prefix file1.jpg file2.zip files.txt
 `file1.jpg` -> `1.jpg`
 `file2.zip` -> `2.zip`
 `files.txt` -> `s.txt`

Specify script in command line (B<-e>) but do not actually rename files (B<-d>,
dry-run mode):

 $ perlmv -de 's/\d+//g' *
 DRYRUN: `B1` -> `B`
 DRYRUN: `c2.txt` -> `c.txt`
 DRYRUN: `D3.pl` -> `D.pl`
 DRYRUN: `D4.pl` -> `D.pl.1`

Really rename the files this time:

 $ perlmv -e 's/\d+//g' *

Execute multiple scriptlets and/or command-line code:

 $ ls -1
 a.txt
 b.html
 c.ini

 $ perlmv -vx to-number-ext -e '"file$_"' *
 `a.txt` -> `1.txt` -> `file1.txt`
 `b.html` -> `2.html` -> `file2.html`
 `c.ini` -> `3.ini` -> `file3.ini`

Save Perl code as scriptlet (in ~/.perlmv/scriptlets/):

 $ perlmv -e 's/\d+//g' -w remove-digits

List all scriptlets (add B<-v> to also show their contents):

 $ perlmv -l
 lc
 uc
 remove-digits

Show (B<-s>) source code of scriptlet:

 $ perlmv -s remove-digits
 s/\d+//g

Remove (B<-D>) scriptlet:

 $ perlmv -D remove-digits

=head2 More examples

From my real-world usage.

Rename .flv partial files from my Firefox browser cache into ordered names
(part01.flv, part02.flv, and so on):

 $ ls --sort=t -r *01 | xargs perlmv -Tdx to-number-ext -e '"part$_.flv"'
 DRYRUN: `15D9F85Ad01` -> `01` -> `part01.flv`
 DRYRUN: `FF8EB240d01` -> `02` -> `part02.flv`
 DRYRUN: `9031E9A8d01` -> `03` -> `part03.flv`
 ...

=head1 DESCRIPTION

Perlmv lets you rename files using Perl code. All the Perl code needs to do is
modify the filename in C<$_> (or just return a value) and perlmv will do the rest
(actual renaming, recursive renaming, handling filename conflicts, dry-run mode,
etc.).

New filename will be taken from the new value of $_ (if it is modified) or the
last expression. So '$_ = lc' and 'lc' are equivalent.

The following variables are also available to the Perl code:

=over 4

=item * C<$FILES> (arrayref)

The list of items to be processed. The Perl code is not supposed to modify this
value.

=item * C<$FILE> (string)

The filename of item (without path). The Perl code is not supposed to modify
this value.

=item * C<$DIR> (string)

The directory of item (absolute path). The Perl code is not supposed to modify
this value.

=item * C<$PARENT> (string)

The name of parent (containing) directory of the item (without path). The Perl
code is not supposed to modify this value.

=back

Perl code will first be run (eval-ed) once at the beginning for testing, with
C<-TEST> as the filename in C<$_> (and C<$TESTING> will be set to true). Perl
code is not run under strict/warnings. Perl code is run under
C<App::perlmv::code> namespace.

When using the L<App::perlmv> library, there is also the the so-called cleaning
phase (in which Perl code will be run with $CLEANING set to true) that can be
used for cleaning purposes in between runs (call to rename()). This is however
irrelevant to B<perlmv> command as B<perlmv> only calls rename() once.

Perl code can be specified directly from the command line (using B<-e>), or by
name in C<~/.perlmv/scriptlets/NAME>, or in C</usr/share/perlmv/scriptlets/>, or
in C<%scriptlets> in L<App::perlmv::scriptlets>, or in C<%scriptlets> in
L<App::perlmv::scriptlets::std>.

=head1 OPTIONS

 -c  (--compile) Only test compile code, do not run it on the arguments
 -D <NAME> (--delete) Delete scriptlet
 -d  (--dry-run) Dry-run (implies -v)
 -e <CODE> (--execute) Specify Perl code to rename file (\$_). Can be specified
     multiple times.
 -f  (--files) Only process files, do not process directories
 -h  (--help) Show this help
 -l  (--list) list all scriptlets
 -M <MODE> (--mode) Specify mode, default is 'rename' (or 'r'). Use 'copy' or 'c'
     to copy instead of rename, 'symlink' or 's' to create a symbolic link, and
     'link' or 'l' to create a (hard) link.
 -o  (--overwrite) Overwrite (by default, ".1", ".2", and so on will be appended
     to avoid overwriting existing files)
 -p  (--parents) Create intermediate directories
 -R  (--recursive) Recursive
 -r  (--reverse) reverse order of processing (by default order is asciibetically)
 -S  (--no-symlinks) Do not process symlinks
 -s <NAME> (--show) Show source code for scriptlet
 -T  (--no-sort) do not sort files (default is sort ascibetically)
 -V  (--version) Print version and exit
 -v  (--verbose) Verbose
 -w <NAME> (--write) Write code specified in -e as scriptlet
 -x <NAME> Execute a scriptlet. Can be specified multiple times. -x is optional
     if there is only one scriptlet to execute, and scriptlet name is specified
     as the first argument, and there is no -e specified.

=head1 FAQ

=head2 Why should I use perlmv? There is prename (File::Rename) already?

Yes, there is a very similar script called B<prename> (also accessible via
B<rename> in Debian) which comes with Perl. This script reinvents prename and
offers more features, e.g.: automatic renaming in case of conflicts, recursive
mode, and saving and loading scriptlets.

=head2 And there is also pmv (File::PerlMove)!

Okay, you got me. I didn't do my homework. The "rename files using Perl
code/expression" is pretty obvious and has surely come up on other CPAN authors'
minds. To be honest, this is a script which I wrote years ago (at least in 2003,
or earlier) and have been using for years, personally. Admittedly I uploaded this
script to CPAN without careful checking of existing solutions on CPAN. But then,
lots of other CPAN modules are also overlapping in functionality with one
another.

Anyway, I plan to improve perlmv as I see fit, mainly for my own needs. I plan to
borrow some features from prename/pmv, and welcome them borrowing features from
perlmv. I welcome patches. And I am willing to submit patches to prename/pmv
after some discussions with the respective authors. And lastly, I am also open to
the idea of merging perlmv to either pername/pmv, if I can get all the features I
love in perlmv into those projects.

=head1 BUGS/TODOS

=over 4

=item * Patches for Windows welcome.

=item * Scriptlet should be able to receive arguments.

=back

=head1 SEE ALSO

L<prename> (L<File::Rename>)

L<pm> (L<File::PerlMove>)

=cut