#!/usr/bin/perl # workaround for dzil package perlmv; BEGIN { $perlmv::VERSION = '0.41'; } use strict; use warnings; use FindBin '$Bin'; use lib "$Bin/../lib"; use App::perlmv; my $pmv = App::perlmv->new; $pmv->{ 'mode' } = 'rename'; $pmv->run; 1; =pod =head1 NAME perlmv =head1 VERSION version 0.41 =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<>) 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 source code of scriptlet: $ perlmv -s remove-digits s/\d+//g Remove 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 arrayref $FILES is also available to the Perl code. It contains, unsurprisingly, the list of files to be processed. The Perl code is not supposed to modify this value. Perl code will first be run (eval-ed) once at the beginning for testing, with C<-TEST> as the filename in C<$_> (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 NAME perlmv - Rename files using Perl code =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>) =head1 AUTHOR Steven Haryanto <stevenharyanto@gmail.com> =head1 COPYRIGHT AND LICENSE This software is copyright (c) 2010 by Steven Haryanto. This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself. =cut __END__