Do you have a good Perl template script?
Pergunta
I do a lot of programming in Perl and was wondering if people had a "default" template Perl script that they use and willing to share.
I started copying one of my older scripts which has Getopt
functions. I am thinking people would have done something similar?
Solução
As people say before I have my methods templates in a module: use PMG::PMGBase;
and for the initial script escafolding, as an emacs user, I have my perl-insert-start and perl-add-getoption templates, but writing things like:
(defun perl-insert-start ()
"Places #!..perl at the start of the script"
(interactive)
(goto-char (point-min))
(insert "#!/usr/bin/env perl\n\n")
(insert "=head1 [progam_name]\n\n")
(insert " description:\n\n")
(insert "=cut\n\n")
(insert "use feature ':5.10';\n")
(insert "use strict;\n")
(insert "#use warnings;\n")
(insert "#use Data::Dumper;\n")
)
is a bit tiresome. So at the end is easier for me to have a Perl template script (see below), and call it with run-command-on-region: C-u M-| :~/scripts/perl-start-template.pl
from Emacs after selecting one space in a blank buffer:
#!/usr/bin/env perl
=head1 [progam_name]
description:
=cut
use feature ':5.10';
use strict;
use Getopt::Long;
my $prog = $0;
my $usage = <<EOQ;
Usage for $0:
>$prog [-test -help -verbose]
EOQ
my $help;
my $test;
my $debug;
my $verbose =1;
my $ok = GetOptions(
'test' => \$test,
'debug:i' => \$debug,
'verbose:i' => \$verbose,
'help' => \$help,
);
if ($help || !$ok ) {
print $usage;
exit;
}
print template();
sub template {
##
### Here start the template code
##
return <<'EOT';
#!/usr/bin/env perl
=head1 [progam_name]
description: This script prints a template for new perl scripts
=cut
use feature ':5.10';
use strict;
#use warnings;
#use Data::Dumper;
use Getopt::Long;
# use Template;
# use PMG::PMGBase;
# use File::Temp qw/ tempfile tempdir /;
# use File::Slurp;
# use File::Copy;
# use File::Path;
# use File::Spec;
# use File::Basename qw(basename dirname);
# use List::Util qw(reduce max min);
# use List::MoreUtils qw(uniq indexes each_arrayref natatime);
# my $PMGbase = PMG::PMGBase->new();
my $prog = $0;
my $usage = <<EOQ;
Usage for $0:
>$prog [-test -help -verbose]
EOQ
my $date = get_date();
my $help;
my $test;
my $debug;
my $verbose =1;
my $bsub;
my $log;
my $stdout;
my $stdin;
my $run;
my $dry_run;
my $ok = GetOptions(
'test' => \$test,
'debug:i' => \$debug,
'verbose:i' => \$verbose,
'help' => \$help,
'log' => \$log,
'bsub' => \$bsub,
'stdout' => \$stdout,
'stdin' => \$stdin,
'run' => \$run,
'dry_run' => \$dry_run,
);
if ($help || !$ok ) {
print $usage;
exit;
}
sub get_date {
my ($day, $mon, $year) = (localtime)[3..5] ;
return my $date= sprintf "%04d-%02d-%02d", $year+1900, $mon+1, $day;
}
sub parse_csv_args {
my $csv_str =shift;
return [split ',', $csv_str];
}
EOT
}
Outras dicas
In my .vimrc
file I have
au BufNewFile *.pl s-^-#!/usr/bin/perl\r\ruse strict;\ruse warnings;\r\r-
which writes
#!/usr/bin/perl
use strict;
use warnings;
to any new Perl script. I also have
au BufNewFile *.pm s-^-package XXX;\r\ruse strict;\ruse warnings;\r\r1;-
for modules, but I tend to use Module::Starter
for those anyway.
When I need a basic template for many similar scripts, I just turn the similar parts into a module. The script then reduces to something like:
use App::Foo;
App::Foo->run( @ARGV );
The App::Foo
would inherit from the template module and override whatever was different:
package App::Foo;
use parent qw(App::Template);
...
In the App::Template
module, you put in whatever you need:
package App::Template;
sub run {
my( $class, @args ) = @_;
my $self = $class->new( ... );
$self->init;
$self->process_command_line( ... );
...
}
sub process_command_line { ... }
...
There are some frameworks on CPAN for this sort of thing, but I think it's just as easy to do it yourself and get exactly what you need without dealing with the parts you don't need.
Mine is pretty simple.
#!/usr/bin/perl
use Modern::Perl
When it comes to things like getopt, there aren't enough commonalities among the scripts I write to make it worth while having a more verbose template.
I have two. An old one which is little more than a wrapper to a perl one-liner and a second that has more functions and examples that I often find useful:
#!/usr/bin/perl
# name_of_script ver 0.01 YYYYMMDD authors@email.address
use strict;
no strict "refs";
sub footer
{
my $this_year=`date +%Y`; chop($this_year);
print "Copyright 2003-$this_year You or Company\n";
# This isn't how copyright works - the dates cove the time when the code
# was created.
}
sub help
{
print "Usage: $0\n";
&footer;
exit(0);
}
if( ($ARGV[0] =~ /^-+h/i) || (!$ARGV[0]) )
{
&help;
}
##### code
##### end of code
print "Done that\n";
exit(0);
I use the above for quick test but more often I use the following, (when I'm not hacking a full module.)
#!/usr/bin/perl
# name_of_script ver 0.01 YYYYMMDD authors@email.address
use strict;
{
no strict "refs"; # this helps bypass frustration when I'm doing it wrong.
}
=head1 NAME
name_of_script
=head1 VERSION
0.01
=cut
our $VERSION = 0.01;
=head1 ABSTRACT
A synopsis of the new script
=head1 DESCRIPTION
Provide an overview of functionality and purpose of
this script
=head1 OPTIONS
%opt stores the global variables
%ignore overrides %opt
=cut
my (%opt,%ignore);
=head2 ARGS
=over 8
=item B<-h> send for help (just spits out this POD by default, but we can chose something else if we like
=back
=head3 other arguments and flags that are valid
For when GetOpt is too heavy
-d -v -i[!] (value)
=cut
for(my $args=0;$args<=(@ARGV -1);$args++){
if ($ARGV[$args]=~m/^-+h/i){ &help; }
elsif ($ARGV[$args] eq '-d'){ $opt{D}++; }
elsif ($ARGV[$args] eq '-v'){ $opt{verbose}++; print "Verbose output not implemented yet - try debug\n";}
elsif ($ARGV[$args]=~m/-+i!(.+)/){ delete($ignore{$1}); }
elsif ($ARGV[$args]=~m/-+record(.+)/){ $opt{record_data}++; }
elsif ($ARGV[$args]=~m/-+w(ipe_home_dirs)?/){ $opt{wipe_home_dirs}++; }
elsif ($ARGV[$args]=~m/-+i(.+)/){ $ignore{$1}=1; }
elsif ($ARGV[$args]=~m/-+path(.+)/){ $opt{BASE_PATH} = $1; }
elsif ($ARGV[$args]=~m/-+path/){ $args++; $opt{BASE_PATH} = $ARGV[$args]; }
elsif ($ARGV[$args]=~m/-+dir(.+)/){ $opt{BASE_PATH} = $1; }
elsif ($ARGV[$args] eq '-no-xml'||$ARGV[$args] eq '-no_xml'){ delete $opt{xml}; }
elsif ($ARGV[$args] eq '-no-mkdir'||$ARGV[$args] eq '-no_mkdir'){ delete $opt{mkdir}; }
elsif ($ARGV[$args] !~m/^-/ && -d "$ARGV[$args]"){ push @{ $opt{paths} }, $ARGV[$args] }
else{ print "what is this $ARGV[$args] you talk of?\n"; &help; }
}
=head1 METHODS
=head3 footer
Adds the Copyright line to any output that needs it
=cut
sub footer { print "perldoc $0 \nCopyright 2011 You or Company\n"; }
=head3 help
Just the help output
=cut
sub help {
print `perldoc $0`;
#print "Usage: $0\n";
#&footer;
exit(0);
}
##### code
##### end of code
=head1 BUGS AND LIMITATIONS
There are no known problems with this script.
Please report any bugs or feature requests
=head1 SEE ALSO
#L<My::Modules>
=head1 MAINTAINER
is the AUTHOR
=head1 AUTHOR
Some Person, C<<some.person at example.com>>
=head1 LICENSE AND COPYRIGHT
Copyright 2011 Alexx Roche, all rights reserved.
This program is free software; you can redistribute it and/or modify it
under the terms of either: Eclipse Public License, Version 1.0 ;
the GNU Lesser General Public License as published
by the Free Software Foundation; or the Artistic License.
See http://www.opensource.org/licenses/ for more information.
=cut
print "Done that\n" if $opt{verbose}>=1;
exit(0);
__END__
__END__ is usually only used if we are going to have POD after the code If you move the "Done that" above the POD then __END__ makes more sense to me.
Feel free to hack these two about as much as you like. I make no claims to good style or practices here, (and I sometimes start with the short one and paste in blocks from the longer one as I need them ending up with two code styles for the trolols.)