But since your code fragment doesn't show any print statements, I expect your real code is different anyway.

Abigail

-- Randal L. Schwartz, Perl hacker

I agree, testing your code you need to either do as rollyguy has suggested, or it does work doing:


open (PROG, "+>>$0");
[download]

In running your code withthe + in it, I have found your regexp does not work as well.
Here is my version, and followed by the output:



#!/usr/bin/perl

use strict;
use warnings;

my $new_value = 10;
my $data=45;

open (PROG, "+>>$0");
seek (PROG, 0, 0);     #not working 
while (<PROG>)
{
  s/^(my .data\s*=\s*)\d+(\s*;)/$1.$new_value.$2/e;
  print;
}
close (PROG);
[download]

Output:


#!/usr/bin/perl

use strict;
use warnings;

my $new_value = 10;
my $data=10;

open (PROG, "+>>$0");
seek (PROG, 0, 0);     #not working 
while (<PROG>)
{
  s/^(my .data\s*=\s*)\d+(\s*;)/$1.$new_value.$2/e;
  print;
}
close (PROG);
[download]


Note: I'm not writing to the file itself, now that I'm thinking about it you probably want to write a temp file, then copy it over....

Hope this helps!

Thanks to all you monks for the precious suggestions you gave to me!
I know my code is still terrible!!

#!perl

use strict;
use warnings;

&changemyself();
sub changemyself{
  my $new_value = 10;
  my $data=6666666666666;   #next time this value will be 10

 open (COPIA, ">copia");
 open (PROG, "+>>$0");
 seek (PROG, 0, 0);         #I don't know why but it's NECESSARY!

 while (<PROG>)
 {
   s/^(  my .data\s*=\s*)\d+(\s*;)/$1.$new_value.$2/e;
   print COPIA;
 }
 close (PROG);              #
 close (COPIA);             #
                            #   the terrible imitation 
                            #     of the
 open (COPIA, "<copia");    #   three cards game
 open (PROG, ">$0");        #
 while (<COPIA>){print PROG}#
 close (PROG);              #
 close (COPIA);             #
 unlink ("copia");          #
}
[download]

Special thank to Helter!
greetings from roma, lor*
PS I never noticed the +>> operator!!! l*

I don't really advocate what you are doing, better I think to store your vars in a seperate file and read them in at runtime, but if you insist, you could use Inline::Files, which make what your doing fairly trivial.

#! perl -sw
use strict;
use Inline::Files -backup;
BEGIN{
    $::name1 = <MY_VARS>; chomp($::name1);
    $::name2 = <MY_VARS>; chomp($::name2);
}

print "This time I have $::name1, and $::name2\n";

$::name1 = 'frederica';
$::name2 = 'willamina';

END{
    open MY_VARS, '>' or die $!;
    print MY_VARS "$::name1\n$::name2\n";
    close MY_VARS;
}
__MY_VARS__
fred
bill
__DATA__

C:\test>202246
This time I have fred, and bill

C:\test>202246
This time I have frederica, and willamina

C:\test>
[download]

Nice things is that the module is tested, proven and has a backup facility (though only one level of backup).

Also, doing this way you don't have to worry about making changes to your code and then having to modify your regexes, or risk forgetting and totally screwing your code.

Using the BEGIN and END blocks makes sure it get done at the right time.

This is a safer way to do it. It will keep a history of modifications, and you can "undo" it using an editor if you want:

open (ME, ">>$0");
print ME "data\n";
close(ME);
__DATA__
[download]

open (PROG, "+<$filename");
[download]