Improving lexical scoping outside subs

I always start with use strict;, but still I have problems with lexical variables scoped outside of subroutines. Here is some simple code which demonstrates the problem:

use strict;

my $fn="myfile";
print readfile($fn);

sub readfile
{
  my $out="";
  open(FH, $fn) or die "Can't read $fn";
  $out .= $_ while(<FH>);
  return $out;
}
[download]

I meant to use $fn as passed to the subroutine, but strict didn't help me catch my mistake.

Here is the improved version:

use strict;

{
  my $fn="myfile";
  print readfile($fn);
}

sub readfile
{
  my $out="";
  open(FH, $fn) or die "Can't read $fn";
  $out .= $_ while(<FH>);
  return $out;
}
[download]

In the improved version, strict catches the mistake. The improvement is to write { } around code outside the subs.

This helps me with debugging, especially casual programs.

It should work perfectly the first time! - toma

Comment on Improving lexical scoping outside subs Select or Download Code

Replies are listed 'Best First'.
Re: Improving lexical scoping outside subs by broquaint (Abbot) on Aug 18, 2003 at 08:51 UTC
This style of putting subroutine definitions below the main body of code is explored at where do you put your subs with many arguments going either way. In terms of not being bitten it's saner to put subs at the top of your code, but if you're happy to get into the habit of putting the main body of code in bare blocks, then it's no longer an issue (unless, of course, you want to use bareword sub calls). Another method is to have the main body of code in a subroutine and just have a call to that with `@ARGV` and `exit` e.g `exit main(@ARGV); sub main { ... }` [download] For which tye listed four good reasons for using that style at Re^2: END failed --- call queue aborted (or was it three ...). HTH `_________ broquaint`	[reply] [d/l]
Re: Improving lexical scoping outside subs by greenFox (Vicar) on Aug 18, 2003 at 08:45 UTC
I recommend that you read the responses to where do you put your subs for further discussions on this topic and other techniques to avoid it. -- Do not seek to follow in the footsteps of the wise. Seek what they sought. -Basho	[reply]
Re: Improving lexical scoping outside subs by dragonchild (Archbishop) on Aug 18, 2003 at 13:20 UTC
While the comments thus far have been good in having Perl discover your error, I would put forward that there's a human way as well. Whenever I write a function, I always put the following template in: `1 sub foo 2 { 4a my $self = shift; 4b my () = @_; 3 }` [download] The numbers are the order I type the lines in. (4a is used only if I'm writing a method.) By doing this, I guarantee that I have a `my () = @_;` line. I suspect that if you had had that line in your code, you would've remembered to put the variable in there. Also, I try very hard to not use the same name in a sub as I would outside the sub. There's a good reason for it, too. In a sub, you're dealing with an abstract flow, so a name like $filename or $fh is useful. Outside, you're dealing with concrete things, so names like $out_filename or $data_fh are much better. So, what you would have is `my $data_fh = IO::File->new($data_filename) \|\| die "Cannot open '$data_filename' for reading: $!\n"; my $stuff = read_file_with_format($data_fh); ... my $second_fh = IO::File->new($second_filename) \|\| die "Cannot open '$second_filename' for reading: $!\n"; my $other_stuff = read_file_with_format($second_fh); ... sub read_file_with_format { my ($fh) = @_; ... }` [download] ------ We are the carpenters and bricklayers of the Information Age. The idea is a little like C++ templates, except not quite so brain-meltingly complicated. -- TheDamian, Exegesis 6 Please remember that I'm crufty and crochety. All opinions are purely mine and all code is untested, unless otherwise specified.	[reply] [d/l] [select]
Re: Improving lexical scoping outside subs by fletcher_the_dog (Friar) on Aug 18, 2003 at 14:33 UTC
A good idea that a friend introduced me to was using a "MAIN" label. This way you don't have to make a call to a "main" subroutine(this is nice because you can use @ARGV like you normally would), you prevent variables from having too large of scope and you make clear what the main part of your program is. Your script would look something like this: `MAIN:{ my $fn="myfile"; print readfile($fn); } sub readfile { my $out=""; open(FH, $fn) or die "Can't read $fn"; $out .= $_ while(<FH>); return $out; }` [download]	[reply] [d/l]