Stripping Comments

hangon has asked for the wisdom of the Perl Monks concerning the following question:

I need to strip comments from a configuration file. The comments start with # and go to the end of the line, unless the initial # is escaped. This is what I have:

while (my $line = <FH>){

    $line =~ s/[^\\]#.*//;
    $line =~ s/^#.*//;
    $line =~ s/\\#/#/g;

    print "$line";
}
[download]

The second regex is to handle cases where the line begins with #. The third just removes the escape character. This works, but is there a better way or can it be simplified with with fewer regexes? Thanks.

sample data:

\#Not a comment
#Comment
\#Not a comment  #But this is a comment
arbitrary text   #Comment     
arbitrary text   \#Not a comment
arbitrary text   \#Not a comment  #But this is a comment
arbitrary text   #Comment         \#This is part of the comment
[download]

desired output:

#Not a comment

#Not a comment
arbitrary text
arbitrary text   #Not a comment
arbitrary text   #Not a comment  
arbitrary text
[download]

UPDATE: Got what I needed, and a little education too. Thanks for the help everyone.

Comment on Stripping Comments Select or Download Code

Replies are listed 'Best First'.
Re: Stripping Comments by varian (Chaplain) on Jun 09, 2007 at 09:18 UTC
Try this: `#!/usr/bin/perl use strict;use warnings; while (my $line=<DATA>) { $line=~s/(?<!\\)#.*//; print $line; } __DATA__ \#Not a comment #Comment \#Not a comment #But this is a comment arbitrary text #Comment arbitrary text \#Not a comment arbitrary text \#Not a comment #But this is a comment arbitrary text #Comment \#This is part of the comment` [download] Prints: `\#Not a comment \#Not a comment arbitrary text arbitrary text \#Not a comment arbitrary text \#Not a comment arbitrary text` [download] The '(?<!' construct looks backward to ensure that the escape character is not preceeding the comment character.	[reply] [d/l] [select]
Re: Stripping Comments (one) by tye (Sage) on Jun 10, 2007 at 02:28 UTC
`no warnings 'uninitialized'; s<\\(#)\|\s#[^\n]$><$1>g` [download] Note that I meet your examples as you entered them and remove the spaces before the "#" comment, unlike your code or most of the other code in this thread. I use `[^\n]` instead of just `.` since it is more explicit, especially since in this case it is important to stop at a newline. - tye	[reply] [d/l] [select]
Re: Stripping Comments by moritz (Cardinal) on Jun 09, 2007 at 09:10 UTC
I don't see why you need the first regex. Perl 6 in German	[reply]
Re^2: Stripping Comments by betterworld (Curate) on Jun 10, 2007 at 00:54 UTC
The problem with the first RE is that it removes the character before the hash mark. In your examples, it's always whitespace, so you didn't recognize it. You could replace the first regex with `s/([^\\])#.*/$1/;` [download] Better yet, replace the first and second RE with the RE that varian provides. You'll need the third RE nonetheless.	[reply] [d/l]


Just another Perl shrine
	PerlMonks