Line(s) extraction

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

I am relatively new to Perl, and as such am having a few questions. I am currently working on a text extraction script. I want to extract only a specific range of lines starting from the line or lines on which my reg exp is found. I am trying to extract from the line on which the expression is located + 3 lines, for a total of four displayed lines. Instead of using reg exps for range limiters (due to multiple finds), I want to use the $. variable. I have tried a few different ways in attempting to implement this variable. I have mainly modified line range examples I have found in bound documentation but to no avail. I have tried code such as:

while(<F>) {
 if(/\bREGEXP\b/) {
  $seen++;
  push(@lines, $.);
  print "\nMatch found:\n\n\tFILENAME: $dir/$file\n\tLINE: $. : '$`~$&
+~$'";
     if(($.) .. ($. + 3) { print; }     # my range extract code
     # OR
     print "$`$&$'\n" if(($.)..($. + 3));   # my range extract code
}else {
  next;
}
}
[download]

Is it possible to use the $. variable as a definitive start and end for a range of text extraction? If so, how can I go about implementing this for my line number range? Thank you. -paurus

Comment on Line(s) extraction Download Code

Replies are listed 'Best First'.
Re: Line(s) extraction by BrowserUk (Patriarch) on Apr 01, 2004 at 06:55 UTC
Rather than bothering with $., why not just print the line containing the regex and then 3 more lines? `#! perl -sw use strict; while( <DATA> ) { next unless m[two]; print; print scalar <DATA> for 1..3; } __DATA__ one two three four five six` [download] Examine what is said, not who speaks. "Efficiency is intelligent laziness." -David Dunham "Think for yourself!" - Abigail	[reply] [d/l]
Re: Line(s) extraction by kvale (Monsignor) on Apr 01, 2004 at 06:56 UTC
Here is a simple solution that doesn't use $. `while (my $line = <F>) { next unless $line =~ /\bREGEXP\b/; $line .= <F>; $line .= <F>; $line .= <F>; print $line; }` [download] -Mark	[reply] [d/l]