Re^4: text matching

I am running the code in cygwin and here is my complete code.

#!/bin/perl5.8.6

use warnings;
use diagnostics;
use File::Basename;

@files = <../tryremove/*>;

#This is the destination path where the comment removed files are crea
+ted.
$path1 = "C:/Projects/tryremove/c";

foreach my $file (@files) {
  my $filename =basename($file);
  open(file1,"$file") or die "The file cannot be opened the $file:$!\n
+";
  open(file2,">$path1/$filename") or die "The file cannot be opened:$!
+\n";

while(!eof(file1)){
  $/ = undef;
  $_ = <file1>;
  s# .* report [s]? .* ; .* ##sx;
  print file2 ($_);
}

close(file1);
close(file2);
}
[download]

Comment on Re^4: text matching Download Code

Replies are listed 'Best First'.
Re^5: text matching by frozenwithjoy (Priest) on Jun 16, 2012 at 07:09 UTC
Thanks. Please delete the line: `$/ = undef;` and let me know if it works.	[reply] [d/l]
Re^6: text matching by prassi (Acolyte) on Jun 16, 2012 at 07:20 UTC
it worked, can you help out what actually `$/ = undef;` [download] would do. I am a newbie to perl try things to understand	[reply] [d/l]
Re^7: text matching by frozenwithjoy (Priest) on Jun 16, 2012 at 07:58 UTC
A few things to keep in mind as you write more code. Always `use strict;` `use autodie;` When opening file handles, don't use a bare file handle. For example, use `$file` instead of `file` When opening file handles, use three arguments: `open my $fh, '>', $filename` (see this for a brief explanation) I strongly recommend explicitly naming your variables if you and/or your colleagues are learning perl. It can make the code much more readable; for example, compare your while loop to mine. I quickly cleaned up the code you posted according to my suggestions so you can see them side-by-side: `#!/bin/perl5.8.6 use strict; use warnings; use diagnostics; use File::Basename; my @files = <../tryremove/>; #This is the destination path where the comment removed files are crea +ted. my $path1 = "C:/Projects/tryremove/c"; foreach my $file (@files) { my $filename = basename($file); open $file1, "<", $file; open $file2, ">", "$path1/$filename"; while ( my $line = <$file1> ) { $line =~ s# . report s? .* ; .* ##sx; print $file2 $line; } close $file1; close $file2; }` [download] Disclaimer: It's late and I'm pretty sleepy. Although I think it should be fine, test it to make sure it behaves the same as your fixed script! - Mike	[reply] [d/l] [select]
Re^8: text matching by prassi (Acolyte) on Jun 16, 2012 at 14:02 UTC
Re^7: text matching by Anonymous Monk on Jun 16, 2012 at 07:27 UTC
see perlvar#$/	[reply]
Re^7: text matching by frozenwithjoy (Priest) on Jun 16, 2012 at 07:34 UTC
That's good to hear! `$/` is the input record separator and is essentially used to specify what character should be used to delimit lines. The default input record separator is `\n` (a.k.a. the newline character). However, in the context that you have it, `$/ = undef;` tells perl that there is no input value separator so that the entire line gets read into a scalar on the subsequent line of your script: `$_ = <file1>;`	[reply] [d/l] [select]
Re^5: text matching by Cristoforo (Curate) on Jun 17, 2012 at 01:13 UTC
Here is the program using a different approach with the range operator. `#!/usr/bin/perl use strict; use warnings; while (<DATA>) { if ((my $first = /^\s#if/) .. (my $last = /^\s#endif/)) { #print if $first \|\| $last; print unless /^\s*reports?/; } else { print; } }` [download] Its a neat use of the range, ( .. ), operator. Chris Update: changed the first print line to more accurately perform what you needed.	[reply] [d/l]