Perl regex replace

Question

I am trying to take a file, and copy it to an output file and then find all the <k> and make them into <p> instead. I want to use regex.

  while (<INPUTFILE>)
  {
         if($_ != "<p>")
         {
                print OUTPUTFILE "$_\n";  
          }

          else
          {
              print OUTPUTFILE "<k>";  
            }

  }

I was trying something like this but I guess the $_ variable takes each line at a time. So It would prob be better to use write the entire file then try the regex search and replace? for that I have tried:

  $_ =~ s/<k>/<p>/g;

Do you have a sample of the file content? Is it simply either <p> or <k> on lines or is there more content? — Nic Gibson
– Nic Gibson, Commented Oct 24, 2013 at 19:39
Your code seems to be looking for <p> in the input and printing <k> when it finds it. Your description has the inverse logic. Which is right? — Ted Hopp
– Ted Hopp, Commented Oct 24, 2013 at 19:40
Always use use strict; use warnings;. It would have found a major error. — ikegami
– ikegami, Commented Oct 24, 2013 at 19:40
Unless you use chop or chomp, $_ already ends with a newline. So you shouldn't add one when you print OUTPUTFILE "$_";. — Barmar
– Barmar, Commented Oct 24, 2013 at 19:42
To expand on @Barmar's comment: since you aren't chomping the input, $_ contains a newline, your if($_ != "<p>") will always be !=, since your test literal has no newline - unless your input file ends with a line that isn't newline-terminated — Jason Clark
– Jason Clark, Commented Oct 24, 2013 at 19:44

ikegami · Accepted Answer · 2013-10-24 19:41:20Z

5

A line at a time is not a problem.

while (<>)
   s/<k>/<p>/g;
   print;
}

answered Oct 24, 2013 at 19:41

ikegami

391k17 gold badges291 silver badges555 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Nic Gibson · Accepted Answer · 2013-10-24 19:44:41Z

1

As @ikegami suggests, one line at a time is no problem. Alternatively, you could do this in a single search and replace:

{
   local $/ = undef;
   my $data = <INPUTFILE>;
   $data =~ s/<k>/<p>/g;
}

This would do the search and replace in a single operation and might be faster and/or more efficient for larger files.

answered Oct 24, 2013 at 19:44

Nic Gibson

7,1634 gold badges33 silver badges40 bronze badges

Comments

vogomatix · Accepted Answer · 2013-10-24 19:46:42Z

0

From the command line

perl -pe 's/<k>/<p>/g' <infile >outfile

Simples.... :-)

Look up 'change record separator' if you are concerned about newline effects

answered Oct 24, 2013 at 19:46

vogomatix

5,1332 gold badges27 silver badges51 bronze badges

Comments

Gilles Quénot · Accepted Answer · 2013-10-24 19:49:18Z

0

If you really want to treat the whole file in one time :

BEGIN { $/ = undef; $\ = undef; }
LINE: while (defined($_ = <ARGV>)) {
    s/<k>/<p>/g;
    print $_;
}

From :

perl -0777 -MO=Deparse -pe 's/<k>/<p>/g; print;'

edited Oct 24, 2013 at 19:49

answered Oct 24, 2013 at 19:41

Gilles Quénot

188k43 gold badges232 silver badges229 bronze badges

1 Comment

ikegami Over a year ago

I think you were trying to slurp the file, but 1) you weren't (so I fixed it), and 2) it's useless. // And now you added /m (which changes what unused ^ and $ match) and /s (which changes what unused . matches). Also useless.

Collectives™ on Stack Overflow

Perl regex replace

4 Answers 4

Comments

Comments

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related