formatting text

texuser74 has asked for the wisdom of the Perl Monks concerning the following question:

Pls help me in optimizing my code.
I have an input file test.txt, in which i need to remove extra enter marks and spaces in the specific lines that are starting with <item>.

input: test.txt

<item>sample   item text one


<item>sample item text two

<item>sample item  text three


<item>sample item text         four
[download]

My present code does it, but i want to do it with fewer lines of code

open(IN, "test.txt") || die "\nCan't open test.txt \n";
open(OUT, ">test.out");
$/="";
{
   local $/ = '<item>';
   print OUT scalar <IN>;
   for (<IN>) {
      s@([\d\D]*?<item>)@
      my $var = $1;
      $var =~ s!\s+! !g;
      $var @e;
      print OUT;
   }
}
close(IN);
close(OUT);
[download]

The required output is:

<item>sample item text one
<item>sample item text two
<item>sample item text three
<item>sample item text four
[download]

Comment on formatting text Select or Download Code

Replies are listed 'Best First'.
Re: formatting text by ikegami (Patriarch) on Jun 20, 2008 at 03:00 UTC
Two quick points first. `/[\d\D]/`? That looks like a weird way to say `/./s` or even `/(?s:.)/`. `for (<IN>)` is similar to `while (<IN>)` except it takes way more memory since it reads the entire file into memory. Use `while (<IN>)`. Now back to the question, `open ...; open ...; $/ = ''; while (<IN>) { s/^\s+//; s/\s(?=\s)//g; print OUT; }` [download] Or as a one-liner `perl -pe"BEGIN{$/=''} s/^\s+//; s/\s(?=\s)//g" test.txt > test.out` [download] I'm sure there are many HTML formatters out there, though. You'd probably be better off using one of them.	[reply] [d/l] [select]
Re: formatting text by Narveson (Chaplain) on Jun 20, 2008 at 03:22 UTC
Using your own global substitution `s!\s+! !g`, but shortening everything else: `while (<DATA>) { next if !/^<item>/; s!\s+! !g; print "$_\n"; } __DATA__ <item>sample item text one <item>sample item text two <item>sample item text three <item>sample item text four` [download]	[reply] [d/l] [select]
Re: formatting text by waldner (Beadle) on Jun 20, 2008 at 07:40 UTC
a simple one liner is `perl -ne 'if (/^<item>/){s/[ \t]+/ /g;print;}' test.txt` [download] I don't use \s because that would remove the newline character at the end of the line too. Edit: made it a little shorter and added /g (which is of course the most important thing).	[reply] [d/l]
Re: formatting text by jwkrahn (Abbot) on Jun 20, 2008 at 03:51 UTC
Perhaps this will do what you require: `open IN, '<', 'test.txt' or die "Can't open 'test.txt' $!"; open OUT, '>', 'test.out' or die "Can't open 'test.out' $!"; while ( <IN> ) { next unless /\S/; tr/ \t/ /s; print OUT; # Updated -- thanks Narveson } close IN; close OUT;` [download]	[reply] [d/l]
Re^2: formatting text by Narveson (Chaplain) on Jun 20, 2008 at 04:00 UTC
Revised I set up `test.txt` and ran the first version of this code and got no output. Added `use warnings` and posted the warning messages here with no comment. Sorry if that came across as unfriendly. Read more... (738 Bytes)	[reply] [d/l] [select]


Think about Loose Coupling
	PerlMonks