=head1 NAME

record_separator.pl - demonstrates how the $/ record 
separator works

=head1 AUTHOR

zeno

=head1 SYNOPSIS

perl record_separator.pl

=head1 DESCRIPTION

This program explains how the $/ record separator charactor 
works. Setting it equal to different values changes the 
behavior of the angle brackets operators.

I got interested in this when I tried to explain to someone
how to read more than one line at a time from a data file
to use a regex to get rid of duplicate lines.
See also:
How can I extract just the unique elements of an array?
http://www.perlmonks.com/index.pl?node_id=609


Note the interesting way we have to "rewind" the DATA file
handle. I got this from perlmonks:
"re-opening DATA, printing any character"
http://www.perlmonks.com/index.pl?node_id=23403

Understand that the $/ is not a regex.  It can't contain,
for example, character classes or alternatives.  While it
can consist of more than one character, trying to
consistantly use this to read, for example, paragraphs,
won't work, because a paragraph can end in different
combinations of characters.
See "Setting Input record separator to chunk on paragraphs":
http://www.perlmonks.com/index.pl?node_id=28681

=cut

use strict;
use warnings;


# record the start of the DATA (if we don't do this and try 
# to seek(DATA,0,0), we start reading again from the start 
# of the program!  interesting, but not very useful.
my $startOfData= tell DATA;

print "HOW DOES THE END OF RECORD SEPARATOR WORK?\n";
print "Normal behavior (\$/ = newline)\n";
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

seek( DATA, $startOfData, 0 ) or die "Can't seek: $!";
print "\nRead word for word (\$/ = ' ')\n";
print "Careful here! there isn't always a space ";
print "between words!  \n";
print "And sometimes there's more than one!\n";
$/ = " ";
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

seek( DATA, $startOfData, 0 ) or die "Can't seek: $!";
print "\nRead sentences (\$/ = '.')\n";
$/ = ".";
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

seek( DATA, $startOfData, 0 ) or die "Can't seek: $!";
print "\nRead psuedo-paragraphs (\$/ = '.\\n')\n";
print "Careful here! Not all paragraphs end this way!\n";
$/ = ".\n";
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

seek( DATA, $startOfData, 0 ) or die "Can't seek: $!";
print "\nsuck mode (undef \$/)\n";
undef $/;
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

seek( DATA, $startOfData, 0 ) or die "Can't seek: $!";
print "\nParagraph mode (\$/ = '')\n";
$/ = '';
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

seek( DATA, $startOfData, 0 ) or die "Can't seek: $!";
print "\nRead a certain number of chars (\n\$/ = \\12)\n";
$/ = \12;
while (<DATA>) {
  chomp;
  print "I read ->$_<-\n";
}

__DATA__
Hello there mom.
I'm sorry I haven't
written    for a while.

All is well!
Everything is fine.  You're great.
[download]