mav3067 has asked for the wisdom of the Perl Monks concerning the following question:

Hey Everyone, I wrote a script which was originally a one off thing, but now more people have expressed interest in using it and I would like to make it a little more robust. I would like to make my script handle all types of new line characters LF,CR, and CRLF so that the users can remain in their blissfully unaware state about such things ;) Here are the file loading subs:
sub load_parents
{
    my ($parent_index) = @_;
    my %parents;

    open (FH, "< $parent_index") or die "could not open file $parent_i
+ndex for reading:$!";

    while (my $line = <FH>) {
      chomp($line);
      my ($index, $parent) = split(/\t/, $line);
      $parents{$index} = $parent;
    }
    return \%parents;

}

sub load_primers
{
    my ($primer_index) = @_;
    my %primers;

    open (FH, "< $primer_index") or die "could not open file $primer_i
+ndex for reading:$!";

    while (my $line = <FH>) {
      chomp($line);
      my ($index, $primer) = split(/\t/, $line);
      $primers{$index} = $primer;
    }
    return \%primers;

}

[download]
I woud appreciate any advice on what I could do to hanle the different new line characters, or if you see a blatent flaw in my logic don't be affraid to point that out either.
Thanks, Mav

Replies are listed 'Best First'.
Re: handling different new line characters
by Velaki (Chaplain) on Aug 28, 2006 at 15:09 UTC

    Instead of chomp($line), you could use a regex: 

    $line =~ s/[\n\r]//g;

    [download]

    This way you're handling both the CR and the LF.

    Hope that helped,
    -v.

    "Perl. There is no substitute."
[reply]
[d/l]
[select]
Re: handling different new line characters
by swampyankee (Parson) on Aug 28, 2006 at 16:23 UTC

    If the program is to be used for files resident on a given platform – Windows, Macintosh, *ixen, etc. – chomp manages that transparently. If you need to deal with end-of-record markers over something like NFS, my technique would be to determine the EOR marker for the source machine and reset $/ appropriately for the machine on which the program is running.

    emc

    Only two things are infinite, the universe and human stupidity, and I'm not sure about the former.

    Albert Einstein
[reply]
Re: handling different new line characters
by McDarren (Abbot) on Aug 29, 2006 at 02:28 UTC
    "..if you see a blatent flaw in my logic.."

    Not a flaw in your logic per se, but more a comment on the way you are opening your files.
    It is generally accepted that the three argument form of open is a better choice. That is, instead of doing: 

     open (FH, "< $parent_index") or die "could not open file $parent_inde
+x for reading:$!";

    [download]
    ..you do: 
     open (FH, '<', $parent_index) or die "could not open file $parent_ind
+ex for reading:$!";

    [download]

    The main advantage of this is that it separates the mode from the filename, thereby avoiding any confusion between the two (ie. what if the filename begins with "<" ?). It also forces you to be explicit about which mode you want - again avoiding any possible confusion.

    And if you are using Perl 5.8.0+, you can even do: 

     open (my $fh, '<', $parent_index) or die "could not open file $parent
+_index for reading:$!";

    [download]

    See the documentation for open for further reading, and this topic is also covered quite well in Intermediate Perl, if you happen to have a copy.

    Hope this helps,
    Darren :)

[reply]
[d/l]
[select]
Re: handling different new line characters
by planetscape (Chancellor) on Aug 29, 2006 at 08:39 UTC

    A bit different approach, using an existing utility:

    In the quest for a utility to convert between *nix and DOS newlines on a regular basis, I have found flip: Newline conversion between Unix, Macintosh and MS-DOS ASCII files to be useful. It not only converts between DOS and Unix file formats, but also includes a handy -t command-line switch to tell you which format a given file uses.

    HTH,

    planetscape
[reply]
[d/l]
Re: handling different new line characters
by jaa (Friar) on Aug 29, 2006 at 08:05 UTC
    No comment on the NL problem really, except that reading is usually not a problem in Perl - you have to be more careful if writing files for distribution to *nix or WinDOS. If you are looking at making your code more robust, I would second the previous recommendations on file handling - remember FH style globs are global.

    You might even consider using IO::File which IMHO has an easy peasy and pretty interface, and automatically closes in a kernel resource friendly manner! 

    use IO::File;

my $fh = IO::File->new($filename, "r")
  or die "unable to open $filename: $!";
 
for my $line ( $fh->getlines() ) {
  chomp $line;
  print "[$line]\n";
}

    [download]
    I would also add the comment that your two subs duplicate each other - why not just have one? Or if you really like having the two names then... 
    
sub load_primers {
  return _loadIndex( @_);
}

sub load_parents {
  return _loadIndex( @_);
}

    [download]
    and move all the code into _loadIndex();

    $0.02

[reply]
[d/l]
[select]
Re: handling different new line characters
by chromatic (Archbishop) on Aug 31, 2006 at 01:00 UTC
[reply]

Back to Seekers of Perl Wisdom