John W. Krahn wrote:
>
> Daniel Liston wrote:
> >
> > Here is a tool I use to unfold long lines in LDAP outputs.
> > It also works on mailboxes that have the Received: lines
> > or Content-*: lines folded.
>
> There are two things in your code that may cause a problem.
>
>
> > #!/usr/bin/perl
> >
> > #syntax: unfold.pl filename > newfilename
> >
> > if ($ARGV[0]) {
> > local $/ = '';
> > open(FILE, "<$ARGV[0]") or die "can't open $ARGV[0]: $!\n";
> > @records = <FILE>;
> > close(FILE);
> > }
> > foreach $record (@records) {
> > $record =~ s/\n\s//g;
> > print "$record\n\n";
> > }
>
> You are removing a newline and a single whitespace character. If your
> input looks like this:
>
> one
> two
> three
>
> Your substitution will result in:
>
> one two three
>
> And if your input looks like this:
>
> one
> two
> three
>
> Your substitution will result in:
>
> onetwothree
>
> However if you change the substitution it will screw up the record
> separator. You should chomp the records when you read from the file and
> substitute all whitespace after the newline with a single space.
>
> #!/usr/bin/perl
>
> #syntax: unfold.pl filename > newfilename
>
> @ARGV or die "usage: $0 filename\n";
>
> my @records = do {
> local ( $/, *FILE ) = '';
> open FILE, $ARGV[0] or die "can't open $ARGV[0]: $!";
> grep chomp, <FILE>;
This will drop the last line of the file if there is no
terminating separator.
grep {chomp, 1} <FILE>;
>
> foreach my $record ( @records ) {
> $record =~ s/\n\s+/ /g;
> print "$record\n\n";
> }
>
>
...and even then the original problem was that the input
file doesn't contain empty lines, just blank ones with
whitespace before the record separator. Setting $/ = ''
won't work in this case, and the record split has to be
coded explicitly. If only you could write
$/ = qr/\n(\s*\n)+/
then all would be well.
Rob
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
