print "$_\n" for "'Pugh.Pugh'.Barney.McGrew.Cuthbert.Dibble.Grub"=~/('
+.+?'|".+?"|[^\.]+).?/g
[download]

/('(?:[^'\\]*|\\.)+'|"(?:[^"\\]*|\\.)+"|[^\.]+).?/g
[download]

Thanks a lot for the answer -- this is perfect. The first regex works just fine for me and seems to handle nested "''" and '""' ok too, which is an advantage. You seem like the regex king; the 2nd regex looks very impressive -- erm what does it do?

The second regex is for escaping quotes with '. Like: 'Blah.Blah\'blah'. I'm not sure about it tho...

This format looks like a CSV with a dot as record separator. You could just use Text::CSV_XS:

my $data = "'Pugh.Pugh'.Barney.McGrew.Cuthbert.Dibble.Grub";

use Text::CSV_XS;

my $csv = Text::CSV_XS->new({ sep_char => '.',  quote_char => "'" });
$csv->parse($data) or die "Cannot parse data";
my @fields = $csv->fields;
[download]

--
Ilya Martynov (http://martynov.org/)

Thanks for the answer. This also works great and, although not as fast as the regex method, it is probably easier for regex ludites like myself to get to grips with. Thanks again.

$_ = "'Pugh.Pugh'.Barney.McGrew.Cuthbert.Dibble.Grub";
my @keepers = grep defined $_, /'(.*?)'|([^.]+)/g;
print map "<$_>\n", @keepers;
[download]

-- Randal L. Schwartz, Perl hacker

my @result = $string =~ /^([^\.]\.[^\.])(:?\.([^\.]))*$/;
[download]

Or use split and some after-working:

my ($firstPart, @result) = split(/\./, $string);
$result[0] = $firstPart . "." . $result[0];
[download]

Best regards,
perl -le "s==*F=e=>y~\*martinF~stronat~=>s~[^\w]~~g=>chop,print"

Thanks for the reply. I should have explained that my "Pugh.Pugh.Barney.McGrew.Cuthbert.Dibble.Grub" string was just a typical string that might occur. The 'word.groupings' would actually come anywhere, not just at the start. Thanks again though.

use Text::ParseWords;

my $line = "'Pugh.Pugh'.Barney.McGrew.Cuthbert.Dibble.Grub";
my @words = quotewords('\.', 0, $line);
[download]

print "$_\n" foreach (split(m/\G(([^\.]+)\.\2)|(?:(?!\2\.)\.)/g, "Pugh
+.Pugh.Barney.McGrew.Cuthbert.Dibble.Grub"));
[download]

Thanks for the reply. I guess I should have explained that my "Pugh.Pugh.Barney.McGrew.Cuthbert.Dibble.Grub" string was just an example of a typical string that might occur. The 'word.groupings' could actually come anywhere, not just at the start, so this method won't work for me. Thanks again though.

A very quick and dirty way of doing this would be:

$line = "Pugh.Pugh.Barney.McGrew.Cuthbert.Dibble.Grub";
$line =~ s/\./\n/g;
$line =~ s/([^\n]*)\n/$1./;
print $line;
[download]

This page is intentionally left justified.

Thanks for the reply. I guess I should have explained that my "Pugh.Pugh.Barney.McGrew.Cuthbert.Dibble.Grub" string was just an example of a typical string that might occur. The 'word.groupings' could actually come anywhere, not just at the start, so this method also won't work for me. Thanks again though.

#!perl -w
use strict;

my $file = "anyfile.txt";
my $sep = ';';

unless (open (CSV, $file)){
    die "Error: $!\n";
}
else {
    while (<CSV>){
    next if $. == 1; # kill headline: dirty
    my @list = &ExtractFields($_, $sep);
    print join ":_:", @list;
    } # while
    close (CSV);
} # else
# ------------------------------------------------------------
sub ExtractFields {
    my ($string, $sep) = @_;
    my @csv = &FilterIndexList($string, $sep);
    my $start = 0;
    my @list = ();

    foreach my $j (@csv){
    my $end = $j-1;
#    print "$start-$end  ";
    push (@list, substr($string, $start, $end-$start+1));
    $start = $j+1;
    } # foreach

    # filter leading and trailing "
    foreach (@list){    s/^\"(.*)\"$/$1/;    }
    
#    print join("(_|_)", @list);
    return (@list);
} # ExtractFields
# ------------------------------------------------------------
sub FilterIndexList {
    my ($string, $sep) = @_;
    
    my @sep = &GetIndexList($string, $sep);
    my @hc = &GetIndexList($string, '"');

    # try to find connected " and remove 
    # the positions within from @sep
    my $i = 0;
    foreach (;;){
    my ($start) = grep {$_ == $hc[$i]-1 } @sep;
    if ($start){
        $i++;
        my ($end) = grep {$_ == $hc[$i]+1 } @sep;
        if ($end){
#        print "found at $start-$end: $hc[$i]\n";        
        # kill positions in @sep within $start and $end
        @sep = grep { $_ <= $start or $_ >= $end } @sep;
        $i++;
        }
        else { # invalid end; throw away end and start over again
        splice(@hc, $i, 1);
        $i--;
        }
    }
    else { # invalid begin; throw away start
        splice(@hc, $i, 1);
    }
    last if $i > $#hc; # exit loop if no more " to test
    }
    return (@sep);
} # FilterIndexList
# ------------------------------------------------------------
# Return a list of incices of positions of $sep in $string
sub GetIndexList {
    my ($string, $subStr) = @_;
    my @list = (); 

    my $pos = -1; # startposition
    while (1){
        # search for next $subStr
    $pos = index($string, $subStr, $pos+1);
        # if startposition again or not found, return
    last if $pos == -1; 
        # else push found position onto list
    push (@list, $pos);
    }
    return (@list);
} # GetIndexList
# ------------------------------------------------------------
[download]

Best regards,
perl -le "s==*F=e=>y~\*martinF~stronat~=>s~[^\w]~~g=>chop,print"

To use split, you can reverse the string with the names, only split enough times (6) so that the last string contains the '.', reverse the order of the elements in the foreach loop, then (un)reverse the string when you print it.

my @names = split(/\./, reverse("Pugh.Pugh.Barney.McGrew.Cuthbert.Dibb
+le.Grub"), 6);
foreach(reverse(@names)) {print reverse . "\n";}

__RESULT__
Pugh.Pugh
Barney
McGrew
Cuthbert
Dibble
Grub
[download]

Thanks for the reply. I guess I should have explained that my "Pugh.Pugh.Barney.McGrew.Cuthbert.Dibble.Grub" string was just an example of a typical string that might occur. The 'word.groupings' could actually come anywhere, not just at the start, so this method won't work for me too. Thanks again though.

my $string = (join(".", sort split(/\./, "Pugh.Barney.Pugh.Barney.McGr
+ew.Cuthbert.Dibble.Grub")));
my @array = ($string =~ m/(\S+)\.\1/g); # Capture duplicate
$string =~ s/(\S+)\.\1//g;    # Eliminate duplicates
$string =~ s/\./\n/g;         # Change \. to \n
foreach (@array) {            # Print duplicates
    print "\n$_.$_";
}
print "$string";              # Print the rest of the list
[download]

    $string =~ s/(\S+)\.\1/'$1.$1'/g;    # Add quotes

and plugged into merlyn's reply

$_ = join(".", sort split(/\./, "Pugh.Barney.Pugh.Barney.McGrew.Cuthbe
+rt.Dibble.Grub"));
$_ =~ s/(\S+)\.\1/'$1.$1'/g;      # Add quotes
my @keepers = grep defined $_, /'(.*?)'|([^.]+)/g;
print map "$_\n", @keepers;
[download]