Re: Finding Start/End Position of the Uppercase Substring

Using pos and extracting character positions and according actual offsets (useful to substr) it could look as outlined below. I'm not saying however, that this 'solution' is better (just another possible approach):

use strict;
use warnings;

my @strings = do { local $/; split /\n/, <DATA> };

foreach my $str (@strings) {
    my $ret = offset($str);
    my $substring = substr($ret->[0], $ret->[2][0], $ret->[2][1]);

    print <<"EOT";
$substring
start character: $ret->[1][0]
end   character: $ret->[1][1]
start    offset: $ret->[2][0]
end      offset: $ret->[2][1]

EOT

}

sub offset {
    my $str = shift;

    my $hyphens = 0;
    $hyphens++ while $str =~ /-/g;

    $str =~ /[A-Z]/g and my $pos_start = pos($str);
    $str =~ /[a-z]/g and my $pos_end   = pos($str);

    return [ $str,
             [ ($pos_start - $hyphens), ($pos_end - $hyphens) - 1 ],
             [ $pos_start - 1, ($pos_end - $pos_start) - 1        ]
           ];
}

__DATA__
ccaatTTTGACACACACAGAAgggca
--aatTTTGACACACACAGAAgggca
[download]

outputs

TTTGACACACACAGA
start character: 6
end   character: 21
start    offset: 5
end      offset: 15

TTTGACACACACAGA
start character: 4
end   character: 19
start    offset: 5
end      offset: 15
[download]

Update: fix formatting.

Comment on Re: Finding Start/End Position of the Uppercase Substring Select or Download Code

Replies are listed 'Best First'.
Re^2: Finding Start/End Position of the Uppercase Substring by johngg (Canon) on Jun 25, 2007 at 09:49 UTC
`my $hyphens = 0; $hyphens++ while $str =~ /-/g;` [download] Although the data given seems clean in this regard, your code will give wrong results if there are any hyphens in the string other than leading ones. Capturing zero or more hyphens at the beginning of the string and finding the length of the capture might be safer. `my $hyphens = length $1 if $str =~ m{\A(-)};` [download] The match will always succeed so if there are no leading hyphens the length of the capture will be zero. `$ perl -Mstrict -Mwarnings -le ' > my @strings = qw{--aacgtACG ctgGTTAtga}; > foreach my $str ( @strings ) > { > my $hyphens = length $1 if $str =~ m{\A(-)}; > print qq{$str - $hyphens}; > }' --aacgtACG - 2 ctgGTTAtga - 0 $` [download] Cheers, JohnGG	[reply] [d/l] [select]

Replies are listed 'Best First'.

Re^2: Finding Start/End Position of the Uppercase Substring
by johngg (Canon) on Jun 25, 2007 at 09:49 UTC

my $hyphens = 0; $hyphens++ while $str =~ /-/g;
[download]

Although the data given seems clean in this regard, your code will give wrong results if there are any hyphens in the string other than leading ones. Capturing zero or more hyphens at the beginning of the string and finding the length of the capture might be safer.

    my $hyphens = length $1 if $str =~ m{\A(-*)};
[download]

The match will always succeed so if there are no leading hyphens the length of the capture will be zero.

$ perl -Mstrict -Mwarnings -le '
> my @strings = qw{--aacgtACG ctgGTTAtga};
> foreach my $str ( @strings )
> {
>     my $hyphens = length $1 if $str =~ m{\A(-*)};
>     print qq{$str - $hyphens};
> }'
--aacgtACG - 2
ctgGTTAtga - 0
$
[download]

Cheers,

JohnGG

[reply]
[d/l]
[select]