«

I am speak in poor English

Sorry ..

The characters in the innermost parentheses, I want to extract

(( A ) aa) <- extract A
(bb (B)) <- extract B

(A (AB) (BC)) <- extract AB and BC

"NamSa" <> wrote in message
news:c36cb0c9-46ca-487d-a049-...

> The characters in the innermost parentheses, I want to extract
>
> (( A ) aa) <- extract A
> (bb (B)) <- extract B
>
> (A (AB) (BC)) <- extract AB and BC

my $text = "(A (AB) (BC))";
my @extracted = $text =~ /\(\s*([A-Z]+)\s*\)/g;
print "@extracted";

That works for all the examples you have given...

Bigus

In article
<c36cb0c9-46ca-487d-a049->,
NamSa <> wrote:

> I am speak in poor English
>
> Sorry ..
>
> The characters in the innermost parentheses, I want to extract
>
> (( A ) aa) <- extract A
> (bb (B)) <- extract B
>
> (A (AB) (BC)) <- extract AB and BC
>

Suggestions:

1. Use Text::Balanced
2. Read 'perldoc -q balanced'

--
Jim Gibson

On Fri, 1 May 2009 09:20:13 -0700 (PDT), NamSa <> wrote:

>
>I am speak in poor English
>
>Sorry ..
>
>The characters in the innermost parentheses, I want to extract
>
>(( A ) aa) <- extract A
>(bb (B)) <- extract B
>
>(A (AB) (BC)) <- extract AB and BC

A single regex could do it.

-sln

output:

'AA'
'B'
'A,B'
'BC'

---------------------------------
## Capture.pl
## (will capture all inner parenth characters stripping off enclosing white space)

use strict;
use warnings;

# Shrunk down regex:
# /\(\s*((??!\s+\))[^()])+)\s*\)/

while (<DATA>)
{
if (my @array = $_ =~
/
\( # Find first open parenth '('
\s* # Trim zero or more white space
( # Start Capture group 1

(?: # grouping

(?!\s+\)) # look ahead, cannot be 1 or more white space followed by ')' char
[^()] # all is ok, grab a character that is not '(' nor ')'

)+ # end grouping, do 1 or more times

) # End Capture group 1, (done only once)
\s* # Trim zero or more white space
\) # Find very next closing parenth ')'
/xg)

{
print "'$_'\n" for (@array);
}
}

__DATA__

(( AA ) aa) <- extract A
(bb (B)) <- extract B

(A (A,B ) (BC)) <- extract AB and BC

( )