Many, many thanks ikegami.
... Nevermind. @- and/or @+ become all wonky if there's a (?&...) inside the (?<...>...).
It seems to me that I have found the reason it becomes wonky: see the node

Strange behavior of @- and @+ in perl5.10 regexps.

The following version of your rel_cap subroutine seems to work: 

pl@nereida:~/Lperltesting$ cat calc510withactions4.pl
#!/usr/local/lib/perl/5.10.1/bin//perl5.10.1
use v5.10;

# Infix to postfix translator using 5.10 regexp
# Original grammar:

#  exp ->   exp [-+] term
#         | term
#  term ->   term [*/] digits
#          | digits

# Applying left-recursion elimination we have:

#  exp  ->   term re
#  re   ->   [+-] term re
#          | # empty
#  term ->   digits rt
#  rt   ->   [*/] rt
#          | # empty

sub rc {
  my $ofs = - shift;

  my $np = @-;
  substr($_, $-[$ofs], $+[$np+$ofs] - $-[$ofs])
}

my $input;
my @stack;

my $regexp = qr{
    (?&exp)

    (?(DEFINE)
        (?<exp>    (?&term) (?&re)
                     (?{ say "exp -> term re" })
        )

        (?<re>     \s* ([+-]) (?&term) \s* (?{ push @stack, $^N }) (?&
+re)
                     (?{ say "re -> [+-] term re" })
                 | # empty
                     (?{ say "re -> empty" })
        )

        (?<term>   ((?&digits))
                      (?{ # intermediate action
                          push @stack, $^N
                      })
                   (?&rt)
                      (?{
                          say "term-> digits($^N) rt";
                      })
        )

        (?<rt>     \s*([*/]) ((?&digits)) \s*
                           (?{  # intermediate action
                                push @stack, rc(1), rc(2)
                            })
                   (?&rt) # end of <rt> definition
                           (?{
                                say "rt -> [*/] digits($^N) rt"
                            })
                  |  # empty
                     (?{ say "rt -> empty" })
        )

        (?<digits>  \s* \d+
        )
    )
}xms;

$input = <>;
chomp($input);
if ($input =~ $regexp) {
  say "matches: $&\nStack=(@stack)";
}
else {
  say "does not match";
}

[download]
Now I can access the attributes of the previous symbols. See the line 
push @stack, rc(1), rc(2)

[download]
Follows an execution: 
pl@nereida:~/Lperltesting$ ./calc510withactions4.pl
2-8/4/2-1
rt -> empty
term-> digits(2) rt
rt -> empty
rt -> [*/] digits(2) rt
rt -> [*/] digits(4) rt
term-> digits(8) rt
rt -> empty
term-> digits(1) rt
re -> empty
re -> [+-] term re
re -> [+-] term re
exp -> term re
matches: 2-8/4/2-1
Stack=(2 8 4 / 2 / - 1 -)

[download]

In reply to Re^2: Backreference variables in code embedded inside Perl 5.10 regexps (try2) by casiano
in thread Backreference variables in code embedded inside Perl 5.10 regexps by casiano

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.