AS 5.8 segfaults -- but only on the second attempt?

BrowserUk has asked for the wisdom of the Perl Monks concerning the following question:

Calling test() the first succeeds. When it is called the second time in the loop, it traps?

I know it is non-sensical code, but it is the smallest reduction of the original that still demonstrates the problem. I just cannot see why?

#! perl -slw
use strict;

our $s = 'testtesttest';
our $t;

sub test{
    $t = $s;
    local *_ = \$t;
    our $i = 0;
    *_ = \substr($t, $i) while s[([a-z])(?{ $i=pos()+1 })][\U$1];
}

test();
print "1 pass:$t";

test() for 1..2;
print "After 2 pass:$t"; ## Never executed.
[download]

Examine what is said, not who speaks.

"Efficiency is intelligent laziness." -David Dunham
"When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller

Comment on AS 5.8 segfaults -- but only on the second attempt? Download Code

Replies are listed 'Best First'.
Re: AS 5.8 segfaults -- but only on the second attempt? by hv (Prior) on May 08, 2003 at 01:20 UTC
There are certainly still some known instabilities introduced by substring references. You might want to browse the bugs database at http://bugs.perl.org for some other examples. For what it's worth, the latest bleading-edge perl doesn't have a problem with this testcase, so there's a good chance it'll also be fixed in 5.8.1. Hugo	[reply]
Re: Re: AS 5.8 segfaults -- but only on the second attempt? by BrowserUk (Patriarch) on May 08, 2003 at 01:49 UTC
Thanks Hugo. That the problem is gone in bleed perl is reassuring. I just created a user account at bugs.perl.org and received the password. It won't let me log in yet, but I'll give it a few. Examine what is said, not who speaks. "Efficiency is intelligent laziness." -David Dunham "When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller	[reply]
Re: AS 5.8 segfaults -- but only on the second attempt? by The Mad Hatter (Priest) on May 08, 2003 at 00:15 UTC
Hmmm. Seems like a bug in 5.8 to me, but I'm not really in a position to give an answer. BTW, I get the same results on v5.8.0 built for i686-linux-thread-multi-ld.	[reply]
Re: AS 5.8 segfaults -- but only on the second attempt? by cciulla (Friar) on May 08, 2003 at 00:47 UTC
Me, too. Also, it causes Vim 6.1 to die a rather horrible death whilst using "perldo". C²	[reply]