[prev in list] [next in list] [prev in thread] [next in thread] 

List:       perl5-porters
Subject:    Re: [perl #68192] Tied scalar numeric FETCH corruption after
From:       Dave Mitchell <davem () iabyn ! com>
Date:       2010-05-30 23:30:26
Message-ID: 20100530233026.GU2976 () iabyn ! com
[Download RAW message or body]

On Sun, May 30, 2010 at 02:17:15PM -0700, Father Chrysostomos wrote:
> Since you asked, I had a quick look and saw that you missed readline and 
> glob. I'll fix those, today if possible.

I had deliberately left those two, as the glob version is currently broken
in a major way - <$complex_expression> completely by-passes pp_glob and
calls &CORE::GLOBAL::glob instead:

[davem@pigeon bleed]$ cone -e '<$a[0]>'
1  <0> enter 
2  <;> nextstate(main 59 -e:1) v:{
3  <0> pushmark s
4  <#> gv[*a] s
5  <1> rv2av sKR/1
6  <$> const[IV 0] s
7  <2> aelem sKM/LVDEFER,2
8  <$> const[IV 0] sM
9  <#> gv[*CORE::GLOBAL::glob] s
a  <1> entersub[t3] vKS/TARG,1
b  <@> leave[1 ref] vKP/REFC

So I was intending to discuss at some point with p5p what the correct
semantics of <...> should be as regards overloading.

>
> Also, have you seen ticket 71286? (I don't plan to work on it, but there 
> is a patch there.)

Well, I'm still officially working on tie bugs rather than overload bugs;
I only got involved in overload since the ticket was also tie related.

-- 
Indomitable in retreat, invincible in advance, insufferable in victory
    -- Churchill on Montgomery
[prev in list] [next in list] [prev in thread] [next in thread]