Front page | perl.perl5.porters |
Postings from February 2000
On Mon, Feb 28, 2000 at 09:32:33AM -0500, John Tobey wrote:
> Run "perl -MO=Terse -e '%x=()'" or "perl -Dx -ce '%x=()'" to see the
> tree structure we're stepping around in.
Ooooh, B::Terse. I'll have to remember that. Much more readable than
-Dx. Thanks.
As long as I'm asking, is there an equivalent to Data::Dumper for C
debugging? I find myself wanting to inspect the internals of SVs and
AVs, but all the inspection stuff is macros.
> Hmmm. By using a single flag (OPf_SPECIAL) in the AASSIGN op, we lose
> the chance to specify for each child on the left-hand side whether it
> uses the hash syntax or not.
No, its okay. OPf_SPECIAL is correctly only set on pseudohashes and
hashes. Arrays and scalars don't get flagged since only OP_RV2HV and
OP_PADHV get hit. I've verified this.
> I don't know how relem got to be 'whatever', but the fact $extra[0]
> survives is consistent with your change in Perl_ck_aassign().
That's what I thought, but its not. OPf_SPECIAL is not set on the op
and @extra is properly treated as an array in aassign().
> > In 28 %$avhv is getting completely cleared.
>
> I would expect @extra to keep its first element here, too. Does it?
> If not, is OPf_SPECIAL actually set?
%$avhv has OPf_SPECIAL set on it, @extra does not, as expected.
@extra is properly handled by aassign, nothing odd hangs around. The
problem is that %$avhv is completely empty by the time it reaches
aassign, which makes me think its being optimized somehow and that
optimization is not aware that its a pseudohash. The odd part is that
av_clear() is not called (before aassign). This is going to be hard
to track down without divine intervention.
--
Michael G Schwern http://www.pobox.com/~schwern/ schwern@pobox.com
But folks don't get me wrong. Jews are not antichrists. Only Karl Marx was for he has created communism.
--Alex Chiu, Immortality Guy
Thread Previous
|
Thread Next