Re: PL/Perl Does not Like vstrings

On 01/04/2012 08:32 PM, Tom Lane wrote:
> Andrew Dunstan<andrew(at)dunslane(dot)net> writes:
>> Tom said:
>>> 2. A slightly cleaner fix for this should be to duplicate the SV and
>>> then release the copy around the SvPVutf8 call, only if it's readonly.
>>> "Fixing" it in do_util_elog is entirely the wrong thing.
>> How do we tell if it's readonly?
> SvREADONLY(sv) macro.
>
>

OK, so this seems to fix The issue David found:

diff --git a/src/pl/plperl/plperl_helpers.h b/src/pl/plperl/plperl_helpers.h
index ac0a97d..f0e1afa 100644
--- a/src/pl/plperl/plperl_helpers.h
+++ b/src/pl/plperl/plperl_helpers.h
@@ -51,7 +51,10 @@ sv2cstr(SV *sv)
/*
* get a utf8 encoded char * out of perl. *note* it may not be
valid utf8!
*/
- val = SvPVutf8(sv, len);
+ if (SvREADONLY(sv))
+ val = SvPVutf8(sv_mortalcopy(sv), len);
+ else
+ val = SvPVutf8(sv, len);

/*
* we use perls length in the event we had an embedded null byte to
ensure

but it doesn't fix the one I found which passes a typeglob to elog():

do '$foo=1; elog(NOTICE,*foo);' language plperl;

That still crashes, but doesn't if we use sv_mortalcopy unconditionally.

cheers

andrew