On Wed Aug 10 19:50:01 2016, sprout wrote: > But rv2gv can sometimes return &PL_sv_undef. To avoid a change in > behaviour, I think a better fix would be to use > > if (PL_last_in_gv && PL_last_in_gv != &PL_sv_undef) { > > in mg.c (and to leave the assertion as-is, since having PL_last_in_gv > set to anything else is a bug we want to be alerted about). With that change (with a cast to make the pointers the same type), should this from pp_readline be simplified?: PL_last_in_gv = MUTABLE_GV(*PL_stack_sp--); if (PL_last_in_gv == (GV *)&PL_sv_undef) PL_last_in_gv = NULL; else assert(isGV_with_GP(PL_last_in_gv)); Tony --- via perlbug: queue: perl5 status: open https://rt.perl.org/Ticket/Display.html?id=128263Thread Previous