Front page | perl.perl5.porters |
Postings from March 2008
B::Size failure. Please report there. SVOP's are beasts. Esp. with this GV. See http://www.perlfoundation.org/perl5/index.cgi?optree_guts However with my threaded perl aelemfast is no SVOP, it's a PADOP, so the sv access does not fail. Need to check this with a non-threaded perl, but I have non. But I doubt that is a core failure, I would blame B::Size. It should check if the op_flag is SPECIAL. Only then there's a SV, otherwise a PAD at op_targ. Rather use Devel::Size, which is much better. B::Size should be thrown away. I tried to fix it for recent releases but gave up. $ perl bug52284.pl walkoptree: 0. UNOP (0x681738) leavesub walkoptree: 1. LISTOP (0x681180) lineseq walkoptree: 2. COP (0x680eb8) nextstate walkoptree: 2. LISTOP (0x675358) list walkoptree: 3. OP (0x681140) pushmark walkoptree: 3. OP (0x6810c8) padsv walkoptree: 3. OP (0x681110) padsv walkoptree: 2. COP (0x681308) nextstate walkoptree: 2. BINOP (0x6812e0) aassign walkoptree: 3. UNOP (0x681298) null walkoptree: 4. OP (0x6812c0) pushmark walkoptree: 4. UNOP (0x675328) keys walkoptree: 5. UNOP (0x6811f8) rv2hv walkoptree: 6. OP (0x6811d8) padsv walkoptree: 3. UNOP (0x681250) null walkoptree: 4. OP (0x681278) pushmark walkoptree: 4. OP (0x6811a8) padav walkoptree: 2. COP (0x681480) nextstate walkoptree: 2. BINOP (0x681458) sassign walkoptree: 3. BINOP (0x681430) helem walkoptree: 4. UNOP (0x681410) rv2hv walkoptree: 5. OP (0x681388) padsv walkoptree: 4. UNOP (0x6813e8) null walkoptree: 5. PADOP (0x6813a8) aelemfast walkoptree: 5. OP (0x6813c8) null walkoptree: 3. OP (0x681368) padsv walkoptree: 2. COP (0x681608) nextstate walkoptree: 2. BINOP (0x6815e0) sassign walkoptree: 3. UNOP (0x6815b8) null walkoptree: 4. PADOP (0x681578) aelemfast walkoptree: 4. OP (0x681598) null walkoptree: 3. BINOP (0x681550) helem walkoptree: 4. UNOP (0x681530) rv2hv walkoptree: 5. OP (0x6814e0) padsv walkoptree: 4. SVOP (0x681510) const -- Reini UrbanThread Next