Re: perl 5.22.0 blockers, March 22nd

On Mon, Mar 23, 2015 at 6:28 PM, Dave Mitchell <dav...@iabyn.com> wrote:

> PERL_OP_PARENT is most definitely *not* intended to be a supported> configuration for 5.22. It will become so (and maybe considered to become> enabled by default) at such point when most (for some definition of> "most") CPAN modules that mess with op_sibling have been modified to> handle it.>> But I *would* like any new API macros and functions related to the changes> in op_sibling to get into 5.22 if possible.>> (I've been away the last few days; I'll reply properly to your op_sibling> posts properly when I've had time to read and digest them)>

Any suggestions as to what I should change for the following code (used in
Syntax::Feature::QwComments and ::Loop)?

STATIC OP* remove_sub_call(pTHX_ OP* entersubop) {
#define remove_sub_call(a) remove_sub_call(aTHX_ a)
   OP* pushop;
   OP* realop;

   pushop = cUNOPx(entersubop)->op_first;
   if (!pushop->op_sibling)
      pushop = cUNOPx(pushop)->op_first;

   realop = pushop->op_sibling;
   if (!realop || !realop->op_sibling)
      return entersubop;

   pushop->op_sibling = realop->op_sibling;
   realop->op_sibling = NULL;
   op_free(entersubop);
   return realop;
}