Front page | perl.perl5.porters |
Postings from June 2016
On Sun, Jun 05, 2016 at 08:39:20PM +0200, Vincent Pit (VPIT) wrote:
> I just have a small question regarding commit ded730eb9 ("get su_init/su_pop
> working under 5.23.8"). In su_localize, you added this snippet :
>
> =======================================================================
>
> +
> +/* new perl context implementation frees savestack *before* restoring
> + * PL_curcop. Temporarily restore it prematurely to make gv_fetch*
> + * looks up unqualified var names in the caller's package */
> +#ifdef SU_HAS_NEW_CXT
> + COP *old_cop = PL_curcop;
> + PL_curcop = CX_CUR()->blk_oldcop;
> +#endif
> +
> #ifdef gv_fetchsv
> gv = gv_fetchsv(sv, GV_ADDMULTI, t);
> #else
> @@ -896,13 +926,16 @@ static void su_localize(pTHX_ void *ud_) {
> const char *name = SvPV_const(sv, len);
> gv = gv_fetchpvn_flags(name, len, GV_ADDMULTI, t);
> #endif
> +#ifdef SU_HAS_NEW_CXT
> + CX_CUR()->blk_oldcop = PL_curcop;
> +#endif
>
> =======================================================================
>
> but as far as I can tell this old_cop variable is never used. Since the
> tests pass with this change, does it mean that it can be removed?
Hmmm... the second chunk of that diff was *supposed* to be
PL_curcop = old_cop;
which would make it logically correct, although I *think* that in
practice, since su_pop() frees all items in the current savestack frame
below itself before calling su_localize(), so su_localize() is effectively
always called after all the other savestack items have been processed, its
hard to think of test case where it matters. I'd recommend still changing
it to PL_curcop = old_cop rather than removing it though.
--
A major Starfleet emergency breaks out near the Enterprise, but
fortunately some other ships in the area are able to deal with it to
everyone's satisfaction.
-- Things That Never Happen in "Star Trek" #13
Thread Previous
|
Thread Next