develooper Front page | perl.perl5.porters | Postings from August 2001

Re: Custom Ops

Thread Previous
From:
Simon Cozens
Date:
August 27, 2001 05:50
Subject:
Re: Custom Ops
Message ID:
20010827133851.A351@netthink.co.uk
On Mon, Aug 27, 2001 at 02:39:54PM +0200, Arthur Bergman wrote:
> Indeed, and I do not question this. But I find atleast on case that does.
> 
> if (o->op_type == OP_NULL)
>         {
>             Perl_dump_indent(aTHX_ level, file, "  (was %s)\n",
> PL_op_name[o->op_t\
> arg]);
> 
> and I was just wondering if we shouldn't do the right thing here?

Well, this case is right, because it's part of the low-level dumping
stuff. By the time something's been optimized away to a null, the
user is never ever going to see the op. Why? Error messages containing
op_name or op_desc appear either before optimization or at runtime.
Before optimization, we haven't nulled out any ops; at runtime, if
we've nulled out an op, it'll be pp_null and won't generate any
error messages!

Simon

Thread Previous


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About