Re: [C++ Patch] PR 54191

[Paolo Carlini <paolo dot carlini at oracle dot com>]

On 08/27/2012 08:04 PM, Jason Merrill wrote:
> On 08/27/2012 01:36 PM, Paolo Carlini wrote:
> > Thanks for the feedback. Indeed, as I tried to explain in an older
> > message, that was also my understanding. The practical difficulty with
> > the mapping (which you can also guess from the current structure of
> > get_delta_difference_1, the if (kind == bk_inaccessible || kind ==
> > bk_ambig) check) is that ba_quiet, as used by lookup_base, doesn't just
> > control the error calls - like normally tf_error does - it *also*
> > changes the return value, error_mark_node or NULL_TREE. I'm still not
> > sure about the correct (and concise! see get_delta_difference_1, again)
> > way to handle this...
>
> I think the right way is to adjust the callers to handle 
> error_mark_node return properly.  Macros may need to become functions.
Ah, ok. I guess the prospect of touching those macros was somehow 
psychologically "blocking" me., eh ;) But actually, all in all, the code 
doesn't become more complex, because only the cases where we are now 
passing tf_none require checking for error_mark_node too where we were 
simply checking != NULL_TREE as return value of lookup_base and most of 
the calls are otherwise already Ok, or even become simpler, where we 
were, on a case by case basis, tweaking to | ba_quiet basing on complain 
(note: I left DERIVED_FROM_P alone, because we weren't passing ba_quiet 
to it)

The below passes testing on x86_64-linux.

Thanks!
Paolo.

//////////////////////////

Attachment: patch_54191_2
Description: Text document