On 11/07/2011 09:39 PM, Jason Merrill wrote:
On 10/28/2011 08:10 PM, Paolo Carlini wrote:
I reverted it. Had inadvertently tested with checking disabled, the
problem with checking enabled happens earlier than that.
Was there a problem with the patch? It still looks correct even if it
doesn't fix this testcase, so there's no need to revert it.
Oh, thanks for paying attention to that. Sure, I can reapply it, without
the testcase, however, because it fails when checking mode is enabled,
because, as I briefly mentioned, the problem with an unwanted NAMESPACE
node happens earlier in case COMPONENT_REF.
By the way, for the ice-on-invalid side of these issues - I'm not
talking about the non-trivial reject-valid ones - I'm still wondering
whether we couldn't reject the code much earlier (if you remember, at
some point, I tried to even patch the parser, and apparently things were
working, personally I wasn't able to construct snippets wrongly rejected)
Paolo.
