Re: [lmi] Input-sequence anomaly

[Greg Chicares]

On 02/11/2016 11:13 PM, Vadim Zeitlin wrote:
[...]
>  It took me embarrassingly long to understand what was going on here but,
> finally, it's just a bug in wx[*] which was fixed in this commit:

Then your threshold of embarrassment must be awfully low, and you
might consider liberalizing it.

>  Sorry for the inconvenience, but you will have to update wxWidgets once
> again to fix this.

The timing works out well. Kim is going to get a new office computer
next week and will have to rebuild everything from scratch then anyway.

I'm just really glad you were able to resolve this. Even though we
hadn't observed it with msw-7, I figured it must simply have been
masked there, and would manifest itself at some inconvenient moment.

> [*] Initially I suspected a bug in g++, but actually it's in its right to
>     do what it does, it's just not really helpful neither, and I opened
>     https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69777 in an attempt to
>     change this.

I was going to ask one question about this, but it's exactly the question
you posed there.

>     But, to be honest, I don't have a lot of hope that
>     anything is really going to happen here.

I think that prediction is correct, but your suggestion is most reasonable.
I did quite a bit of web searching on __cxa_pure_virtual, and never saw
this gcc oddity mentioned; it's certain to cause someone else misery, and
it would seem so easy to emit a diagnostic instead of placing a landmine.