Re: [lmi] DataView assertion fires [Was: Automated GUI testing, revisited]

[Greg Chicares]

On 2014-11-13 01:53Z, Vadim Zeitlin wrote:
> On Wed, 12 Nov 2014 21:28:42 +0000 Greg Chicares <address@hidden> wrote:
> 
> GC> On 2014-11-11 23:46Z, Boutin, Wendy wrote:
> GC> [...]
> GC> > benchmark_census: ERROR (Assertion failure: should have non zero page 
> size
> GC> > [file ../src/generic/datavgen.cpp, line 3632, in OnChar()].)
> GC> 
> GC> Vadim--Kim reports exactly the same error. I don't see it myself.
> GC> I'm using msw-xp; they have -7. We're all using wx trunk as of
> GC> the same SHA1 (06ddf44a276524d9e154cbe6c7dd4f3d8442f912).
> 
>  I don't know why I don't see it (I'm testing under Windows 7 too), but the
> assert seems pretty useless anyhow, so I've just removed it from wxWidgets
> in r78146 (2f3969724b0b25d7d8e239fc0d7265446bf508bf). Unfortunately I
> strongly suspect that the real problem is not the assert itself but
> whatever leads to its appearance and the test will still fail -- but at
> least perhaps now we can get more information about why does it fail?

Okay, I've updated wx as of a few minutes ago, and confirmed that it
and lmi both build, and that lmi opens as expected. In a moment I'll
publish new secret decoder ring settings that will enable those who
use msw7 to retest this issue, and perhaps to report more informative
diagnostics.