Re: [lmi] UTF-16 output from automated GUI test?

[Vadim Zeitlin]

On Thu, 20 Oct 2016 23:15:52 +0000 Greg Chicares <address@hidden> wrote:

GC> However, in reality we want to change from std::wcout to std::cout, right?

 Yes, I think this is the best solution.

GC> But now I'm confused about that, too, because grepping for 'wcout' finds
GC> nothing--maybe 'wcout' is implicitly used by wxPuts().

 Yes, this was an oversimplification on my part: current code uses wxPuts()
and wxPrintf() which use wide character IO. The real goal is to switch to
using non-Unicode IO and the most trivial to achieve it could seem to be
replacing them with puts() and printf(). However this is unappealing in
practice because the latter is unsafe and painful to use with string
objects, so the real plan is to replace (wide) wxPrintf() with (narrow)
std::cout instead.

GC> I'm having an acute sinus attack and can't focus on anything else,

 Sorry to hear about this and hope you get better soon!

GC> so please ignore whatever I say--you know the right thing to do.

 I think I do, I'm just still not sure about the urgency of it. I'll try to
do it relatively soon but won't treat it as absolutely critical unless you
tell me to.

 Thanks,
VZ