Re: [lmi] wx assertion failure: invalid font

[Vadim Zeitlin]

On Tue, 9 Oct 2018 21:31:12 +0000 Greg Chicares <address@hidden> wrote:

GC> On 10/2/18 6:41 PM, Vadim Zeitlin wrote:
GC> > On Tue, 2 Oct 2018 17:44:43 +0000 Greg Chicares <address@hidden> wrote:
GC> > 
GC> > GC>  - When I use the original instructions to reproduce the reported 
problem,
GC> > GC> a PDF is created, but some lines look wrong.
GC> 
GC> [tiny letters, weird spacing between words]
GC> 
GC> >  Yes, they do, and I completely failed to notice this even though I did
GC> > examine the generated file after my fix. And, of course, it makes perfect
GC> > sense: if an invalid font was used before, resulting in an assert, now a
GC> > valid but wrong font is used instead. I really should have taken time to
GC> > stop and think about what I was doing, sorry...
GC> > 
GC> >  I'm going to take time to really understand why does the invalid or wrong
GC> > font end up being used and I'll create another PR with a proper fix later.
GC> 
GC> Can the reason be as follows?

 Yes, indeed, this is what I found back on Saturday, but somehow I didn't
understand why did it work or, rather, why the original version didn't.
It's a bit embarrassing to admit this, as in hindsight the reason is
perfectly obvious and I saw it immediately when I wrote a function for
dumping wxHtmlCell contents and used it here, but for some reason I just
didn't realize it before, sorry.

GC> It's just a wild guess that I arrived at by comparing the TAG_HANDLERS
GC> for <P> and <HEADER>. This does seem to prevent the problem for our
GC> reproducible 'sample2gpp' test case.

 Yes, it does, and the reason for it is explained in the commit message and
a comment in https://github.com/vadz/lmi/pull/98


 I'd like to submit some other changes, cleaning up things in the related
code a little and will try to do this a.s.a.p., but for now the fix above
should definitely be already applied.

 Sorry again for taking so long to make this trivial fix,
VZ