Re: '__gnuplot_has_feature__' undefined

[Daniel J Sebald]

On 09/02/2016 10:43 AM, Michael Godfrey wrote:
> Rik,
>
> The fewer versions supported the better.  At some point a final version
> should be chosen so that more work can be done on the fltk and qt systems.
>
> Michael

You're making it sound as though gnuplot support has been a drain on 
resources for the OpenGL class of drivers.  I don't see it that way, as 
those working on upkeep for gnuplot aren't necessarily inclined to spend 
effort on Qt graphics.  Had some of the fixes that existed for gnuplot 
for years been applied all along, it wouldn't have gotten so far out of 
function.

For a large group of people, what's important is the quality of the 
"printed" plot, i.e., the crispness of lines, fonts, etc., the 
flexibility of output formats for use in manuscripts, etc.  If fltk/qt 
with its faster, more seamless graphics could supplant that it would be 
fine to commit to one graphics engine, but that doesn't seem to be the 
case.

The strategy of creating a bitmapped plot and then using another tool to 
convert that to whichever format regardless of whether that format 
supports vector graphics is almost a non-starter for me.  Then there is 
the problem that the Qt graphics has some issues that look to be related 
to threading, but I don't know the design principle of that interchange. 
 If I knew the concept were solid and it was a matter of just going in 
to find a bug here and there, that's one thing.  But if it's a case of 
trying to find a workaround for some structural problem, one has to 
start evaluating effort against quality of output.

Then there is the fact that there are still bugs downstream in the 
OpenGL drivers.  I've gone through a lot of effort to build/fix some of 
those, but trying to get fixes accepted into the OpenGL code isn't an 
easy task.  I can at least have some influence on bug fixes in gnuplot.

Dan


> On 09/02/2016 04:22 PM, Rik wrote:
> > Of course, one of the bigger questions is whether we should even be trying
> > to support multiple versions of gnuplot.  In particular, we are having
> > trouble with the 4.X branch of gnuplot and there has a suggestion to drop
> > support for it in this release (https://savannah.gnu.org/bugs/?48033).  If
> > we did that, I wouldn't be worrying about testing conformance with so many
> > different versions.