On 04.09.05, Simson L. Garfinkel wrote:
> >Beside that you'll note, that the styles are not aware of axis brakes
> >currently. This is an open issue in the current graph design. In case
> >you want me to start working on that and first of all thinking about
> >how to solve that in the current design, just do a feature request ...
> I don't actually understand this question.
You probably missed the implicit <ironic> tags here. The point is that
we can add such a feature of broken bars at split axes. I started to
thought about some details during the last days ... and it's really
not that difficult. We just don't have it at the moment and upto now
nobody was in need of such a feature. Still, as soon as I'm told that
at lease somebody really needs that and it's needed not just a single
time (where you could probably code something qick'n'dirty) but again
and again, it's about time to be implement.
> If I was to make a proper
> feature request, it would be for a design document that explains the
> concepts behind the class hiearchy and explains when named parameters are
> passed and when arrays of objects are passed.
Well, the doc is really a problem. Currently I'm thinking about
explaining the details of the examples much better. More cook-book
like. I think currently we don't have enough resoures to start a real
users manual ... but from various discussions I got the impression,
that many features shown (and intended to be shown) by the examples
are just not really recognized. It seems to be, that a few comments in
the source are just not helping a lot ... we need to better explain
things. I'll try to start working in that area before the 0.9 release.
> The second request would be
> for more examples that are based on graphing data, rather than graphing
For that we need some data. And originally we had no concept of
distributing the data together with the examples. This has changed in
the mean ... so it's probably a good idea to better use that feature.
> 2. Sorry that PyX doesn't have anything for 3D bargraphs. I threw something
> together with gnuplot but the results were horrible. Good enough for a
> slide show that I'm doing this week, but still pretty raw.
> Have you thought about having PyX interface to OpenGL? That could make some
> fancy graphs.
Originally we didn't had any prospect for PyX beside PostScript. In
the mean we already did PDF and I do have the vision of a native SVG
support as well. Additionally, something Jörg came up with some time
ago was the idea of a cairo support. I kind of like the idea, but I'm
not at all sure whether we'll ever make it. But anyway, compared to a
OpenGL backend I think something like cairo is much more suitable.
> 3. Yes, I was able to download your PDF from PhysRevB. Did you do all of
> the graphs with PyX?
Sure I did. ;-)
> I would like to look at the source; Your system
> doesn't do hidden line removal, but that is really easy for printing bar
> graphs (just run from back to front.)
I'll send you a ready to run version. BTW such a hidden parts removal
is also done in our 3d example on our webpage. The (very crude)
sources of this example is also somewhere in CVS, but for the plot's I
used some improved version of the code. Still, it's just something for
people to build own 3d plots out of some very basic central projection
functionality. Nothing suitable for general purpose ... we'll need a
general purpose 3d graph geometry for that.
> BTW, I can send you a PDF of your PDF, if you wish.
Thank you a lot, but I don't need to have a copy. Additionally, from
time to time, I do have access to certain journals when I'm a guest at
an institute. I do not worry about not having a copy at hand ...
by _ _ _ Dr. André Wobst
/ \ \ / ) wobsta@..., http://www.wobsta.de/
/ _ \ \/\/ / PyX - High quality PostScript and PDF figures
(_/ \_)_/\_/ with Python & TeX: visit http://pyx.sourceforge.net/