Hi David: I agree with your assessment of PLplot vs. other plotting
packages. I made the same survey a while back and came to the same
conclusion. PLplot actually has a pretty nice build process (IMHO) and is
a very healthy and active open-source project.
Craig--You are working from a Mac, no? What is the problem getting PLplot
to compile? I don't have much insight into this, but I would bet someone
on the plplot list would.
Regards,
Doug
[email protected]
Software Engineer
UCAR - COSMIC, Tel. (303) 497-2611
On Mon, 8 Feb 2010, David Mertens wrote:
On Mon, Feb 8, 2010 at 12:54 AM, Craig DeForest <[email protected]>
wrote:
There are good reasons to adopt-and-fix something (be it PLPlot or, say,
Gnuplot) that already does high level formatting. There are also
good reasons to write a new interface layer from scratch.
I should be sleeping, but this comment got me to finally looking around the web
to see what's available. The list of C-libraries is smaller than you'd
expect, and there appear to be more Python graphing libraries than C ones. Now
that I've actually looked at it, I would honestly vote that we go with
PLplot. Other viable options that I found include dislin, ncar graphcis,
mathgl, and PGPLOT of course. If you want to take a look, I had decent success
googling (without quotes) 'c plotting library.'
The reason I vote for PLplot is that I believe that it is a pretty healthy
open-source project (i.e. it should stick around for a while), it's LGPL, it's
already got PDL bindings, it has output-to-memory (piddle), and it seems
comparable to the other projects. Dislin is apparently developed by one guy and
has its own license - including a paid license for commercial use - though it
has lots of installers for various systems. Ncar seems to have a spaghetti
of licenses and may be difficult to install, though it's supported by the
National Center for Atmospheric Research, so I don't imagine it's going
anywhere any time soon. MathGL looks really pretty, but it's only developed by
a single developer; it might be a good option for TIMTOWDI. PGPLOT
requires a fortran compiler and cannot be used in a commercial setting.
I don't think we'd want to use Gnuplot because I believe you must communicate
to it through a pipe. That is, there's no low-level interface to Gnuplot,
last I checked. If we're going that route, I'd recommend Asymptote (for which
I've already written a piping module!).
Our only problem with PLplot is getting it to build. That seems to me to be an
easier problem to solve than writing a whole new set of bindings for
Dislin, NCAR, or MathGL.
Just my 2c.
David
_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
