gnuplot development

Hello,

When passing scripts to GNUPlot, I find that it does not understand
things like using 3,1415 instead of 3.1415, so I convert all such
numeric expressions to the 'C' locale style.

This leads me to a more general question: is there a standard for
character encoding of strings in GNUPlot?  Does it depend at all on the
selected locale, or is it always iso-8859-1 or UTF-8 or so?  Does it
depend on the terminal used for output?

Cheers,
John

> 
> From: Petr Mikulik <mikulik@...>
> Date: 2005/11/27 Sun PM 12:48:25 EST
> To: Ethan A Merritt <merritt@...>
> CC: gnuplot <gnuplot-beta@...>, 
>         Daniel J Sebald <daniel.sebald@...>
> Subject: Re: Question: intent of "set datafile binary"?
> 
> > So far as I can tell, the command
> >    set datafile binary
> > does not accomplish anything at all.
> >
> > I had assumed it was intended to make subsequent input default
> > to binary, but perhaps not.
> >
> > That is, I expected the two following sequences would produce
> > the same result.  But they don't.  The second sequence tries
> > to read the input as ascii.
> >
> > # All in-line
> > plot 'foo' binary with image
> >
> > # Set defaults
> > set datafile binary
> > plot 'foo' with image
> 
> I think the command "set datafile binary" should not switch to binary 
> reading because there is no an opposite switch like "plot 'bla' ascii".
> 
> I think it should behave this way:
> gnuplot> set datafile binary
>  			     ^
>  			option expected
> 

I've no strong opinion on this.  Could program either way.

> 
> BTW: "show datafile binary" says
>    "int64" (8) -- processor does not support this size
>    "uint64" (8) -- processor does not support this size
> but I think they are supported as "long long int" on i386.

I guess I did see the syntax "long long" the other day.  (Personally, I'd never expect such thing in a programming language.  The imprecision of it is that a programmer could easily jump to the conclusion that long long just means long.)  I can fix that.

Dan

On Tue, Nov 29, 2005 at 11:36:25AM +0100, Hans-Bernhard Broeker wrote:
> gnuplot is not a "web extension" (whatever exactly that may be supposed=
=20
> to mean).  It's an interactive program, which can also be used as a=20
> back-end by other programs if you redirect its input and/or output.  If=
=20
> that's not information for you, I'm afraid gnuplot isn't what you need.

    However it would be quite interesting to have gnuplot's core
function available as a library. I am afraid this is not the current
policy but it would indeed be interesting to be able to use gnuplot's
powerfull plotting engine in other programme (think SciPy for instance).

--=20
    Ga=EBl

Hi, I am a bioinformatics student from India,
I have an assignment to develop a web-based graph plotting program.
I would like to how can I implement  the gnuplot as a web extension, any 
seperate lib/modules are available. If you could provide some light on the 
same, it would be  of great help. please tell me about any sample cgi 
scripts also. 

 

Mr. Shameer Khadar
Bioinformatics & Systems Biology Group
Dr. D.Y. Patil Institute for Biotechnology & Bioinformatics
Plot 15 / Sector 50 - CBD Belapur - Navi Mumbai - 400614 - Maharashtra
INDIA <skhadar@...>
 ------------------------------------------------------------------------
Imagination is important than knowledge - A Einstein
 ------------------------------------------------------------------------

So far as I can tell, the command
    set datafile binary
does not accomplish anything at all.

I had assumed it was intended to make subsequent input default
to binary, but perhaps not.

That is, I expected the two following sequences would produce
the same result.  But they don't.  The second sequence tries
to read the input as ascii.

# All in-line
plot 'foo' binary with image

# Set defaults
set datafile binary
plot 'foo' with image


-- 
Ethan A Merritt
Biomolecular Structure Center
University of Washington, Seattle 98195-7742

On Nov 27, 2005, at 05:09, gnuplot-beta-request@...  
wrote:

>
> Message: 2
> Date: Sat, 26 Nov 2005 09:47:37 -0500
> From: Jeremy Conlin <jeremit0@...>
> To: gnuplot-beta@...
> Subject: Error using Aqua Term with gnuplot 4.1
>
> I have an Apple PowerBook running OS 10.4.  I just installed gnuplot
> version 4.1 because I need to plot errorbars with histeps.  I only had
> one problem with the installation but was able to get around it.  The
> problem was in the LaTeX tutorial and I simply disabled it.  Now when
> I run gnuplot, every time it tries to plot, Aqua Term opens but a
> window never opens and a plot is never displayed.  I finally get an
> error that says:
>
> 2005-11-26 09:00:16.828 gnuplot[2941] *** Uncaught exception:
> <NSInvalidReceivePortException> connection went invalid while waiting
> for a reply
> Trace/BPT trap

Just FYI, this has been resolved off-list as Jeremy contacted me  
directly too.

In case the problem pops up again, here is an explanation and solution:

The problem was caused by having two versions a AquaTerm/libaquaterm  
installed in different places, one installed by a package-manager  
(Fink) and the other by the official installer from  
aquaterm.sourceforge.net.

The Fink version of libaquaterm got picked up by gnuplot's configure  
since Fink adds itself at the beginning of the path, which is not a  
problem in itself, but the env var AQUATERM_APP was used to  
explicitly select *the other* installment of AquaTerm at runtime.

The solution is to either 1) unset AQUATERM_APP and go with the Fink  
version, or 2) remove/hide Fink's version of libaquaterm from configure.

/Per

I have an Apple PowerBook running OS 10.4.  I just installed gnuplot
version 4.1 because I need to plot errorbars with histeps.  I only had
one problem with the installation but was able to get around it.  The
problem was in the LaTeX tutorial and I simply disabled it.  Now when
I run gnuplot, every time it tries to plot, Aqua Term opens but a
window never opens and a plot is never displayed.  I finally get an
error that says:

2005-11-26 09:00:16.828 gnuplot[2941] *** Uncaught exception:
<NSInvalidReceivePortException> connection went invalid while waiting
for a reply
Trace/BPT trap

Can anyone help me with this?  What do I need to do to get this to work?
Thanks,
Jeremy

ps.  Please forgive if this gets posted twice.

I am now getting error build errors from the latex tutorial
subdirectory:

Making all in tutorial
make[2]: Entering directory `/home/merritt/cvs/gnuplot-cvs/tutorial'

set xtics ('$-\pi$' -pi, '$-\frac{\pi}{2}$' -pi/2, "0" 0, '$\frac{\pi}{2}$' 
pi/2, '$\pi$' pi)
                                               ^
"eg5.plt", line 14: invalid expression

make[2]: *** [eg5.tex] Error 1



I suspect the culprit is this patch:

2005-11-24  Juergen Wieferink  <wieferink@...>
        * src/util.c (parse_sq) src/scanner.c (scanner) docs/gnuplot.doc:
        Fixed \' in single-quoted strings.

I'm not entirely certain, however.  In any event it broke very recently.

-- 
Ethan A Merritt
Biomolecular Structure Center
University of Washington, Seattle 98195-7742

>> $ ident `which gnuplot`
>> /usr/local/bin/gnuplot:
>>     $Header: /cvsroot/gnuplot/gnuplot/term/tgif.trm,v 1.36 2005/08/07 
>> 09:43:34 mikulik Exp $
>
> This looks to be a string printed in every output file of "set term tgif". 
> This message makes no sense there.
>
> I propose it is replaced by the same message which appears in postscript 
> files: fprintf(...,
>  "Creator: gnuplot %s patchlevel %s\n", gnuplot_version, gnuplot_patchlevel)

OK, I've changed that in cvs.

---
PM

Petr Mikulik wrote:

>>> enabling HIDDEN3D_QUADTREE.
[...]
> Unfortunately it is a compile-time switch.
> If it works, and gives good results, then make it as default.

I will.

>> The only real alternative would be to a full-featured, 
>> object-precision 3D rendering engine to replace both pm3d (polygons) 
>> and hidden3d (wireframe) output.  Such a beast would probably be even 
>> slower than hidden3d already is.

> Do we really want (need) a full virtual reality engine in gnuplot?

It doesn't have to be a VR engine (no interaction or real-time 
efficiency strictly needed), but yes, to resolve the problems inherent 
in the naive approach to 3D sorting currently being discussed for pm3d 
here, it would need to be quite a complete 3D engine.  Worse yet, I 
don't think any such renderer exists yet --- pretty much nothing out 
there is done in object precision (i.e. returns results in terms of real 
drawable objects, not just coloured screen pixels), these days.

Whether we *want* such a beast inside gnuplot, is the tricky question. 
Sorting pm3d polygons as if they're points will create nasty artifacts, 
and they'll be far from rare.

As I said to Ethan before: if you you want to hide non-polygon stuff, 
hidden3d is the way to do it.

What about embedding all these new commands into a single "set" command, 
e.g.

set allterms resolution 300
set allterms size 5, 7

so that when setting a new terminal afterwards, such as "set term png", it 
would take the above values as default.

---
PM

> In the pm3d.dem example "color lines: 'splot sin(y)/(y) with linespoints 
> palette'" are the points as well as the lines supposed to be hidden?  In 
> general, the points appear to be hidden, but when zooming about sometimes I 
> see the points come through.

I think it would be sufficent there to plot "with lines" only.

---
PM