Version 1.3 has been released

Dear pgfplots users,

we are prode to announce the release of pgfplots 1.3!

Version 1.3 provides
- improvements for two dimensional visualization, among them
axis equal, color bars, nodes near coords, jumps in plots, improved 
description
positioning, reverseable axis directions, simpler alignment of adjacent 
axes,
units and a simplified user interface,

- new three dimensional line, scatter, mesh and surface plots,

- a copy of the automatic pdf externalization library,

- an improved manual enhanced with a lot of pdf cross references.

Versions are available on CTAN at
http://tug.ctan.org/cgi-bin/ctanPackageInformation.py?id=pgfplots
and on sourceforge under
http://sourceforge.net/projects/pgfplots/

Best regards

Your pgfplots team

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support

Gnuplot parametric option

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 

_______________________________________________
Pgfplots-features mailing list
Pgfplots-features@...
https://lists.sourceforge.net/lists/listinfo/pgfplots-features

Re: Gnuplot parametric option

Hello Gabriel,

thank you for your detailed bug report! I will investige the problem. 
Your modifications appear to be a quite good idea...

Best regards

Christian

PS
You suggestion for line plots with gnuplot,
 > set parametric;
 > splot [u=0:5*pi] sin(u),cos(u),2*u/5/pi
draws, indeed, a line plot - but is samples a lattice. It just appears 
to be a line plot because the 'v' variable is constant. If you set
'set table "P.dat"'
you'll see that gnuplot generates 10 isocurves a 100 points each and 
that is considers the output to be a surface...

Gabriel Lera schrieb:
> Dear Christian:
> 
> I have just downloaded the new version of pgfplots, and let me thank you 
> again for the impresive work done!
> 
> I had recently discovered that in the previous version there was a bug 
> with gnuplot's parametric option. Although setting it correctly added 
> "set parametric;" to de gnuplot file, the independent variable was not 
> set to "t", as gnuplot expected. I saw two ways to correct it:
> 
> 1) Use the "raw gnuplot" option (that work ok)
> 2) Change the code in "pgfplotscoordprocessing.code.tex" (previous version):
> 
> % pgfplots code:
> %                plot [x=\pgfplots <at> plot <at> domain] #4;%
> % my version 22/12/2009
>                 \iftikz <at> plot <at> parametric   
>                     plot [t=\pgfplots <at> plot <at> domain] #4;%
>                 \else
>                     plot [x=\pgfplots <at> plot <at> domain] #4;%
>                 \fi
> % my version 22/12/2009
> 
> In the new pgfplots 1.3 somehow things got worse.  Use of the parametric 
> option does nothing on the gnuplot file (My guess is that with the new 
> way of processing options "\iftikz <at> plot <at> parametric" is not set before 
> processing this part of the code).
> 
> The relevant part of the new code is in lines 4148 to 4161 of 
> pgfplotscoordprocessing.code.tex
> 
>             \iftikz <at> plot <at> parametric    set parametric;\fi
>             \ifpgfplots <at> curplot <at> threedim
>                 \if0\b <at> pgfplots <at> should <at> sample <at> LINE
>                     % Samples twodimensionally (a lattice):
>                     % and the isosamples thing confuses me.
>                     set isosamples 
> \pgfkeysvalueof{/pgfplots/samples}\if0\b <at> pgfplots <at> should <at> sample <at> LINE, 
> \pgfkeysvalueof{/pgfplots/samples y}\fi;
>                     splot [x=\pgfplots <at> plot <at> domain] 
> [y=\pgfplots <at> plot <at> ydomain] \pgfplots <at> gnuplotcode;%
>                 \else
>                     % *should* sample a line, but I don't know how.
>                     splot [x=\pgfplots <at> plot <at> domain] \pgfplots <at> gnuplotcode;%
>                 \fi
>             \else
>                 plot [x=\pgfplots <at> plot <at> domain] \pgfplots <at> gnuplotcode;%
>             \fi
> 
> In my opinion, after fixing the problem with the parametric option, some 
> code should be added to set the independent variable to "t"
> 
> For 3D parametric plots gnuplot expect independent variables "u" and "v".
> 
> Best regards,
> 
> Gabriel Lera
> 
> PD: By the way, to "sample a line" in gnuplot you can use parametric 
> mode using just one independent variable:
> 
> set parametric;
> splot [u=0:5*pi] sin(u),cos(u),2*u/5/pi 
> 
> The previous gnuplot code draws the helix of page 78 of pgfplots 1.3 manual.
> 
> 
> ------------------------------------------------------------------------
> 
> ------------------------------------------------------------------------------
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast and easy
> Join now and get one step closer to millions of Verizon customers
> http://p.sf.net/sfu/verizon-dev2dev 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Pgfplots-features mailing list
> Pgfplots-features@...
> https://lists.sourceforge.net/lists/listinfo/pgfplots-features

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 

Re: pgfplots Avira AntiVir warning

Hello Hans-Joerg,

thank you for the note... I am aware of that problem, but unfortunately, 
I can't do anything except to disable the javascript library 
'clickable'. Perhaps I should upload a further manual version which 
doesn't use it...

In any case, it is a false alarm: the manual is virus free! I wrote all 
of the  javascript code in the manual by myselfes and I am pretty sure 
that I don't accidentally write viruses...

I've had the same problem with Avira Antivir some time ago when I had a 
javascript snippet on our web-pages (which was also virus-free). It 
seems the virus heuristics of Avira are quite bad when it comes to 
javascript.

Is it possible to configure Antivir to make an exception for this 
specific file?

Best regards

Christian

Hans-Joerg Schulz schrieb:
> Hi,
> 
> I just want to inform you that Avira AntiVir (running under Windows 7)
> gives a warning when unzipping the file pgfplots.pdf from the latest
> pgfplots release. It says that the signature of the HMTL script virus
> "HTML/Malicious.PDF.Gen" was found and urges me to delete it.
> 
> Maybe the virus heuristics are just overreacting, but I thought I'd
> better let you know...
> 
> Best regards,
> Hans-Jörg.
> 
> --
> Hans-Jörg Schulz
> University of Rostock
> Institute for Computer Science
> 18051 Rostock, Germany
> eMail: hjschulz@...
> www: http://vcg.informatik.uni-rostock.de/~hs162
> phone: +49-381-4987441
> fax: +49-381-4987482
> -----------------------------------------------------

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 

Re: Gnuplot parametric option

Hello Gabriel,

I managed to find the two bugs: it was indeed the case the 'parametric' 
was ignored whereas '/tikz/parametric' works.

So, use '/tikz/parametric' instead of 'parametric' to avoid the problem. 
The developer version now contains a fix to do this automatically.

To solve the problem with the dummy variables, I see two possibilities. 
I am not really sure which I'll take, perhaps you can give me your opinion:

First possibility: use the same names which are used for plot expression 
in pgfplots. These names can be customized within pgfplots using 
'variable=\x, variable y=\y' which will use the gnuplot variables 'x' 
and 'y'.

Second possibility: force the use of 't' (or 'u', 'v') for parametric 
plots.

I'd prefer the first choice because it is more consistent with pgfplots, 
but I'd like to hear your opinion as experienced gnuplot/pgfplots user. 
What do you think? The problem is: if someone has ever used gnuplot with 
parametric plots, he might rely on 't'. Hm.

Best regards

Christian

Gabriel Lera schrieb:
> Dear Christian:
> 
> I have just downloaded the new version of pgfplots, and let me thank you 
> again for the impresive work done!
> 
> I had recently discovered that in the previous version there was a bug 
> with gnuplot's parametric option. Although setting it correctly added 
> "set parametric;" to de gnuplot file, the independent variable was not 
> set to "t", as gnuplot expected. I saw two ways to correct it:
> 
> 1) Use the "raw gnuplot" option (that work ok)
> 2) Change the code in "pgfplotscoordprocessing.code.tex" (previous version):
> 
> % pgfplots code:
> %                plot [x=\pgfplots <at> plot <at> domain] #4;%
> % my version 22/12/2009
>                 \iftikz <at> plot <at> parametric   
>                     plot [t=\pgfplots <at> plot <at> domain] #4;%
>                 \else
>                     plot [x=\pgfplots <at> plot <at> domain] #4;%
>                 \fi
> % my version 22/12/2009
> 
> In the new pgfplots 1.3 somehow things got worse.  Use of the parametric 
> option does nothing on the gnuplot file (My guess is that with the new 
> way of processing options "\iftikz <at> plot <at> parametric" is not set before 
> processing this part of the code).
> 
> The relevant part of the new code is in lines 4148 to 4161 of 
> pgfplotscoordprocessing.code.tex
> 
>             \iftikz <at> plot <at> parametric    set parametric;\fi
>             \ifpgfplots <at> curplot <at> threedim
>                 \if0\b <at> pgfplots <at> should <at> sample <at> LINE
>                     % Samples twodimensionally (a lattice):
>                     % and the isosamples thing confuses me.
>                     set isosamples 
> \pgfkeysvalueof{/pgfplots/samples}\if0\b <at> pgfplots <at> should <at> sample <at> LINE, 
> \pgfkeysvalueof{/pgfplots/samples y}\fi;
>                     splot [x=\pgfplots <at> plot <at> domain] 
> [y=\pgfplots <at> plot <at> ydomain] \pgfplots <at> gnuplotcode;%
>                 \else
>                     % *should* sample a line, but I don't know how.
>                     splot [x=\pgfplots <at> plot <at> domain] \pgfplots <at> gnuplotcode;%
>                 \fi
>             \else
>                 plot [x=\pgfplots <at> plot <at> domain] \pgfplots <at> gnuplotcode;%
>             \fi
> 
> In my opinion, after fixing the problem with the parametric option, some 
> code should be added to set the independent variable to "t"
> 
> For 3D parametric plots gnuplot expect independent variables "u" and "v".
> 
> Best regards,
> 
> Gabriel Lera
> 
> PD: By the way, to "sample a line" in gnuplot you can use parametric 
> mode using just one independent variable:
> 
> set parametric;
> splot [u=0:5*pi] sin(u),cos(u),2*u/5/pi 
> 
> The previous gnuplot code draws the helix of page 78 of pgfplots 1.3 manual.
> 
> 
> ------------------------------------------------------------------------
> 
> ------------------------------------------------------------------------------
> This SF.Net email is sponsored by the Verizon Developer Community
> Take advantage of Verizon's best-in-class app development support
> A streamlined, 14 day to market process makes app distribution fast and easy
> Join now and get one step closer to millions of Verizon customers
> http://p.sf.net/sfu/verizon-dev2dev 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Pgfplots-features mailing list
> Pgfplots-features@...
> https://lists.sourceforge.net/lists/listinfo/pgfplots-features

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 

Re: Gnuplot parametric option

[Fwd: Re: pgfplots Avira AntiVir warning]

gmane.comp.tex.pgfplots

From: Hans-Joerg Schulz <hs162@...>
Subject: Re: pgfplots Avira AntiVir warning
Date: 2010-01-09 12:37:32 GMT

On 08.01.2010 19:40, Christian Feuersaenger wrote:

> Is it possible to configure Antivir to make an exception for this
> specific file?

Yes, it is possible and that's what I did: just click "ignore" _and_
"remember my choice for this file (dangerous)" 

All the best,
Hans-Jörg.

--
Hans-Jörg Schulz
University of Rostock
Institute for Computer Science
18051 Rostock, Germany
eMail: hjschulz@...
www: http://vcg.informatik.uni-rostock.de/~hs162
phone: +49-381-4987441
fax: +49-381-4987482
-----------------------------------------------------

------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev 

_______________________________________________
Pgfplots-features mailing list
Pgfplots-features@...
https://lists.sourceforge.net/lists/listinfo/pgfplots-features

contour plots and giving labels to points added by \addplot table

Hi all,

has somebody thought of adding contour plots to pgfplots, now where 3D graphs
are available?

Then I have a question regarding the labeling or pinning some data points which
where plotted by the \addplot table command. Is it possible to add a label or
pin to a somehow given data point from that table? For example point number 7
shows a maximum and one wants to add a pin/label/node to this point.

It would also be nice to be able to add a node to a drawn plot where the data
come from a table like in

    \begin{tikzpicture}
        \draw (0,0) .. controls (6,1) and (9,1) ..
            node[near start,sloped,above]    {near start}
            node                             {midway}
            node[very near end,sloped,below] {very near end}
                (12,0);
    \end{tikzpicture}

Up to now it just seems to be possible to add a node to the last drawn point.

All the best,
Stefan Pinnow

------------------------------------------------------------------------------
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev

Re: Noch kleine Frage zu pgfplots

 <at> pgfplots-features This (german) post handles the request to set 
specific characters of an input file on an "ignore" list.
Maybe the example is of interest to others as well, so I CC the answer. 
Handle such changes with care!

Hi Daniel,

Es gibt tatsaechlich die Moeglichkeit, einzelne Symbole auf eine ignore 
liste zu setzen.
Und zwar geht das wie folgt:
\begin{tikzpicture}
    \begin{axis}

    \catcode`\(=9
    \catcode`\)=9
    \addplot table[col sep=comma] {P.dat};   

    \end{axis}
\end{tikzpicture}

bei einer Datei, die so aussieht:
(1,2),4,5
(3,4),6,7

Das \catcode`\   ist der Befehl, um sogenannte category-codes fuer ein 
nachfolgendes Zeichen einzustellen. Die Ziffer "9" heisst "Ignored 
Character".

Diese Aenderung gilt nicht nur fuer Dateien, sondern fuer alle 
nachfolgenden Zeichen. Deshalb solltest Du sowas mit Vorsicht geniessen. 
Zum Beispiel waere es nicht so gut, wenn Du ein 't' ausblendest, dann 
kann man nicht mehr 'plot' schreiben. Falls Du die Kommas auch 
ausblendest, wird aus der ersten Datenzeile "1245" weil der Rest 
ausgeblendet wird. Du kannst auch \catcode`\,=10 setzen, dann bekommt 
das Komma den speziellen Charaktercode "10", der "space" bedeuted. Dann 
kannst Du Kommata eingeben und es kommt dasselbe raus, wie wenn Du 
spaces eingibst. Ist aber nicht zu empfehlen, weil Kommas noch woanders 
wichtig sind, daher besser '\addplot table' mit '[col sep=comma]' nehmen 
und nur die runden klammern ausblenden.

Es ist auch zu empfehlen, diese gesamte Aenderung innerhalb von 
geschweiften Klammern zu schreiben:
{
\catcode...
\addplot...
}
Dann hoert diese \catcode Gueltigkeit nach der Endklammer auf.

Ich hoffe, das hilft Dir.

Mit liebem Gruss

Christian

Daniel Wissel schrieb:
> Guten Morgen Christian :)
>
> kleine Frage noch:
> Wenn ich Werte in der Koordinaten-Schreibweise "(x, y)" aus einem File
> lesen will, gibt es Fehler, weil die Schreibweise "x y" (ohne Klammern
> und Komma) erwartet wird.
>
> Wie umschifft man dieses Problem leichtesten?
> Kann man einzelne Symbole im File auf eine "ignore-Liste" setzen?
>
> Danke und liebe Grüße,
> Daniel
>
>   

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com

Dots in ticklabels

Hello pgfplots users,

I think I've found a bug (or maybe a feature) when trying to use dots as 
the final character in tick labels like in the following example :

\documentclass{minimal}
\usepackage{pgfplots}
\begin{document}
\begin{tikzpicture}
        % works fine
	\begin{axis}[ybar, enlargelimits=0.15, legend pos=north west, legend columns=2, bar width=15pt,
								xtick=data, xticklabels={{Mr}, {Mrs}}]
		\addplot coordinates {(0,0.1624) (1,0.1950)};
		\addplot coordinates {(0,0.1831) (1,0.2469)};
		\legend{$\min$, $\max$}
	\end{axis}
\end{tikzpicture}

\vspace{2cm}

\begin{tikzpicture}
        % does not work
	\begin{axis}[ybar, enlargelimits=0.15, legend pos=north west, legend columns=2, bar width=15pt, 
								xtick=data, xticklabels={{Mr.}, {Mrs.}}]
		\addplot coordinates {(0,0.1624) (1,0.1950)};
		\addplot coordinates {(0,0.1831) (1,0.2469)};
		\legend{$\min$, $\max$}
	\end{axis}

\end{tikzpicture}
\end{document} 

The bug only appears when the dot is used at the end, when it is the 
first character or it appears anywhere in the middle it does work.

Sebastian

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com