Re: [Maxima-discuss] Error in draw package of v.5.34

[Martin K. <kr...@fh...>]

The problem was that I did not correctly downloaded the file from the
repository (I got the html wrapper rather than the file). I fixed this but
the problem with the coma example remains. Does that work for you?

 

(%i38)
apply(plot,append(['damping_ratio(f),xrange=[0,5]],[file_name="D:\\FHB\\Soft
ware\\SMath\\SMath Skript\\SMath\\Maxima\\p5"]));

draw2d (explicit): non defined variable

#0: plot(f='damping_ratio((s+a)/(s^3+a*(s^2+1))),opts=[xrange =
[0,5],file_name = "D:\\FHB\\Software\\SMath\\SMath
Skript\\SMath\\Maxima\\p5"])(coma.mac line 546)

-- an error. To debug this try: debugmode(true);

(%i40) plot('damping_ratio(f),xrange=[0,5]);

draw2d (explicit): non defined variable

#0: plot(f='damping_ratio((s+a)/(s^3+a*(s^2+1))),opts=[xrange =
[0,5]])(coma.mac line 546)

-- an error. To debug this try: debugmode(true);

 

Best regards, Martin

Von: Gunter Königsmann [mailto:gu...@pe...] 
Gesendet: Donnerstag, 20. November 2014 11:39
An: maxima-discuss
Betreff: Re: [Maxima-discuss] Error in draw package of v.5.34

 

I assume the cause of this error message is that your system still contains
a draw.o file from a previous maxima version.

Just delete the folder c:\Users\<your user name>\maxima\binary and
everything should work fine: There is a fix for this problem in the current
git version of maxima so this step won't be necessary any more in the next
release.

Kind regards,

     Gunter.



On 20.11.2014 09:27, Martin Kraska wrote:

Based on Maxima 5.34.1 on Win 7 updating gnuplot.lisp leads to this error
message:

(I modified maxima.bat such that the temp dir is in the install dir to allow
for co-existing installations)

 

(%i17) load(coma);

"coma v.1.70, (Wilhelm Haager, 2014-09-07)"

 

Compiling /FHB/Software/SMath/2014 11 16 SMath and
Maxima/Maxima-5.34.1/share/maxima/5.34.1/share/draw/grcommon.lisp.

End of Pass 1.  

End of Pass 2.  

OPTIMIZE levels: Safety=2, Space=3, Speed=3

Finished compiling D:/FHB/Software/SMath/2014 11 16 SMath and
Maxima/Maxima-5.34.1/Temp/user/binary/binary-gcl/share/draw/grcommon.o.

Loading D:/FHB/Software/SMath/2014 11 16 SMath and
Maxima/Maxima-5.34.1/Temp/user/binary/binary-gcl/share/draw/grcommon.o

Finished loading D:/FHB/Software/SMath/2014 11 16 SMath and
Maxima/Maxima-5.34.1/Temp/user/binary/binary-gcl/share/draw/grcommon.o

Compiling /FHB/Software/SMath/2014 11 16 SMath and
Maxima/Maxima-5.34.1/share/maxima/5.34.1/share/draw/gnuplot.lisp.

loadfile: failed to load D:/FHB/Software/SMath/2014 11 16 SMath and
Maxima/Maxima-5.34.1/share/maxima/5.34.1/share/draw/draw.lisp

-- an error. To debug this try: debugmode(true);

 

Martin Kraska

 

Von: Mario Rodriguez [mailto:bio...@te...] 
Gesendet: Mittwoch, 19. November 2014 23:32
An: Gunter Königsmann; Wilhelm Haager; max...@li...
Betreff: Re: [Maxima-discuss] Error in draw package of v.5.34

 

 

El 19/11/14 a las 17:18, Gunter Königsmann escribió:

I love maxima on android: this program allowed me to run tests when being
away from my keyboard. Anyway:

This was caused by my request, I'm afraid of telling:
It was hard to convince my colleagues to use maxima as long as on hitting a
variable that was not given a value in an explicit() draw directive did
cause the program to halt for about 30 minutes before it ended with an out
of memory: plotting experimental functions to see if they contain typing
errors seems a natural thing to do. The fix was to test if the function that
is to be plotted free of undefined variables.
Maxima is IMHO now much more usable now. Seems like that test sometimes
returns false positives now. Not being a lisp wizard I don't know how to fix
that, too.

Kind regards,

    Gunter.

 

On 19. November 2014 16:10:36 MEZ, Wilhelm Haager
<mailto:wil...@ht...> <wil...@ht...> wrote: 

Hi,
 
In the new version Maxima 5.34 draw2d is obviously erroneous:
The routine produces an error message, when a delayed evaluation
is necessary, i.e. when the function to be plotted has to be
evaluated for each plotted point separately.
 
For example:
 
   rightest_zero(p):=lmax(map(realpart,map(rhs,allroots(p))));
   draw2d(explicit('rightest_zero(x**4+a*x**2+5*x**2+x+1),a,0,10));
 
produces an error message, whereas
 
   plot2d('rightest_zero(x**4+a*x**2+5*x**2+x+1),[a,0,10]);
 
works fine. Also after replacing the draw routines of V 5.34
by the routines of V 5.31 (and subsequent compilation)
draw2d works correctly.
 
Wilhelm Haager



I have commited a fix to the repository.

Could you copy the new version of 'gnuplot.lisp' to your local share/draw
folder and test it?

Thanks.

--
Mario