gnuplot and powerdevs

[Ezequiel]

Hi all,

I'm a newbie with powerDEVS. I'm experiencing some problems with results plotted in gnuplot. At first, results plotted in gnuplot window closed after the simulation had finished. Then I realized that it was due to the wxt option in the gnuplot script file, so I modified the code of gnuplot block replacing the line strcpy(buff,"\nset terminal wxt persist") by strcpy(buff,"\nset terminal qt persist").
Now I'm having another issue related to that. Buttons in the gnuplot window, such as zomm in/out or grid on/off, do nothing. After a bit of googling I've discovered that it could be caused by the persist option in the script, because since gnuplot closes after simulation, the results window remains open but inactive.

I'm using PowerDEVS 2.4rev981 on Ubuntu 16.04 and gnuplot 4.6.6-3.

[Franz Preyser]

Hi Ezequiel!
I have made the following experience:
With Kubuntu 16.04 I have to use:
strcpy(buff,"\npause -1"); // use this line in kubuntu 16.04

with Kubuntu 14.04 I have to use:
strcpy(buff,"\nset terminal wxt persist"); // use this line in kubuntu 14.04
strcpy(buff,"\npause -1");

Then zooming and everything else worked for me.

[Matias Bonaventura]

In my old Ubuntu 12 I remember I had to install gnuplot-x11 for the wxt persist to work

sudo apt-get install gnuplot-x11

Cheers!

[Federico Bergero]

Sadly we have not found a solution that works on all platforms. Different version of ubuntu have different tweaks for the gnuplot problem.

[Ezequiel]

Hi,

Both solutions that Franz mentioned worked for me. I've chosen the second one, since it implies adding a single line to the original gnuplot block code.

Thank you very much for all your replies!

[Matías Aereal Aeón]

I'm having similar issues and could not find a way to fix it.

When compiling any model that has gnuplot involved, I get a PowerDevs - Message window with the following message: The process terminated abnormally.
Unknown signal 0. After that a GnuPlot window (with id: 0) appears, and everytime I close it, respawns a new one until I have to manually kill the gnuplot process.

When running the model via command line, I get the following output:

matt@ubu:~/workspace/powerdevs/output$ ./model -tf 100
Creating model... 
Model Created, start running ... 
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: Mismatch between the program and library build versions detected.
The library used 3.0 (wchar_t,compiler with C++ ABI 1009,wx containers,compatible with 2.8),
and your program used 3.0 (wchar_t,compiler with C++ ABI 1010,wx containers,compatible with 2.8).
Segmentation fault (core dumped)
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]
Warning: empty y range [0:0], adjusting to [-1:1]

^ and this goes on and on until I manually kill the gnuplot process.

Apparently the mismatch warning does not affect anything for what I have read online, still don't know why I'm getting that segfault and warnings.

I'm running Ubuntu 16.10, 64 bits, in a VirtualBox machine.
GnuPlot version 5, patchlevel 4.
I'm using a slightly modified version of PowerDEVS. I'll come back with more information with the original code from here.

Any ideas?

[Franz Preyser]

Hi!
This may have nothing to do with the GnuPlot-block. They are using void-Pointer in connection with Event-instances to send messages. Thereby they are doing some ugly type - casts between double* , double[], and vector*. Maybe there is something wrong....