Tried that. No change. The plot is still squished as before.
Another thing - during the execution of the above series of commands, an octplot window opens up with a single plot (just the axes and tics in some default locations), not 6 plots in an array.
Supposing that I just wanted to examine the plots without printing them, would that not indicate that subplot has a bug as well.
Should I file the above two kinds of behaviour as bugs or will you take care of that ?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I tired your example file, replacing the plotcmd's everywhere with plot(1:10) and it works for me.
Perhapse the problem is in the plotcmd's themselves -- can you give me an example of them ?
When filing bug reports please add an example which reproduces the bug. The example should be:
a. as simple as possible still causing the bug
b. self contained -- i.e. I should be able to run it on my system (octave+octave-forge+octplot), so please change your "plotcmd11" to something else which produces the bug
Thanks for trying octplot
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Now the plots are no longer squished but largely expanded - so much so that the first column of plots takes up 75% of the space, while 3/4ths of the second one is clipped off.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think I have figured out what is happening with the bug.
The size of the plot o/p seems to be related to the physical size of the octplot window.
Try the following :
Plot your data. Using the mouse, resize the octplot window. Now, issue the print command. Examine the pdf o/p. You will see that the size of the plot would have changed commensurately with the increase in the physical size.
Needless to say, this a bug. The hardcopy should be related to the size of paper settings, not the size of the octplot window.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
You are right that the printout size should not be conneted to the screen size. Please file this as a bug. It will take time to fix thoug. If you need plottign to file only and the screen does not interest you, I highly recommend gnuplot which is the best tool for this job (beter than matlab IMHO).
Butnow you have me confused:
1. if you resize the window to a reasonable size (i.e. the defualt 400x300 pixels) is the output (both in the window and pdf) OK?
2. if not, please please please provide me with a SELF CONTAINED example which reproduces this bug. I cannto fix what I cannot reproduce.
Shai
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am using toggle_octplot :
xlabel('Vec');
subplot(3,2,1)
plotcmd11=octplotcmd('d11',p11,r11,1,3);
eval(plotcmd11);
subplot(3,2,2)
plotcmd11=octplotcmd('d11',p11,r11,1,4);
eval(plotcmd11);
subplot(3,2,3)
plotcmd12=octplotcmd('d12',p12,r12,1,3);
eval(plotcmd12);
subplot(3,2,4)
plotcmd12=octplotcmd('d12',p12,r12,1,4);
eval(plotcmd12);
subplot(3,2,5)
plotcmd13=octplotcmd('d13',p13,r13,1,3);
eval(plotcmd13);
subplot(3,2,6)
plotcmd13=octplotcmd('d13',p13,r13,1,4);
eval(plotcmd13);
print('-r300','-dpdf','200.pdf');
octplotcmd is just a defined command that builds up the plot command.
The above produces a largely blank page with all 6 plots squished in the bottom left corner. How do I fix it ?
this is indeed a bug. However the fix is simple:
1) locate the print command
at the octave prompt type:
which print
now edit this file and replace the line (shouel be line 114):
octplot_command("print",fn,0);
with
octplot_command("print",fn,1);
(i.e. change 0 to 1)
by the way, the -r300 switch to the print command is not necessary since pdf files are reolution independant -- you can zoom as much as you want.
Tried that. No change. The plot is still squished as before.
Another thing - during the execution of the above series of commands, an octplot window opens up with a single plot (just the axes and tics in some default locations), not 6 plots in an array.
Supposing that I just wanted to examine the plots without printing them, would that not indicate that subplot has a bug as well.
Should I file the above two kinds of behaviour as bugs or will you take care of that ?
I tired your example file, replacing the plotcmd's everywhere with plot(1:10) and it works for me.
Perhapse the problem is in the plotcmd's themselves -- can you give me an example of them ?
When filing bug reports please add an example which reproduces the bug. The example should be:
a. as simple as possible still causing the bug
b. self contained -- i.e. I should be able to run it on my system (octave+octave-forge+octplot), so please change your "plotcmd11" to something else which produces the bug
Thanks for trying octplot
Here is plotcmd11 (others are similar) :
plotcmd11 = plot(d11(1:51,1),d11(1:51,4),'-',d11(52:102,1),d11(52:102,4),'-',d11(103:153,1),d11(103:153,4),'-',d11(154:204,1),d11(154:204,4),'-',d11(205:255,1),d11(205:255,4),'-',d11(256:306,1),d11(256:306,4),'-',d11(307:357,1),d11(307:357,4),'-',d11(358:408,1),d11(358:408,4),'-',d11(409:459,1),d11(409:459,4),'-',d11(460:510,1),d11(460:510,4),'-',d11(511:548,1),d11(511:548,4),'-')
Another quirk :
I just added xlabel and ylabel commands :
subplot(3,2,1)
plotcmd11=octplotcmd('d11',p11,r11,1,3);
eval(plotcmd11);
xlabel('Vec');
ylabel('Ic');
subplot(3,2,2)
plotcmd11=octplotcmd('d11',p11,r11,1,4);
eval(plotcmd11);
xlabel('Vec');
ylabel('Ib');
subplot(3,2,3)
plotcmd12=octplotcmd('d12',p12,r12,1,3);
eval(plotcmd12);
xlabel('Vec');
ylabel('Ic');
subplot(3,2,4)
plotcmd12=octplotcmd('d12',p12,r12,1,4);
eval(plotcmd12);
xlabel('Vec');
ylabel('Ib');
subplot(3,2,5)
plotcmd13=octplotcmd('d13',p13,r13,1,3);
eval(plotcmd13);
xlabel('Vec');
ylabel('Ic');
subplot(3,2,6)
plotcmd13=octplotcmd('d13',p13,r13,1,4);
eval(plotcmd13);
xlabel('Vec');
ylabel('Ib');
print('-r300','-dpdf','200.pdf');
Now the plots are no longer squished but largely expanded - so much so that the first column of plots takes up 75% of the space, while 3/4ths of the second one is clipped off.
I think I have figured out what is happening with the bug.
The size of the plot o/p seems to be related to the physical size of the octplot window.
Try the following :
Plot your data. Using the mouse, resize the octplot window. Now, issue the print command. Examine the pdf o/p. You will see that the size of the plot would have changed commensurately with the increase in the physical size.
Needless to say, this a bug. The hardcopy should be related to the size of paper settings, not the size of the octplot window.
You are right that the printout size should not be conneted to the screen size. Please file this as a bug. It will take time to fix thoug. If you need plottign to file only and the screen does not interest you, I highly recommend gnuplot which is the best tool for this job (beter than matlab IMHO).
Butnow you have me confused:
1. if you resize the window to a reasonable size (i.e. the defualt 400x300 pixels) is the output (both in the window and pdf) OK?
2. if not, please please please provide me with a SELF CONTAINED example which reproduces this bug. I cannto fix what I cannot reproduce.
Shai