Prev: multiple variables in if condition in mat lab
Next: Paper currency recognition using hidden markov model
From: drgz on 12 Mar 2010 04:54
Is there any way in MATLAB to reduce the length of the line sample, as it is with i.e. Gnuplot (set key samplen 'sample length')?

When I search in the product help I find examples for changing almost everything else, but can't find anything about this.
From: Walter Roberson on 12 Mar 2010 05:06
drgz wrote:
> Is there any way in MATLAB to reduce the length of the line sample, as
> it is with i.e. Gnuplot (set key samplen 'sample length')?
>
> When I search in the product help I find examples for changing almost
> everything else, but can't find anything about this.

Yes, it is _possible_ to change it, but there probably is no special
control for the length (I could figure out if there is, but it's 4 am
and I have to sleep.)

Legends in matlab are implemented by adding a second axis to the figure.
The lines and text are then placed within that axis. Find that second
axis and you can find its child line objects, and you can then change
their XData properties to change the length and positions of the lines.
Or more simply, make the axis smaller by changing its Position
properties, and then set the XLim property of the axis so that the
left-hand side of the lines are cut off. Saves you altering the
individual lines, and if you were wanting the lines shorter you were
probably wanting to change the axis Position anyhow (either because you
wanted a smaller legend box or because you wanted longer titles and it
would be a nuisance to also change the Position properties of all of the
text() objects to match a shorter line length...)
From: Sadik on 12 Mar 2010 07:01
Hi,

Actually you can. But to do that, you have to find the handle of the line object in the legend.

An example:

plot(1:10)
legend('a')
linesInPlot = findobj('type','line'); % The second one is the line we are looking for
get(linesInPlot(2),'XData') % returns 0.1231 0.7385
set(linesInPlot(2),'XData',[0.1231 0.4]) % so that new length < 0.5*original, right?

That should be it.

Best.

"drgz " <syrehue(a)hotmail.com> wrote in message <hncsnb$o11$1(a)fred.mathworks.com>...
> Is there any way in MATLAB to reduce the length of the line sample, as it is with i.e. Gnuplot (set key samplen 'sample length')?
>
> When I search in the product help I find examples for changing almost everything else, but can't find anything about this.
From: Sadik on 12 Mar 2010 07:03
By the way, I am using Matlab 7, so you had better check if it is really the second element of linesInPlot, that is if the handle is given by linesInPlot(2).


"drgz " <syrehue(a)hotmail.com> wrote in message <hnd2vt$6r$1(a)fred.mathworks.com>...
> Is there any way in MATLAB to reduce the length of the line sample, as it is with i.e. Gnuplot (set key samplen 'sample length')?
>
> When I search in the product help I find examples for changing almost everything else, but can't find anything about this.
From: Sadik on 12 Mar 2010 07:09
I am sorry, it seems that my message did not show up for some reason. I am writing again:

An example:

plot(1:10)
legend('a')
linesInPlot = findobj('type','line'); % linesInPlot(2) is the handle to that line
get(linesInPlot(2),'XData') % returns 0.1231 0.7385
set(linesInPlot(2),'XData',[0.1231 0.4]) % so that new length < 0.5*previous

Best.




"Sadik " <sadik.hava(a)gmail.com> wrote in message <hndahp$3ij$1(a)fred.mathworks.com>...
> By the way, I am using Matlab 7, so you had better check if it is really the second element of linesInPlot, that is if the handle is given by linesInPlot(2).
>
>
> "drgz " <syrehue(a)hotmail.com> wrote in message <hnd2vt$6r$1(a)fred.mathworks.com>...
> > Is there any way in MATLAB to reduce the length of the line sample, as it is with i.e. Gnuplot (set key samplen 'sample length')?
> >
> > When I search in the product help I find examples for changing almost everything else, but can't find anything about this.
 |  Next  |  Last
Pages: 1 2
Prev: multiple variables in if condition in mat lab
Next: Paper currency recognition using hidden markov model