matplotlib-users Mailing List for matplotlib (Page 12)

 I am out of town and won't be able to address this until early next week.
- Charlie
2008年6月12日 John Hunter <jd...@gm...>:
> On Thu, Jun 12, 2008 at 2:58 PM, İsmail Dönmez <is...@na...> wrote:
> > Hi all,
> >
> > I was trying matplotlib 0.98.0 and Qt4 backend works great but Coca
> > backend seems to be broken :
>
> Charlie,
>
> any chance you can try and port cocoaagg over to the new trunk api.
> figure.dpi is no longer a lazy value, but a plain ol number. There
> are probably a few other changes that will need to be made as well.
> I'd like to clear up as many of these problems as we can and shoot for
> a bugfix release next week.
>
> JDH
>

On 2008年6月13日, John Hunter wrote:
> On Thu, Jun 12, 2008 at 9:21 PM, James A. Benson
> wrote:
>>
>> My app uses the wx backend, so I happened to notice
>> that the backend_wx doesn't seem to work with
>> matplotlib 0.98.0. Running the 0.98 example embedding_in_wx.py
>> shows the problem. Only a bit of the lines get plotted. The
>> pan up/down zoom in/out all show various error messages.
>> When i revert back to 0.91.3 and run the same 0.91 example
>> it works fine.
>
> Hey James, we were discussing problems with the wx backend on 0.98
> yesterday on the devel list, and Ken offered to take a look at them,
> so it is possible we will have those fixed for the next bugfix
> release. What wx version are you using?
>
Sorry, I forgot to include that important detail. I
see the problem on a Linux F8 machine with wx version 2.8.7.1
> Note that unless you are running over an x11 connection, there is
> little reason to prefer wx over wxagg, which has better rendering and
> more features. Over an x11 connection, wxagg is slow because it has
> to pass the who bitmap over the wire, but otherwise it works great.
>
> JDH
>
Thanks for the advice and later comments on this thread from
Michael. Sounds like i'm definitely doing the right
thing by switching from the wx backend to perhaps any other backend.
I'm going with the qt4 backend.
Thanks,
Jim

Hi,
I'm not sure if I'm doing something wrong, but I find the sizing of
the axes areas when doing plots a bit problematic.
If I make the figures small, the actual plot area takes up most of the
area and the axes labels aren't visible unless I make the window
bigger. When I export to eps or pdf, they are also cut off.
Is there a way to get the plot area scaled to the figure size in such
a way that the axes labels can be included? I am trying to avoid
having to scale the image when it is included in the document.
Thanks,
Robin

Hi, Andrea. Yes, I bet that upgrading would resolve the issue. However,
the issue isn't getting in my way at the moment, since I use the WXAgg
backend and it works well as far as I can tell. I mainly wanted to point
out for informational purposes -- and I may have been unclear about my point
-- that the Enthought Python distribution is currently at wxPython 2.8.4 and
that the WX backend seems not to work at that version.

Hi Michael,
On Fri, Jun 13, 2008 at 2:23 PM, Michael Droettboom wrote:
> FWIW, I developed the new wx backend using wx 2.8.6.1, and it does currently
> work for me. Unfortunately, it's always hard to know how to write for a
> broad range of point releases for wxPython -- the documentation offers
> little clues about when features were added or changed.
Then it might be possible that something has changed between 2.8.4.0
(the OP wxPython version) and 2.8.6.0 (yours), although it looks like
a wxWidgets difference rather than a wxPython one...
Andrea.
"Imagination Is The Only Weapon In The War Against Reality."
http://xoomer.alice.it/infinity77/

FWIW, I developed the new wx backend using wx 2.8.6.1, and it does 
currently work for me. Unfortunately, it's always hard to know how to 
write for a broad range of point releases for wxPython -- the 
documentation offers little clues about when features were added or changed.
Cheers,
Mike
Andrea Gavana wrote:
> Hi Stan,
>
> On Fri, Jun 13, 2008 at 2:09 PM, Stan West wrote:
> 
>> For what it's worth, the WX backend isn't working for me with wxPython
>> 2.8.4; I get the following when I attempt to show a figure:
>>
>> File "C:\Program
>> Files\Python25\lib\site-packages\matplotlib-0.98.0_r5467-py2.5-win32.egg\mat
>> plotlib\backends\backend_wx.py", line 474, in select
>> self.SelectObject(self.bitmap)
>> AttributeError: GraphicsContextWx instance has no attribute 'SelectObject'
>>
>> I'm at wxPython 2.8.4 simply because it is included with the Enthought
>> Python Distribution that I picked for getting started with Python.
>>
>> However, I don't normally use the WX backend. I've been using WXAgg, so WX
>> is relevant for me only for the functionality inherited by WXAgg.
>> 
>
> Please upgrade wxPython to the latest release (2.8.7.1) from here:
>
> http://www.wxpython.org/download.php#binaries
>
> Or use the very latest pre-release:
>
> http://wxpython.wxcommunity.com/preview/20080608
>
> This will (probably) fix the problem you see with wx.GraphicsContext,
> although I know next to nothing about matplotlib internal handling of
> wxPython things.
>
> Andrea.
>
> "Imagination Is The Only Weapon In The War Against Reality."
> http://xoomer.alice.it/infinity77/
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

Hi Stan,
On Fri, Jun 13, 2008 at 2:09 PM, Stan West wrote:
> For what it's worth, the WX backend isn't working for me with wxPython
> 2.8.4; I get the following when I attempt to show a figure:
>
> File "C:\Program
> Files\Python25\lib\site-packages\matplotlib-0.98.0_r5467-py2.5-win32.egg\mat
> plotlib\backends\backend_wx.py", line 474, in select
> self.SelectObject(self.bitmap)
> AttributeError: GraphicsContextWx instance has no attribute 'SelectObject'
>
> I'm at wxPython 2.8.4 simply because it is included with the Enthought
> Python Distribution that I picked for getting started with Python.
>
> However, I don't normally use the WX backend. I've been using WXAgg, so WX
> is relevant for me only for the functionality inherited by WXAgg.
Please upgrade wxPython to the latest release (2.8.7.1) from here:
http://www.wxpython.org/download.php#binaries
Or use the very latest pre-release:
http://wxpython.wxcommunity.com/preview/20080608
This will (probably) fix the problem you see with wx.GraphicsContext,
although I know next to nothing about matplotlib internal handling of
wxPython things.
Andrea.
"Imagination Is The Only Weapon In The War Against Reality."
http://xoomer.alice.it/infinity77/

For what it's worth, the WX backend isn't working for me with wxPython
2.8.4; I get the following when I attempt to show a figure:
 File "C:\Program
Files\Python25\lib\site-packages\matplotlib-0.98.0_r5467-py2.5-win32.egg\mat
plotlib\backends\backend_wx.py", line 474, in select
 self.SelectObject(self.bitmap)
AttributeError: GraphicsContextWx instance has no attribute 'SelectObject'
I'm at wxPython 2.8.4 simply because it is included with the Enthought
Python Distribution that I picked for getting started with Python.
However, I don't normally use the WX backend. I've been using WXAgg, so WX
is relevant for me only for the functionality inherited by WXAgg.
Stan

Sorry I'm just getting into this discussion now.
The wx backend in 0.98 is almost a complete rewrite to use 
wx.GraphicsContext rather than the older wx.DC interface. It was no 
longer possible to move forward with the new rendering infrastructure 
with the wx.DC interface due to lack of support for polycurves. As the 
wx backend is one of the least used, perhaps, it hasn't seen the same 
amount of testing as the others and you're running into the bleeding 
edge. (And some of us still insert an implicit "beta" everytime we see 
the number 0.98... ;)
Note that in my testing, the new wx backend isn't any faster than wxagg 
over an X11 connection, and locally it's *insanely* slow. On Linux it 
uses Cairo, which has the same bitmap transferring issues that Agg has. 
So that eliminates any advantages it may have had. Possibly if one were 
doing some wx.GraphicsContext rendering interleaved with matplotlib 
rendering it might be useful, but I doubt there's any of that in the 
wild. I would strongly suggest we just remove the wx backend at this 
point and move to wxAgg for everything. I advocated this months ago, 
but forgot to follow through on it --->
http://sourceforge.net/mailarchive/message.php?msg_id=4734927C.3060601%40stsci.edu
Local:
WxAgg: 27.5 fps
Wx: 1.8 fps
Wx: 16.8 fps (without clipping)
Network (X11 with compression):
WxAgg: 2.0 fps
Wx: 1.7 fps
Wx: 2.0 fps (without clipping)
Cheers,
Mike
John Hunter wrote:
> On Thu, Jun 12, 2008 at 9:21 PM, James A. Benson
> <jb...@se...> wrote:
> 
>> My app uses the wx backend, so I happened to notice
>> that the backend_wx doesn't seem to work with
>> matplotlib 0.98.0. Running the 0.98 example embedding_in_wx.py
>> shows the problem. Only a bit of the lines get plotted. The
>> pan up/down zoom in/out all show various error messages.
>> When i revert back to 0.91.3 and run the same 0.91 example
>> it works fine.
>> 
>
> Hey James, we were discussing problems with the wx backend on 0.98
> yesterday on the devel list, and Ken offered to take a look at them,
> so it is possible we will have those fixed for the next bugfix
> release. What wx version are you using?
>
> Note that unless you are running over an x11 connection, there is
> little reason to prefer wx over wxagg, which has better rendering and
> more features. Over an x11 connection, wxagg is slow because it has
> to pass the who bitmap over the wire, but otherwise it works great.
>
> JDH
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

On Fri, Jun 13, 2008 at 5:24 AM, Jatinder Syali <jat...@ya...> wrote:
> Hi all,
>
> I am trying to create pie chart using matplotlib. I have 9 items to be
> plotted in the chart. sometimes the length of the lable appears to be long
> which scatteres them in the chart. Is there any way to create a link/line
> between label and its pie to make the represenation meaningfull and obvi
Someone recently posted a patch to use left or right alignment of text
depending on which side of he pie you are on. I think Michael was
going to commit that if noone objected, and it seems like a good idea
to me. That will probably help. The main poblem I see problem in
your example is the labeldistance is too large, try 1.1 or 0.9 and see
if that helps.
JDH

On Fri, Jun 13, 2008 at 2:43 AM, Dan Murphy <chi...@ma...> wrote:
>>>> import scipy
>>>> scipy.test('1','10')
> gave me this
> Traceback (most recent call last):
> ImportError: Need nose >=0.10 for tests - see
This is just an error in the testing framework. Apparently you need
to upgrade to nose 0.1 to run the tests. It looks like scipy built
and installed correctly since you could import it. If you want to run
the tests, go ahead and upgrade nose, but it is not essential.
> Any thoughts? Since scipy did not seem to install correctly, I did not move
> on to the matplotlib step. Can I move on to matplotlib without scipy?
Sorry I wasn't clear. This installation page is for building a
standard set of tools (MINS - matplotlib, ipython, numpy, scipy) we
use in a course we teach on scientific computing in python.
matplotlib does not require ipython or scipy (but you'll be glad you
have them!). You can proceed with the matplotlib build instructions.
JDH
PS, please always keep your replies on list rather than just replying to me.

On Thu, Jun 12, 2008 at 8:56 PM, T J <tj...@gm...> wrote:
> I am making a scatter plot and want the legend to display the symbols.
> This functionality doesn't seem to exist, so I have followed the
> workaround outlined here:
>
> http://www.nabble.com/Legend-for-a-scatter-plot-based-on-symbols-td17554839.html#a17554839
>
> Are there any plans to make the symbols which are available in plot()
> the same as those available for scatter()? If not, can we at least
> get the diamond symbol the same? I want to pass the same symbol to
> plot() and scatter() and get the same symbol---as it is, I must use
> 'd' in scatter and 'D' in plot.
I think this is worth fixing -- could you file a bug on the
sourceforge site and note any symbols you are aware of that don't
agree between plot and scatter.
>
> Also, how are the markersizes scaled? For example, in scatter(), I am
> using s=30...but if I do plot(...,markersize=30), then the markers are
> not the same size as the markers from the scatter plot. I can go back
> and forth until the scale is right, but is there a better way?
Note sure if this is a bug or a feature, but it is a consequence of
emulating the matlab approach. In scatter, the size often is used to
convey information, and the size parameter give thes area of the
circle that circumscribes the marker. In plot, he marker size is not
generally intended to convey information, only that something happened
here, and the size choice is usually one of aesthetics. In any case,
in plot the size is radius of the circle, the side of the square or
diamond. For most markers, you should be able to square the
markersize to get approximate correspondence between plot and scatter.
JDH

On Thu, Jun 12, 2008 at 9:21 PM, James A. Benson
<jb...@se...> wrote:
>
> My app uses the wx backend, so I happened to notice
> that the backend_wx doesn't seem to work with
> matplotlib 0.98.0. Running the 0.98 example embedding_in_wx.py
> shows the problem. Only a bit of the lines get plotted. The
> pan up/down zoom in/out all show various error messages.
> When i revert back to 0.91.3 and run the same 0.91 example
> it works fine.
Hey James, we were discussing problems with the wx backend on 0.98
yesterday on the devel list, and Ken offered to take a look at them,
so it is possible we will have those fixed for the next bugfix
release. What wx version are you using?
Note that unless you are running over an x11 connection, there is
little reason to prefer wx over wxagg, which has better rendering and
more features. Over an x11 connection, wxagg is slow because it has
to pass the who bitmap over the wire, but otherwise it works great.
JDH

Hello, matplotlib-users.
 Hi! I'm using matplotlib to display dynamic graph of incoming data.
 Separate window with embedded graph in it.
 Everything okay, except of too much, or too less space available for
graph data, since it passed as percentage. For my case better to say
that 20px from left and 20px from bottom are for axes, everything
other is graph data. Is that possible with matplotlib? Or only handle
resize request and update percentage as needed?
-- 
Best regards,
 Anton
 Call2Ru Service
 dat...@ca...

 My app uses the wx backend, so I happened to notice
that the backend_wx doesn't seem to work with
matplotlib 0.98.0. Running the 0.98 example embedding_in_wx.py
shows the problem. Only a bit of the lines get plotted. The
pan up/down zoom in/out all show various error messages.
When i revert back to 0.91.3 and run the same 0.91 example
it works fine.
 Not a big deal for me (perhaps the only person still using
the wx backend), i'm in the process of switching my app to qt4
and the embedding_in_qt4.py example works fine in 0.98.
Jim

I am making a scatter plot and want the legend to display the symbols.
 This functionality doesn't seem to exist, so I have followed the
workaround outlined here:
http://www.nabble.com/Legend-for-a-scatter-plot-based-on-symbols-td17554839.html#a17554839
Are there any plans to make the symbols which are available in plot()
the same as those available for scatter()? If not, can we at least
get the diamond symbol the same? I want to pass the same symbol to
plot() and scatter() and get the same symbol---as it is, I must use
'd' in scatter and 'D' in plot.
Also, how are the markersizes scaled? For example, in scatter(), I am
using s=30...but if I do plot(...,markersize=30), then the markers are
not the same size as the markers from the scatter plot. I can go back
and forth until the scale is right, but is there a better way?
Thanks.

On Thu, Jun 12, 2008 at 4:19 PM, T J <tj...@gm...> wrote:
> I just updated my version of matplotlib to r5496, and now when I type
> show(), nothing happens. What could have gone wrong? Sorry if this is
> simple. I've attached a log of the installation.
Try
 > rm -rf build
 > rm -rf /path/to/site-packages/matplotlib*
 > svn up
 > python setup.py install > build.out
 > python examples/pylab_examples/simple_plot.py --verbose-debug > run.out
and post build.out and run.out.
Also, please post svn problems to the matplotlib-devel list
http://lists.sourceforge.net/mailman/listinfo/matplotlib-devel

I just updated my version of matplotlib to r5496, and now when I type
show(), nothing happens. What could have gone wrong? Sorry if this is
simple. I've attached a log of the installation.
t

On Thu, Jun 12, 2008 at 2:58 PM, İsmail Dönmez <is...@na...> wrote:
> Hi all,
>
> I was trying matplotlib 0.98.0 and Qt4 backend works great but Coca
> backend seems to be broken :
Charlie,
any chance you can try and port cocoaagg over to the new trunk api.
figure.dpi is no longer a lazy value, but a plain ol number. There
are probably a few other changes that will need to be made as well.
I'd like to clear up as many of these problems as we can and shoot for
a bugfix release next week.
JDH

Hi all,
I was trying matplotlib 0.98.0 and Qt4 backend works great but Coca
backend seems to be broken :
Python 2.5.1 (r251:54863, Jan 17 2008, 19:35:17)
[GCC 4.0.1 (Apple Inc. build 5465)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from scipy import lena
>>> import matplotlib
>>> matplotlib.use("CocoaAgg")
>>> import pylab
>>> pylab.imshow(lena())
<matplotlib.image.AxesImage object at 0x2ac0a10>
>>> pylab.show()
Traceback (most recent call last):
 File "<stdin>", line 1, in <module>
 File "/Library/Python/2.5/site-packages/matplotlib-0.98.0-py2.5-macosx-10.3-i386.egg/matplotlib/backends/backend_cocoaagg.py",
line 46, in show
 manager.show()
 File "/Library/Python/2.5/site-packages/matplotlib-0.98.0-py2.5-macosx-10.3-i386.egg/matplotlib/backends/backend_cocoaagg.py",
line 205, in show
 NSApplication.sharedApplication().run()
 File "/Library/Python/2.5/site-packages/matplotlib-0.98.0-py2.5-macosx-10.3-i386.egg/matplotlib/backends/backend_cocoaagg.py",
line 181, in startWithBundle_
 if not bundle.loadNibFile_externalNameTable_withZone_('Matplotlib.nib',
{}, None):
 File "/Library/Python/2.5/site-packages/matplotlib-0.98.0-py2.5-macosx-10.3-i386.egg/matplotlib/backends/backend_cocoaagg.py",
line 88, in awakeFromNib
 self.plotView.windowDidResize_(self)
 File "/Library/Python/2.5/site-packages/matplotlib-0.98.0-py2.5-macosx-10.3-i386.egg/matplotlib/backends/backend_cocoaagg.py",
line 137, in windowDidResize_
 dpi = self.canvas.figure.dpi.get()
AttributeError: 'int' object has no attribute 'get'
>>>
Any ideas?
Regards,
ismail
-- 
Never learn by your mistakes, if you do you may never dare to try again.

John Hunter wrote:
> Could you try the instructions at:
> 
> http://ipython.scipy.org/moin/Py4Science/InstallationOSX
Minor note on that:
"""
I installed
wxPython2.8-osx-unicode-2.8.3.0-universal10.4-py2.5.dmg from the 
pythonmac site and am crossing my fingers that I don't find myself 
entangled in wx version hell.
"""
recent version of the wx back-end no longer rely on compiled code, so wx 
version hell is a think of the past (you will get better performance 
with wx 2.8+, though).
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

John Hunter wrote:
> Could you try the instructions at:
> 
> http://ipython.scipy.org/moin/Py4Science/InstallationOSX
note that these only work on OS-X 10.5 (which was the OP's question).
However, the key issue is if you are building it just for yourself, or 
if you want to be able to re-distribute either your build or 
applications built from it via py2app, etc.
If the latter, you want Charlie Moad's page:
http://ipython.scipy.org/moin/MatplotlibOSXBuildNotes
However, it really is a lot easier to just install the binary egg found 
here:
http://sourceforge.net/project/showfiles.php?group_id=80706&package_id=278194&release_id=603020
It's called:
 	matplotlib-0.98.0-py2.5-macosx-10.3.egg
the "10.3" means "10.3 or greater".
you may need to re-name it:
matplotlib-0.98.0-py2.5.egg
then:
$ easy_install matplotlib-0.98.0-py2.5.egg
(I may not have the re-naming right, I'm not running 10.5 to test.)
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

On Thu, Jun 12, 2008 at 1:11 AM, Dan Murphy <chi...@ma...> wrote:
> I tried to follow the directions per
>
> % python setup.py build
> % python setup.py install
Could you try the instructions at:
http://ipython.scipy.org/moin/Py4Science/InstallationOSX
and let us know how it goes?
Thanks,
JDH

If you can find a way to create a minimal script, that would be better, 
because then we can verify the fix.
But barring anything else, seeing the .eps might help.
Cheers,
Mike
Michael Hearne wrote:
> John - Thanks for the quick fix. I'm having a separate issue now with 
> an EPS file being generated (using savefig() again) that appears to be 
> invalid (can't display it in OS X Preview, or convert to PDF with 
> ps2pdf). Unfortunately, the code that creates this particular file is 
> rather involved, and hard to replicate with one simple script.
>
> If I send the offending EPS file (it's 1.4 MB zipped), is it possible to 
> tell from that what the problem is?
>
> The errors from ps2pdf are:
> Error: /undefined in --get--
> Operand stack:
> true --dict:43/43(ro)(L)-- .notdef
> Execution stack:
> %interp_exit .runexec2 --nostringval-- --nostringval-- 
> --nostringval-- 2 %stopped_push --nostringval-- 
> --nostringval-- --nostringval-- false 1 %stopped_push 1905 
> 1 3 %oparray_pop 1904 1 3 %oparray_pop --nostringval-- 
> 1888 1 3 %oparray_pop 1771 1 3 %oparray_pop 
> --nostringval-- %errorexec_pop .runexec2 --nostringval-- 
> --nostringval-- --nostringval-- 2 %stopped_push 
> --nostringval-- %finish_show --nostringval-- --nostringval-- 5 
> 4 0 --nostringval-- (pdf_text_enum_t) %op_show_continue 
> --nostringval--
> Dictionary stack:
> --dict:1144/1684(ro)(G)-- --dict:0/20(G)-- --dict:75/200(L)-- 
> --dict:61/112(L)-- --dict:20/25(ro)(L)--
> Current allocation mode is local
> Last OS error: 2
> Current file position is 996122
> GPL Ghostscript 8.61: Unrecoverable error, exit code 1
>
> --Mike
>
> John Hunter wrote:
> 
>> On Wed, Jun 11, 2008 at 10:14 AM, John Hunter <jd...@gm...> wrote:
>>
>> 
>> 
>>> Interesting. When I plot it on my screen it looks correct with a *Agg
>>> GUI backend. But when I save it (either from the GUI or using
>>> savefig) it has the problem you describe. This suggests to me that
>>> either some cached information or something is not getting notified of
>>> the different default dpi in savefig. I'll look into it. Since PS is
>>> dpi independent, it would not affect it.
>>> 
>>> 
>> Well, my diagnosis wasn't correct, but I fixed the problem. In the
>> caching of the text layout, we were not taking into account the
>> different renderers in the cache key. Since PS has a nominal dpi
>> setting that is ignored by postscript, the differences in dpi were not
>> getting picked up in the layout cache. The solution I implemented was
>> to simply add the renderer id to the property tuple used for caching.
>> Fixed in svn r5470.
>>
>> JDH
>> 
>> 
>
> 
-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

Hi folks
Using version 0.91.2, I do not manage to push ticks labels to the right
side of the axis, keeping ticks lines on both sides. How can I do it ?
Another question : 
I have two subplot "in-line" (121 and 122) sharing the same yaxis.
How would you set a single set of yticks labels displayed between the
axes ?
-- 
Fabrice Silva <si...@lm...>
LMA UPR CNRS 7051 - équipe S2M