matplotlib-users Mailing List for matplotlib

Hello,

How complicated would it be to obtain the add_patch() capability in 
polar axes? Right now, the script

--------------------
from pylab import *
import matplotlib.patches as patches

ax = axes( [0.1, 0.1, 0.8, 0.8], polar = True )
ptch = patches.RegularPolygon( (pi/2,.5), 5, orientation = 20 )
ax.add_patch()
show()
--------------------

fails, Python complaining that polar axes don't have an add_patch() method.

Now, I believe if we simply mimic what's in the Axes class and try to use

artists.extend(patches)

in PolarAxes, the coordinates are going to be all wacko.

The following works for me, and I wonder whether it is a useful addition 
to PolarAxes (sorry for the wrapped lines):

--------------------
from matplotlib.pylab import *


def RegularPolygon_polar( ax, xy, numVertices, radius = 5, orientation = 
0, **kwargs ):
     import matplotlib.patches as patches
     import matplotlib.transforms as transforms

     ptch = patches.RegularPolygon( (0,0),
                                    numVertices,
                                    radius = radius,
                                    orientation = orientation,
                                    **kwargs )

     trans = transforms.identity_affine()
     trans.set_offset( xy, ax.transData )
     ptch.set_transform( trans )
     return ptch

ax = axes( [0.1, 0.1, 0.8, 0.8], polar = True )
ptch = RegularPolygon_polar( ax, (150*pi/180,.5), 5, radius = 10, 
orientation = 80 )
ax.add_patch( ptch )
show()
--------------------

If this sounds like the way to go, let me know. I'm willing to write the 
few methods similar to the above and add them to PolarAxes.

On the other hand, it might be easier to simply modify the classes in 
patches.py, but I am not sure how they would access the Axes instance.

Cheers,
Dominique

>>>>> "Delbert" == Delbert D Franz <iq...@so...> writes:

    Delbert> I just did another test. mx2num fails for dates prior to
    Delbert> 1970 because it uses the magical Unix seconds from epoch.
    Delbert> We need mx2num to work for the whole range of dates for
    Delbert> which date2num works-at least that is my vote.  Delbert
    Delbert> On Wednesday 02 February 2005 10:10 am, Delbert D. Franz wrote:

If you feel up to making an mx2num that satisfies these requirements,
I'm happy to include it as a patch.

JDH

I just did another test. mx2num fails for dates prior to 1970 because
it uses the magical Unix seconds from epoch.  We need mx2num to work 
for the whole range of dates for which  date2num  works-at least that
is my vote. 
                                  Delbert
On Wednesday 02 February 2005 10:10 am, Delbert D. Franz wrote:
> John, thanks for the updated mx2num but I still found something
> puzzling: date2num takes the timezone from the matplotlibrc file
> but mx2num must take it from some info on my machine.  There is a
> consistent seven hour difference between the two even though the 
> datetime instances  give exactly the same time. 
> 
> I need to point out that in my applications I have no interest in 
> the time zone and I want it always treated as if I am 
> on the Greenwich Meridian.  Here's why:  My simulation software
> has used what is essentially the modified Julian Day number 
> from astronomy with a zero point  around 1878 or so.  I just add 
> the fraction of the day to it.   This works
> fine in my applications because there is no data to run the model 
> earlier than about the 1900!  Also I have users in different time 
> zones sending me test cases and it would be a mess to have to 
> somehow always adjust for time zone.  Thus I have effectively 
> computed the day numbers relative to the local time zone always. 
> 
> Another benefit is that there is a fixed offset between my 
> time scale and the one used in plot_dates.  Makes life a little
> simpler. 
> 
> Questions: 
> 
> 1. Should mx2num use a different time zone than does date2num?
> 
> 2. If so, how do I get mx2num at its invocation only, to use 
>    the time zone in the matplotlibrc file?  I assume I have 
>    to use the set_tzinfo method right before mx2num is invoked
>    and then use set_tzinfo again right after to return to 
>    what it was. 
> 
> Heres an example test file: 
> 
> from matplotlib.dates import mx2num, date2num
> import datetime
> import mx.DateTime
> 
> for ihr in xrange(24):
>     yr = 2003
>     mn = 10
>     day = 20
>     imin = 30
>     second = 0.0
>     isec = 0
> 
>     dt = mx.DateTime.DateTime(yr, mn, day, ihr, imin, second)
>     dt2 = datetime.datetime(yr, mn, day, ihr, imin, isec)
>     print 'hour=', ihr
>     print 'mx dt=',dt
>     print 'dt2=', dt2
>     diff =  mx2num(dt) - date2num(dt2)
>     print 'mx2num(dt)=',mx2num(dt)
>     print 'date2num(dt2) =', date2num(dt2)
>     print 'difference=', diff*24.0
>     print ' '
> 
> 
> Here's some output at my time zone:
> 
> 
> hour= 0
> mx dt= 2003-10-20 00:30:00.00
> dt2= 2003-10-20 00:30:00
> mx2num(dt)= 731508.3125
> date2num(dt2) = 731508.020833
> difference= 6.99999999907
>  
> hour= 1
> mx dt= 2003-10-20 01:30:00.00
> dt2= 2003-10-20 01:30:00
> mx2num(dt)= 731508.354167
> date2num(dt2) = 731508.0625
> difference= 6.99999999907
>  
> hour= 2
> mx dt= 2003-10-20 02:30:00.00
> dt2= 2003-10-20 02:30:00
> mx2num(dt)= 731508.395833
> date2num(dt2) = 731508.104167
> difference= 7.00000000186
>  
> hour= 3
> mx dt= 2003-10-20 03:30:00.00
> dt2= 2003-10-20 03:30:00
> mx2num(dt)= 731508.4375
> date2num(dt2) = 731508.145833
> difference= 6.99999999907
>  
> hour= 4
> mx dt= 2003-10-20 04:30:00.00
> dt2= 2003-10-20 04:30:00
> mx2num(dt)= 731508.479167
> date2num(dt2) = 731508.1875
> difference= 6.99999999907
>  
> hour= 5
> mx dt= 2003-10-20 05:30:00.00
> dt2= 2003-10-20 05:30:00
> mx2num(dt)= 731508.520833
> date2num(dt2) = 731508.229167
> difference= 7.00000000186
>  
>                                 Delbert
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
> Tool for open source databases. Create drag-&-drop reports. Save time
> by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> 

John, thanks for the updated mx2num but I still found something
puzzling: date2num takes the timezone from the matplotlibrc file
but mx2num must take it from some info on my machine.  There is a
consistent seven hour difference between the two even though the 
datetime instances  give exactly the same time. 

I need to point out that in my applications I have no interest in 
the time zone and I want it always treated as if I am 
on the Greenwich Meridian.  Here's why:  My simulation software
has used what is essentially the modified Julian Day number 
from astronomy with a zero point  around 1878 or so.  I just add 
the fraction of the day to it.   This works
fine in my applications because there is no data to run the model 
earlier than about the 1900!  Also I have users in different time 
zones sending me test cases and it would be a mess to have to 
somehow always adjust for time zone.  Thus I have effectively 
computed the day numbers relative to the local time zone always. 

Another benefit is that there is a fixed offset between my 
time scale and the one used in plot_dates.  Makes life a little
simpler. 

Questions: 

1. Should mx2num use a different time zone than does date2num?

2. If so, how do I get mx2num at its invocation only, to use 
   the time zone in the matplotlibrc file?  I assume I have 
   to use the set_tzinfo method right before mx2num is invoked
   and then use set_tzinfo again right after to return to 
   what it was. 

Heres an example test file: 

from matplotlib.dates import mx2num, date2num
import datetime
import mx.DateTime

for ihr in xrange(24):
    yr = 2003
    mn = 10
    day = 20
    imin = 30
    second = 0.0
    isec = 0

    dt = mx.DateTime.DateTime(yr, mn, day, ihr, imin, second)
    dt2 = datetime.datetime(yr, mn, day, ihr, imin, isec)
    print 'hour=', ihr
    print 'mx dt=',dt
    print 'dt2=', dt2
    diff =  mx2num(dt) - date2num(dt2)
    print 'mx2num(dt)=',mx2num(dt)
    print 'date2num(dt2) =', date2num(dt2)
    print 'difference=', diff*24.0
    print ' '


Here's some output at my time zone:


hour= 0
mx dt= 2003-10-20 00:30:00.00
dt2= 2003-10-20 00:30:00
mx2num(dt)= 731508.3125
date2num(dt2) = 731508.020833
difference= 6.99999999907
 
hour= 1
mx dt= 2003-10-20 01:30:00.00
dt2= 2003-10-20 01:30:00
mx2num(dt)= 731508.354167
date2num(dt2) = 731508.0625
difference= 6.99999999907
 
hour= 2
mx dt= 2003-10-20 02:30:00.00
dt2= 2003-10-20 02:30:00
mx2num(dt)= 731508.395833
date2num(dt2) = 731508.104167
difference= 7.00000000186
 
hour= 3
mx dt= 2003-10-20 03:30:00.00
dt2= 2003-10-20 03:30:00
mx2num(dt)= 731508.4375
date2num(dt2) = 731508.145833
difference= 6.99999999907
 
hour= 4
mx dt= 2003-10-20 04:30:00.00
dt2= 2003-10-20 04:30:00
mx2num(dt)= 731508.479167
date2num(dt2) = 731508.1875
difference= 6.99999999907
 
hour= 5
mx dt= 2003-10-20 05:30:00.00
dt2= 2003-10-20 05:30:00
mx2num(dt)= 731508.520833
date2num(dt2) = 731508.229167
difference= 7.00000000186
 
                                Delbert

>>>>> "Delbert" == Delbert D Franz <dd...@lk...> writes:

    Delbert> I am now able to create multiple figures, each with one
    Delbert> or more subplots with my software.  What a great
    Delbert> collection of software.  Thanks for all the work.

Great, glad it helped.  I was pretty sure that was the answer.

    Delbert> I tried using mx.datetime to create an mx.datetime
    Delbert> instance which worked fine. However, mx2num() fails with
    Delbert> python claiming an undefined mxdates.  I checked the
    Delbert> source and found only one mxdates.  No idea how it should
    Delbert> be defined.  In the mean time I am using a datetime
    Delbert> instance but I lose a small bit of precision with only
    Delbert> integer seconds.

Oops, replace mx2num in matplotlib/dates.py with 

def mx2num(mxdates):
    """
    Convert mx datetime instance (or sequence of mx instances) to the
    new date format,
    """
    scalar = False
    if not iterable(mxdates):
        scalar = True
        mxdates = [mxdates]
    ret = epoch2num([m.ticks() for m in mxdates])
    if scalar: return ret[0]
    else: return ret


Thanks for the report!
JDH

>>>>> "Nils" == Nils Wagner <nw...@me...> writes:

    Nils> Hi all, Is it possible to rule the distance between subplots
    Nils> ?  How about the setting of x axis tic marks ? The default
    Nils> is not very promising (see bsp5.py)

For axes placement, see also the axes command, which gives you a finer
degree of control than subplot.

For the tick labeling, you are right, the default xticklabels in your
example are a mess.  We'll take a look at this case to see where the
bug is.  In my experience, getting default ticking and labeling right
is hard, and matplotlib has gotten a lot better at it since the bad
old days but there is a ways to go.

Until we get this fixed, you do have the option of using a custom
ticker, which is explained in the user's guide Chapter 5, and
illustrated in examples/custom_ticker1.py in the matplotlib src
distro.

JDH

>>>>> "Cory" == Cory Davis <cd...@st...> writes:

    Cory> try title('$\mu='+str(p[ic-1])+'$') Cory.

Make sure you quote the string as a raw string with the "r" prefix.  I
also suggest format strings for formatting numeric values, something
like

title(r'$\mu=%1.2f$' % p[ic-1])
      ^

JDH

On Wednesday 02 February 2005 04:51 am, Steve Chaplin wrote:
>
> When resizing the FigureCanvas on my system I notice:
> GTK   sets a FigureCanvas size that can only be increased.
> QtAgg sets a FigureCanvas size that can be increased or decreased
> TkAgg sets a FigureCanvas size that can be increased or decreased to a
> minimum (of perhaps 400 x 300 pixels)
> Wx I do not have, is it like Qt or Tk or is there a 4th variation?

on linux, wx lets me increase or decrease all the way. I wonder if it is 
different on windows.

>
> I looked at changing GTK so the FigureCanvas could be decreased a while
> ago but the problem is that while the axes scale/shrink nicely the text
> stays the same size and becomes misaligned or clipped and gives an odd
> looking graph. Perhaps thats why Tk allows you shrink the window but
> only to a certain size.
>
> Matplotlib currently has 3 (or more) ways of handling window resizing,
> which one of these behaviours is 'correct' or most desirable?
>

I think the most desirable for interactive use would be to allow scaling, but 
if it doesn't scale properly, I think the current behavior is appropriate. I 
guess I just didnt notice it before now.

Darren

try title('$\mu='+str(p[ic-1])+'$')

Cory.

On Wed, 2005-02-02 at 12:48, Nils Wagner wrote:
> Hi all,
> 
> I am going to change the title of each subplot with respect to the 
> actual value of \mu within a loop.
> How can I do that ?
> 
> p = array(([0.1,0.01,0.001,0.0001]))
> ic = 1
> for mu in p:
>  title('$\mu=$'+str(p[ic-1]))  # This version doesn't work
>  ic = ic + 1
> 
> Nils
> 
>  
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
> Tool for open source databases. Create drag-&-drop reports. Save time
> by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
> Download a FREE copy at http://www.intelliview.com/go/osdn_nl
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
-- 
))))))))))))))))))))))))))))))))))))))))))))
Cory Davis
Meteorology
School of GeoSciences
University of Edinburgh
King's Buildings
EDINBURGH EH9 3JZ

ph: +44(0)131 6505092
fax +44(0)131 6505780
cd...@st...
co...@me...
http://www.geos.ed.ac.uk/contacts/homes/cdavis
))))))))))))))))))))))))))))))))))))))))))))

Hi Nils,

On Wed, 2005-02-02 at 12:22, Nils Wagner wrote:
> Hi all,
> 
> Is it possible to rule the distance between subplots ?
> How about the setting of x axis tic marks ? The default is not very 
> promising
> (see bsp5.py)

Here is a function I use instead of the matplotlib subplot.

It has the same arguments as the matlab version except for the 
figpos argument, which determines where in the figure your
array of plots goes, and the axpos argument, which determines where in each 
panel the axes goes.  It returns an axes object, which you can use 
to set xticks etc.  

Hope this helps,
Cheers,
Cory.

Hi all,

I am going to change the title of each subplot with respect to the 
actual value of \mu within a loop.
How can I do that ?

p = array(([0.1,0.01,0.001,0.0001]))
ic = 1
for mu in p:
 title('$\mu=$'+str(p[ic-1]))  # This version doesn't work
 ic = ic + 1

Nils

 

 

>>>>> "dimitri" == dimitri pater <dim...@gm...> writes:

    dimitri> hello, I want to create a barchart on a website, somebody
    dimitri> advised me to use matplotlib and it does look very
    dimitri> promising. The problem that I now face is that I can't
    dimitri> seem to get things right, the following code results in
    dimitri> an import error:

    dimitri> from cgi import escape
 
    dimitri> try: import cgitb cgitb.enable() except: sys.stderr =
    dimitri> sys.stdout

    dimitri> # test import pylab import pylab

    dimitri> Does this mean that the webserver doesn't have pylab
    dimitri> installed or am I missing something? (bit of a newbie in
    dimitri> this area, so please bear with me...)

To run matplotlib in a web app server, you will probably want to set
"backend : Agg" in your .matplotlibrc file.  matplotlib supports
output to a variety of image devices and GUIs, and agg is a good
choice to make nice PNGs for app servers.  See
http://matplotlib.sf.net/.matplotlibrc.  On a standard linux install,
this file would be placed in /usr/share/matplotlib/.matplotlibrc, and
i can be copied to your HOME directory and edited.  Often times HOME
is not net in an app server environment.  You either need to set it,
or take a look at
http://groups-beta.google.com/group/comp.lang.python/msg/09eac15acef8ee40,
which describes some possible solutions.  matplotlib needs a directory
that is can write to to save some font cache information, and the link
above describes the role of the environment variables HOME and
MATPLOTLIBDATA in the choices matplotlib makes.

As for your error, I suggest taking a minimal matplotlib script

import pylab
pylab.plot([1,2,3])
pylab.savefig('test.png')

and running it with

> python myscript.py --verbose-helpful

and report any output and/or errors the script produces.  Saying that
you got an import error, without giving us the exact error message`,
doesn't help us help you.

Good luck!
JDH

Hi all,

Is it possible to rule the distance between subplots ?
How about the setting of x axis tic marks ? The default is not very 
promising
(see bsp5.py)

Any pointer or suggestion how to improve my example are welcome.

Thanks in advance.

        Nils
 

hello,

I want to create a barchart on a website, somebody advised me to use
matplotlib and it does look very promising. The problem that I now
face is that I can't seem to get things right, the following code
results in an import error:

from cgi import escape
 
try:
    import cgitb
    cgitb.enable()
except:
    sys.stderr = sys.stdout

# test import pylab
import pylab

Does this mean that the webserver doesn't have pylab installed or am I
missing something? (bit of a newbie in this area, so please bear with
me...)

best regards,
Dimitri
-- 
Please visit dimitri's website: www.serpia.com

On Tue, 2005-02-01 at 20:17 -0800, Darren Dale wrote:
> I am working in interactive mode, with the GTKAgg backend. When I make a 
> figure, I was not able to resize it on the screen. I tried changing my dpi 
> and default figure size in .matplotlibrc, and now I can increase the figure 
> size, but not decrease it passed its original size. I changed 
> my .matplotlibrc back to its original configuration, and the same behavior 
> persists: I can increase the figure size, but not decrease it passed the 
> original size. This is also true for the GTK backend, but not for WXAgg.
> 
> I rebuilt my atlas libraries today, along with Numeric, and numarray. I dont 
> think that should have anything to do with it, but I rebuilt gtk+, pygtk, and 
> matplotlib just in case. All are the latest version. 
> 
> Everything else seems to work, minimizing, maximizing, closing, but not 
> resizing. Does anyone have a suggestion as to what is wrong, or what I might 
> check?

Its related to the way the FigureCanvasGTK is implemented using
self.set_size_request() - it sets the widgets minimum size.
 
When resizing the FigureCanvas on my system I notice:
GTK   sets a FigureCanvas size that can only be increased.
QtAgg sets a FigureCanvas size that can be increased or decreased
TkAgg sets a FigureCanvas size that can be increased or decreased to a
minimum (of perhaps 400 x 300 pixels)
Wx I do not have, is it like Qt or Tk or is there a 4th variation?

I looked at changing GTK so the FigureCanvas could be decreased a while
ago but the problem is that while the axes scale/shrink nicely the text
stays the same size and becomes misaligned or clipped and gives an odd
looking graph. Perhaps thats why Tk allows you shrink the window but
only to a certain size.

Matplotlib currently has 3 (or more) ways of handling window resizing,
which one of these behaviours is 'correct' or most desirable?

Regards
Steve

John, 

Thanks for the tip.  I had thought of checking that as well as I 
drifted off to sleep.  That was it-in spades.   I had 
input two different subplot codes but somehow only one got saved. 
Then I got suckered in by the simplicity of initializing
the list using [ 2*Subpdata() ] where Subpdata is a class
defining the various attributes of a subplot in my program. 
However, I soon found out that this made my problem worse;
Python went into an endless loop.  Again, this gave me two
subplot codes but they were the same.   The simple list 
initialization binds the same instance object to the two 
list entries.  Not what I wanted.  After using a for statement
to initialize, I got my two subplots.  

I am now able to create multiple figures, each with one or more
subplots with my software.  What a great collection of software. 
Thanks for all the work. 

I tried using mx.datetime to create an mx.datetime instance
which worked fine. However, mx2num() fails with python 
claiming an undefined mxdates.  I checked the source and 
found only one  mxdates.  No idea how it should be defined. 
In the mean time I am using a datetime instance but I lose
a small bit of precision with only integer seconds. 

                            Delbert

 
On Tuesday 01 February 2005 05:38 am, John Hunter wrote:
> >>>>> "Delbert" == Delbert D Franz <dd...@lk...> writes:
> 
>     Delbert> I have made great progress with my GUI to plot
>     Delbert> time-series files from my unsteady-flow modeling
>     Delbert> software.  It works when I have one subplot per figure
>     Delbert> but I have not been able to get two subplots (211 and
>     Delbert> 212) to work.  Only the lower subplot appears and the
>     Delbert> trace or line assigned to the upper subplot (211) appears
>     Delbert> in the lower subplot.  The space for the upper subplot
>     Delbert> appears as blank space in the figure.
> 
> You code looks correct on first glance (except did you mean tht title
> to be in the loop?) .  My guess is your data structure has a bug in
> it.  Add a print statement and make sure the loc code is as you think
> it is.
> 
>         print fg.fgs[ifg].sp[isp].loc_code
>         ax = f.add_subplot(fg.fgs[ifg].sp[isp].loc_code)  
> 
> If they are correct, you need to compose a minimum, free standing
> script that replicates your problem and I can take a look.
> 
> Hope this helps,
> JDH
> 
> 

I am working in interactive mode, with the GTKAgg backend. When I make a 
figure, I was not able to resize it on the screen. I tried changing my dpi 
and default figure size in .matplotlibrc, and now I can increase the figure 
size, but not decrease it passed its original size. I changed 
my .matplotlibrc back to its original configuration, and the same behavior 
persists: I can increase the figure size, but not decrease it passed the 
original size. This is also true for the GTK backend, but not for WXAgg.

I rebuilt my atlas libraries today, along with Numeric, and numarray. I dont 
think that should have anything to do with it, but I rebuilt gtk+, pygtk, and 
matplotlib just in case. All are the latest version. 

Everything else seems to work, minimizing, maximizing, closing, but not 
resizing. Does anyone have a suggestion as to what is wrong, or what I might 
check?

 
Thanks,
Darren

> See my previous message about building matplotlib on OS-X. Now that 
> I've done that, I need to give it to some of my coworkers that don't 
> have the dev tools installed, and are aghast at the idea of typing 
> ./configure; make, make install.
>
> So, how do I make a binary distro?

Wow, serendipity!
I was just sitting down to the challenge of building matplotlib on os-x 
myself.  Please keep the list(s) updated on your progress.  I would 
certainly prefer running an install script or mpkg than trying to 
compile all the dependencies.
*sending karma,
    Brendan

______________________________________________________________________ 

Post your free ad now! http://personals.yahoo.ca

On Feb 1, 2005, at 19:04, Chris Barker wrote:

> See my previous message about building matplotlib on OS-X. Now that 
> I've done that, I need to give it to some of my coworkers that don't 
> have the dev tools installed, and are aghast at the idea of typing 
> ./configure; make, make install.
>
> So, how do I make a binary distro? I know I can get distutils to do 
> it, but it won't included the needed extra libs:
>
> freetype
> libpng
> zlib

zlib ships with OS X

> It seems I have three options:
>
> 1) Just give them a tarball of the compiled libs, ready to be dropped 
> into /usr/local (I don't like that option, what If I clobber 
> something?)
>
> 2) Somehow put the libs inside the matplotlib distro

That's possible

> 3) Statically link those libs into the matplotlib extensions

But this is better

> I have no idea how to do 2 or 3, but could probably figure it out.

This is (3)
Make sure you don't have libpng.dylib or libfreetype.dylib sitting 
around on your link paths...

Build them like this:
./configure --disable-shared --enable static

Make sure you only have libpng.a and libfreetype.a on your link paths.  
It will link statically.

> What do folks suggest? Any pointers?

After you've linked matplotlib statically, use bdist_mpkg from py2app 
to make a redistributable .pkg installer for it.  After installing 
py2app, you should have a tool in /usr/local/bin called "bdist_mpkg" 
that will Just Do It without any setup.py modifications to the target 
lib... so go into the matplotlib directory, type bdist_mpkg, and cross 
your fingers that a dist/matplotlib-xx.pkg will appear.  If you have 
any problems beyond that, let me know.

-bob

Chris,

An alternative way to get freetype, libpng and zlib for OS X is to use  
the  the  i-Installer (http://ii2.sourceforge.net/) distribution  
application.  It is primarily for TeX on OS X, but it provides several  
other useful binaries for OS X.

Larry

On Feb 1, 2005, at 3:50 PM, Chris Barker wrote:

> Hi all,
>
> I just got matplotlib working on my box, so I thought I'd post this,  
> so that it will be in the archives, and because I'd like others to try  
> it and let me know if it doesn't work for you.
>
>
> One question for the knowledgeable folks here:
>
> I have /usr/include/zlib.h on my system, but I have no idea where I  
> got it. I do know that it isn't on other systems in my office without  
> the dev tools. Did it come with the dev tools, or something else I  
> installed?
>
> Anyway, here's what worked for me:
>
> Installing matplotlib on OS-X (10.3.7)
>
> Here are my notes as to what it took to get matplotlib (0.71)  
> installed and working on OS-X. I have so far kept a fink-free system,  
> so that's what I've done here as well. I use it with the AGG back end  
> for generating images for a web site, and hopefully with the wx  
> backend for interactive use and embedding in wx Applications.
>
> 1) Requirements:
> -------------------------------------------
> According to the matplotlib install docs  
> (http://matplotlib.sourceforge.net/installing.html), you need the  
> following:
>
> freetype (>= 2.1.7)
> libpng
> zlib
>
> Personally, I've been avoiding Fink, as it doesn't seem to play well  
> with the rest of OS-X, including the Apple supplied Python, so I've  
> looked elsewhere for these libs.
>
> a) Freetype:
> I seem to have it in:
>
> /usr/X11R6/include/freetype2
>
> I don't think I installed it myself, so it probably came with Apple's  
> X11, Which I did install.
>
> However, I seemed to be having problems with that version, so I  
> looked, and it seems to be:
>
> libfreetype.6.3.dylib
>
> Given that freetype2 is currently at version 2.1.9, I have no idea  
> what to make of that! So off to sourceforge to get a new freetype:
>
> http://freetype.sourceforge.net/index2.html
>
> where I got:
>
> freetype-2.1.9.tar.gz
>
> Following the instructions in docs/INSTALL.UNX:
>
> $ ./configure
> $ make
> $ sudo make install
>
> That puts it in /usr/local/..., which is a good place for it.
>
> b) zlib:
> I have:
> /usr/include/zlib.h
>
> I don't know where I might have gotten it, but there it is.
>
> NOTE: I checked on another system in my office, and it doesn't have  
> zlib. It also doesn't have any developer tools installed, so you  
> couldn't build anything on it anyway. If zlib comes with XCode tools,  
> anyone who can build should have it. It also may have come with some  
> other package I've installed, but I have no idea what. Let me know if  
> you know.
>
> c) libpng:
> This, I didn't have on my system, except inside the wxWidgets source  
> tree, so I went looking for it.
>
> Note that you need zlib to compile libpng, so make sure you have that  
> first.
>
> I did a google search for "libpng OS-X". I found:
>
> http://www.libpng.org/pub/png/pngcode.html
>
> Which led me to the libpng sourceforge site.
>
> From there I downloaded:
>
> libpng-1.2.8.tar.gz
>
> unpacked it, and opened a terminal in the libpng-1.2.8 directory, and  
> did:
>
> $ cp scripts/makefile.darwin ./makefile
>
> (note that according to the INSTALL, there is supposed to be a  
> makefile.macosx, but it wasn't there)
>
> I took a look in the makefile, and found:
>
> ZLIBLIB=/usr/local/lib
> ZLIBINC=/usr/local/include
>
> Which is not where zlib is on my system. However, while I can find  
> zlib.h, I couldn't fine the actual lib, so I tried make without  
> changing anything.
>
> $ make
>
> Which seemed to work fine. zlib must be installed in a standard  
> location, and gcc  found it.
>
> $ sudo make install
>
> to install the lib into /usr/local/ (this was specified in the  
> makefile, and it's a good place for it)
>
> $ make test
> and
> $ ./pngtest pngnow.png
>
> Which both seemed to pass.
>
> d) Numeric or numarray:
>
> I already have Numeric installed, from Bob Ippolito's PIMP (MacPython  
> Package Manager) repostitory. (www.undefined.org/python)
>
> 2) Building matplotlib
> ---------------------------------------------------------
> First, I took a look at the matplotlib setup.py. It had "auto" for the  
> back-end flags, so I thought I would give it a try that way:
>
> $python setup.py build
>
> That worked!
> $ sudo python setup.py install
>
> 4) And now to test!
>
> $python
>
> >>> import pylab
> Could not open font file /Library/Fonts/NISC18030.ttf
> No module named pygtk
> PyGTK version 1.99.16 or greater is required to run the GTK Matplotlib  
> backends
>
> This turns out to be because the matlabrc file sets the GTKAgg back  
> end as the default. You have two choices.
>
> 1) you can set the back end before importing pylab.
> >>> import matplotlib
> >>> matplotlib.use('Agg')
> >>> import pylab
>
> This works fine
>
> 2) Edit the matplotlibrc file. I found it in:
> /System/Library/Frameworks/Python.framework/Versions/2.3/share/ 
> matplotlib/.matplotlibrc
>
> Change the line:
> backend      : GTKAgg    # the default backend
> to
> backend      : Agg    # the default backend
>
> And you're all set to make images for the web, etc.
>
> I'm  going to leave getting it to work with wxPython for another day.
>
> -Chris
>
>
> -- 
> Christopher Barker, Ph.D.
> Oceanographer
>                                     		
> NOAA/OR&R/HAZMAT         (206) 526-6959   voice
> 7600 Sand Point Way NE   (206) 526-6329   fax
> Seattle, WA  98115       (206) 526-6317   main reception
>
> Chr...@no...
> _______________________________________________
> Pythonmac-SIG maillist  -  Pyt...@py...
> http://mail.python.org/mailman/listinfo/pythonmac-sig
>
>
Larry Meyn
Aerospace Operations Modeling Office

M/S 210-10
NASA Ames Research Center
Moffett Field, CA 94035-1000

E-mail: Lar...@na...
Phone:  (650) 604-5038
Fax:    (650) 604-0222
E-Fax:  (425) 944-5526 sent via e-mail


Larry Meyn
Aerospace Operations Modeling Office

M/S 210-10
NASA Ames Research Center
Moffett Field, CA 94035-1000

E-mail: Lar...@na...
Phone:  (650) 604-5038
Fax:    (650) 604-0222
E-Fax:  (425) 944-5526 sent via e-mail 

Hi all,

See my previous message about building matplotlib on OS-X. Now that I've 
done that, I need to give it to some of my coworkers that don't have the 
dev tools installed, and are aghast at the idea of typing ./configure; 
make, make install.

So, how do I make a binary distro? I know I can get distutils to do it, 
but it won't included the needed extra libs:

freetype
libpng
zlib

It seems I have three options:

1) Just give them a tarball of the compiled libs, ready to be dropped 
into /usr/local (I don't like that option, what If I clobber something?)

2) Somehow put the libs inside the matplotlib distro

3) Statically link those libs into the matplotlib extensions

I have no idea how to do 2 or 3, but could probably figure it out.

What do folks suggest? Any pointers?

Also, what's up with the Packman databases? I think I saw a note from 
Bob saying he isn't maintaining his. Jack, are you maintaining yours? If 
either of these is being maintained, how would I get matplotlib into it?

thanks,
-Chris

-- 
Christopher Barker, Ph.D.
Oceanographer
                                     		
NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chr...@no...

>>>>> "Chris" == Chris Barker <Chr...@no...> writes:

    Chris> 1) It seems to me that it belongs as a figure (or canvas, I
    Chris> suppose) method.

Yep, I agree.  savefig is now a figure method.  It stores a ref to
its canvas, which you can access (you probably don't need to) as

  fig.canvas

The pylab interface function savefig now forwards the call to
Figure.savefig.

    Chris> 2) Most critically, it uses the "current figure". I can see
    Chris> where this makes sense with Matlab style interactive use,
    Chris> but in a program, I may be building more than one figure at
    Chris> a time, and it makes more sense to me to do something like:

Yes, managing the current state is one of the jobs of pylab.  But I
don't mind tweaking a few things so you can use pylab w/o the current
state features.

    Chris> Sorry to come off as being critical. I am very impressed
    Chris> with your work on matplotlib. There are been many promising
    Chris> starts to a python plotting package, and NONE of them have
    Chris> come even close to what you've done with matplotlib. That's

No worries, I take your suggestions in the right way, as constructive
criticism.  Thanks for the vote of confidence!

    Chris> I hope as I get more familiar with matplotlib, I'll start
    Chris> to contribute code, rather than just questions and
    Chris> criticism.

    Chris> John, do you want patches that add-to/improve the OO
    Chris> interface?

Yes, certainly, but these should be made with an eye to backwards
compatibility and it would be great if the changes included
documentation and/or examples, both of which are somewhat lacking for
the OO interface.  Documenting what you learn, much as you did for the
OSX install, might be as helpful as any improvements in the API.

Thanks,
JDH