matplotlib-devel Mailing List for matplotlib

On Nov 29, 2007 10:30 PM, Charles Moad <cw...@gm...> wrote:
> For reference and possible critique, I wrote up notes on how I built a
> Universal matplotlib dist.
Thanks for posting these Charlie, I put these up on the wiki along
side mine, and refer to both on the mpl installing page. So if you
need to update them, the best place is probably the wiki.
http://ipython.scipy.org/moin/MatplotlibOSXBuildNotes

On Nov 30, 2007 6:32 AM, Jeff Whitaker <js...@fa...> wrote:
> John: A small thing - my name is mispelled on the credits page (should
> be one 't'). Also, if you get a chance it would be great if you could
> update the basemap docs to the latest svn (I'll release 0.9.8 soon).
Done.

The following run well . But I just want to make the image, with the x and y
as axis and z as the image value ,show on the figure. In the pylab , we can
just give a command "imshow" ,But here , I fail to do it , I have try
ax.imshow() again and again ,but fail. Can sb give me some advice . Thank
you .
import matplotlib
matplotlib.use("WXAgg")
matplotlib.interactive(True)
from matplotlib.backends.backend_wx import FigureCanvasWx
from matplotlib.figure import Figure
from matplotlib.axes import *
from numpy import *
import wx
from pylab import *
 
#---------------------------------------------------
class DynamicPlot():
 def __init__(self,Frame):
 delta = 0.025
 x = arange(-3.0, 3.0, delta)
 y = arange(-2.0, 2.0, delta)
 X, Y = meshgrid(x, y)
 Z1 = bivariate_normal(X, Y, 1.0, 1.0, 0.0, 0.0)
 Z2 = bivariate_normal(X, Y, 1.5, 0.5, 1, 1)
 Z = 10.0 * (Z2 - Z1) 
 levels = arange(-1.2, 1.6, 0.2) 
 self.fig = Figure((8,8), 75)
 self.canvas = FigureCanvasWx(Frame, -1, self.fig)
 ax=self.fig.add_axes([0.1,0.1,0.8,0.8])
 
ax.contour(X,Y,Z,levels,origin='lower',linewidths=2,extent=(-3,3,-2,2))
app = wx.PySimpleApp()
f=wx.Frame(None,size=(600,600))
f.Show(True)
DynamicPlot(f)
app.MainLoop()
-- 
View this message in context: http://www.nabble.com/How-to-show-matplotlib.image---tf4912839.html#a14064300
Sent from the matplotlib - devel mailing list archive at Nabble.com.

Thank u for your help. I did it as your reply. But some questions appear
again . Can you give me some more advice , your help will be appreciated.
Code are as following: You know when I do "self.ax.images.append(self.im)"
in pylab . It do properly , But ,it fail here. Thank you!
# -*- coding:gb2312 -*- 
import matplotlib
matplotlib.use("WXAgg")
matplotlib.interactive(True)
from matplotlib.backends.backend_wx import FigureCanvasWx
from matplotlib.figure import Figure
from matplotlib.axes import *
from numpy import *
import wx
from matplotlib.image import NonUniformImage
import numpy as npy
 
#---------------------------------------------------
class DynamicPlot():
 def __init__(self,Frame):
 x = npy.arange(-4, 4, 0.005)
 y = npy.arange(-4, 4, 0.005)
 z = npy.sqrt(x[npy.newaxis,:]**2 + y[:,npy.newaxis]**2)
 
 self.fig = Figure((8,8), 75)
 self.canvas = FigureCanvasWx(Frame, -1, self.fig)
 self.ax=self.fig.add_subplot(111)
 self.im = NonUniformImage(self.ax)
 self.im.set_data(x, y, z)
 self.ax.images.append(self.im)
 self.ax.imshow(self.im)
app = wx.PySimpleApp()
f=wx.Frame(None,size=(600,600))
f.Show(True)
DynamicPlot(f)
app.MainLoop()
-- 
View this message in context: http://www.nabble.com/matplotlib.image.AxesImage-can%27t-work.-tf4897267.html#a14057748
Sent from the matplotlib - devel mailing list archive at Nabble.com.

John Hunter wrote:
> For the 0.91 release, I have updated the web site, which had become
> woefully out of date. In addition to some 91 specific things like
> "what's new", credits and updating a bunch of links to point to
> pyplot, I also cleaned out some dead wood (removed numeric and
> numarray references, cleaned up installing and backends to an extent).
> I think a lot more needs to be done to get the site fully modernized
> (eg installing could use a lot more work, goals needs to be
> modernized), but if you see something egregious or otherwise in need
> of attention, let me know, because on these kinds of things it is good
> to strike when the iron is hot.
>
> Also, it has been a long time since I updated the credits page, so
> this too is very stale. If you are not mentioned there and would like
> to be, or if I have left out some significant contributions that you
> would like to see there, please drop me a blurb and I'll update it.
>
> JDH
> 
John: A small thing - my name is mispelled on the credits page (should 
be one 't'). Also, if you get a chance it would be great if you could 
update the basemap docs to the latest svn (I'll release 0.9.8 soon).
-Jeff
-- 
Jeffrey S. Whitaker Phone : (303)497-6313
NOAA/OAR/CDC R/PSD1 FAX : (303)497-6449
325 Broadway Boulder, CO, USA 80305-3328

For reference and possible critique, I wrote up notes on how I built a 
Universal matplotlib dist.
# Build instructions for a matplotlib distribution on OSX
# Using python-2.5 package from python.org
# Getting starting
quaternion:mpl-build cmoad$ pwd
/usr/local/src/mpl-build
quaternion:mpl-build cmoad$ ll
total 11408
-rw-r--r--@ 1 cmoad staff 1.2M Jul 2 18:06 freetype-2.3.5.tar.bz2
-rw-r--r--@ 1 cmoad staff 600K Nov 9 07:31 libpng-1.2.23.tar.bz2
-rw-r--r-- 1 cmoad wheel 3.8M Nov 29 21:47 matplotlib-0.91.1.tar.gz
# Setup environment for Universal builds of libpng and freetype
quaternion:mpl-build cmoad$ export MACOSX_DEPLOYMENT_TARGET=10.4
quaternion:mpl-build cmoad$ export CFLAGS="-arch i386 -arch ppc - 
isysroot /Developer/SDKs/MacOSX10.4u.sdk"
quaternion:mpl-build cmoad$ export LDFLAGS="-arch i386 -arch ppc - 
syslibroot,/Developer/SDKs/MacOSX10.4u.sdk"
# Starting with libpng
quaternion:mpl-build cmoad$ tar -xjf libpng-1.2.23.tar.bz2
quaternion:mpl-build cmoad$ cd libpng-1.2.23
# To avoid errors with "-M" flags and Universal builds, pass "-- 
disable-dependency-tracking" to configure
quaternion:libpng-1.2.23 cmoad$ ./configure --disable-dependency- 
tracking
quaternion:libpng-1.2.23 cmoad$ make -j3
# To make sure we don't link against the dynamic libs, copy the static 
library to the libpng source root
quaternion:libpng-1.2.23 cmoad$ cp .libs/libpng.a ./
# To prove we have a Universal binary
quaternion:libpng-1.2.23 cmoad$ file libpng.a
libpng.a: Mach-O universal binary with 2 architectures
libpng.a (for architecture i386): current ar archive random 
library
libpng.a (for architecture ppc): current ar archive random 
library
# Next to freetype
quaternion:libpng-1.2.23 cmoad$ cd ../
quaternion:mpl-build cmoad$ tar -xjf freetype-2.3.5.tar.bz2
quaternion:mpl-build cmoad$ cd freetype-2.3.5
quaternion:freetype-2.3.5 cmoad$ ./configure
quaternion:freetype-2.3.5 cmoad$ make -j3
# Copy the freetype static lib to its source root as well
quaternion:freetype-2.3.5 cmoad$ cp objs/.libs/libfreetype.a ./
# To prove we have a Universal binary
quaternion:freetype-2.3.5 cmoad$ file libfreetype.a
libfreetype.a: Mach-O universal binary with 2 architectures
libfreetype.a (for architecture i386): current ar archive random 
library
libfreetype.a (for architecture ppc): current ar archive random 
library
# Finally to matplotlib
quaternion:freetype-2.3.5 cmoad$ cd ../
# This OSX package from python.org will build Universal python 
extensions without our help
quaternion:mpl-build cmoad$ unset MACOSX_DEPLOYMENT_TARGET
quaternion:mpl-build cmoad$ unset CFLAGS
quaternion:mpl-build cmoad$ unset LDFLAGS
# Point gcc to our new Universal builds of libpng and freetype
quaternion:mpl-build cmoad$ export CFLAGS="-arch ppc -arch i386 -I/usr/ 
local/src/mpl-build/libpng-1.2.23 -I/usr/local/src/mpl-build/ 
freetype-2.3.5/include"
quaternion:mpl-build cmoad$ export LDFLAGS="-arch ppc -arch i386 -L/ 
usr/local/src/mpl-build/libpng-1.2.23 -L/usr/local/src/mpl-build/ 
freetype-2.3.5"
# Now we are ready to build matplotlib
quaternion:mpl-build cmoad$ tar -xzf matplotlib-0.91.1.tar.gz
quaternion:mpl-build cmoad$ cd matplotlib-0.91.1
# Matplotlib's detected environment for those interested (I already 
installed numpy)
quaternion:matplotlib-0.91.1 cmoad$ python setupegg.py
= 
= 
= 
= 
========================================================================
BUILDING MATPLOTLIB
 matplotlib: 0.91.1
 python: 2.5.1 (r251:54869, Apr 18 2007, 22:08:04) [GCC
 4.0.1 (Apple Computer, Inc. build 5367)]
 platform: darwin
REQUIRED DEPENDENCIES
 numpy: 1.0.4
 freetype2: found, but unknown version (no pkg-config)
OPTIONAL BACKEND DEPENDENCIES
 libpng: found, but unknown version (no pkg-config)
 Tkinter: Tkinter: 50704, Tk: 8.4, Tcl: 8.4
 wxPython: no
 * wxPython not found
 Gtk+: no
 * Building for Gtk+ requires pygtk; you must 
be able
 * to "import gtk" in your build/install 
environment
 Qt: no
 Qt4: no
 Cairo: no
OPTIONAL DATE/TIMEZONE DEPENDENCIES
 datetime: present, version unknown
 dateutil: matplotlib will provide
 pytz: matplotlib will provide
OPTIONAL USETEX DEPENDENCIES
 dvipng: no
 ghostscript: /bin/sh: gs: command not found
 latex: no
EXPERIMENTAL CONFIG PACKAGE DEPENDENCIES
 configobj: matplotlib will provide
 enthought.traits: matplotlib will provide
# Finally build the matplotlib egg
quaternion:matplotlib-0.91.1 cmoad$ python setupegg.py bdist_egg
# The resulting eggs which should be **cross fingers** 10.3+ compatible
quaternion:matplotlib-0.91.1 cmoad$ ll dist/
total 12184
-rw-r--r-- 1 cmoad wheel 5.9M Nov 29 23:03 matplotlib-0.91.1-py2.5- 
macosx-10.3-fat.egg
# END
Attached as well in case line breaks destroy the readability.

 I posted the 0.91.1 source and mac build. Please test it out. 
I held off on the announcement since I haven't built the windows 
binaries.
 I did run into several problems that stemmed from files not 
being mentioned in "MANIFEST.in". Some included "CXX/*.h", "ttconv/ 
*.h" and "setup.cfg.template". Basically these files were not being 
included in the sdist output. I thought I would mention it in case 
people notice missing files.
- Charlie

On Thursday 29 November 2007 10:36:48 pm Darren Dale wrote:
> On Thursday 29 November 2007 9:19:37 pm John Hunter wrote:
> > Great, thanks for he pointers -- in general I would like to give as
> > much credit as possible to everyone who has contributed, so any
> > omissions are merely oversight or laziness and not intentional
> > slights. I've made a few additions to cover your suggestions, and
> > feel free to make additions or updates as they occur to you for you or
> > anyone else.
>
> Fernando wrote TConfig, which is so far the most difficult and elegant part
> of the new config package (aside from Traits itself).
Sorry, one more: James Amundson did the initial work on porting the qt backend 
to qt4.

On Thursday 29 November 2007 9:19:37 pm John Hunter wrote:
> Great, thanks for he pointers -- in general I would like to give as
> much credit as possible to everyone who has contributed, so any
> omissions are merely oversight or laziness and not intentional
> slights. I've made a few additions to cover your suggestions, and
> feel free to make additions or updates as they occur to you for you or
> anyone else.
Fernando wrote TConfig, which is so far the most difficult and elegant part of 
the new config package (aside from Traits itself).

On Nov 29, 2007 4:46 PM, Darren Dale <dar...@co...> wrote:
> Not to be a glory hog, but I spent a fair bit of time on backend_ps and
> backend_qt*. It seems like Eric's fingerprints can be found all over
> matplotlib, but I dont have a suggestion for what to add to his nugget.
Great, thanks for he pointers -- in general I would like to give as
much credit as possible to everyone who has contributed, so any
omissions are merely oversight or laziness and not intentional
slights. I've made a few additions to cover your suggestions, and
feel free to make additions or updates as they occur to you for you or
anyone else.
JDH

On Nov 29, 2007 7:45 PM, Charlie Moad <cw...@gm...> wrote:
> So here's my plan. I just got an iMac a few weeks ago and I had to
> spend a little time getting parallels setup with VS2003... yada yada
> yada. I plan on cutting a 0.91.1 release tomorrow followed shortly by
> windows and mac builds. Hopefully nothing radical has snuck into the
> svn tree since the 0.91.0 build.
No nothing radical, but there have been a few of changes and patches
so at a minimum, be sure and run backend driver to test.
Good luck,
JDH

So here's my plan. I just got an iMac a few weeks ago and I had to
spend a little time getting parallels setup with VS2003... yada yada
yada. I plan on cutting a 0.91.1 release tomorrow followed shortly by
windows and mac builds. Hopefully nothing radical has snuck into the
svn tree since the 0.91.0 build.
- Charlie
On Nov 28, 2007 8:32 AM, Charlie Moad <cw...@gm...> wrote:
> My 1 year old only let me get the source release pushed last night and
> build the mac release. I'll try to get the windows builds posted
> tonight.
>
>
> On Nov 28, 2007 8:23 AM, Rob Hetland <he...@ta...> wrote:
> >
> > On Nov 28, 2007, at 12:03 AM, John Hunter wrote:
> >
> >
> > > I think native tcl/tk is preferable, but this is not a terribly
> > > informed opinion. If you don't hear otherwise from someone else, just
> > > build under that assumption.
> >
> > I am still having issues with Tk on my machine (native? version
> > 8.4.7). In particular, event.key does not register (which I use
> > quite a bit in my interactive grid generation tools). Not even the
> > 'g' to toggle the grid. I also get an error when creating a figure
> > instance:
> >
> > 2007年11月28日 14:19:46.440 Python[19291] *** _NSAutoreleaseNoPool():
> > Object 0x17ca2f30 of class NSCarbonWindowContentView autoreleased
> > with no pool in place - just leaking
> > <matplotlib.figure.Figure instance at 0x711300>
> >
> > Other backends (tried Wx 2.8.6.1 and Qt4) _do_ work fine on the
> > latest svn.
> >
> > I don't think anyone else is experiencing this problem. I reported
> > it earlier, and John said it works fine for him -- nobody else chimed
> > in to say they had the same problem...
> >
> > I just wanted to be sure you all were aware of potential Tk problems
> > before your release.
> >
> > -Rob
> >
> > ----
> > Rob Hetland, Associate Professor
> > Dept. of Oceanography, Texas A&M University
> > http://pong.tamu.edu/~rob
> > phone: 979-458-0096, fax: 979-845-6331
> >
> >
> >
>

On Thursday 29 November 2007 03:34:19 pm John Hunter wrote:
> For the 0.91 release, I have updated the web site, which had become
> woefully out of date. In addition to some 91 specific things like
> "what's new", credits and updating a bunch of links to point to
> pyplot, I also cleaned out some dead wood (removed numeric and
> numarray references, cleaned up installing and backends to an extent).
> I think a lot more needs to be done to get the site fully modernized
> (eg installing could use a lot more work, goals needs to be
> modernized), but if you see something egregious or otherwise in need
> of attention, let me know, because on these kinds of things it is good
> to strike when the iron is hot.
>
> Also, it has been a long time since I updated the credits page, so
> this too is very stale. If you are not mentioned there and would like
> to be, or if I have left out some significant contributions that you
> would like to see there, please drop me a blurb and I'll update it.
Not to be a glory hog, but I spent a fair bit of time on backend_ps and 
backend_qt*. It seems like Eric's fingerprints can be found all over 
matplotlib, but I dont have a suggestion for what to add to his nugget.
Darren

On Nov 29, 2007 4:02 PM, Ludwig Schwardt <lud...@gm...> wrote:
> I'm glad you liked my Tk patch! Along the same line, I patched axes3d.py to
> allow the creation of non-interactive 3D plots.
OK, since noone else is maintaining this package right now, what you
say goes :-)
Committed in r4509
JDH

On Nov 29, 2007 2:30 PM, Michael Droettboom <md...@st...> wrote:
> There are some Gtk-specific functions at the bottom of mlab.py. It
> tries to "import gtk" and then will either define them or not.
Ahh, I hadn't considered import time in the try/except block. In
light of this, we probably will want to put them in another module. I
was inclined to something like matplotlib.gui.gtktools, but tere is
an analogous problem with mlab.rec2excel which conditionally imports
pyExcelerator (you probably did not notice it in your timing since it
wasn't installed). Perhaps matplotlib.external or
matplotlib.optional as top-level packages for both (and related).
Something like
matplotlib.optional.exceltools
matplotlib.optional.gtktools
This will serve the dual purpose of cleaning mlab and making Jarrod's
job a bit easier when he decides what to bring over to scipy.
JDH

For the 0.91 release, I have updated the web site, which had become
woefully out of date. In addition to some 91 specific things like
"what's new", credits and updating a bunch of links to point to
pyplot, I also cleaned out some dead wood (removed numeric and
numarray references, cleaned up installing and backends to an extent).
 I think a lot more needs to be done to get the site fully modernized
(eg installing could use a lot more work, goals needs to be
modernized), but if you see something egregious or otherwise in need
of attention, let me know, because on these kinds of things it is good
to strike when the iron is hot.
Also, it has been a long time since I updated the credits page, so
this too is very stale. If you are not mentioned there and would like
to be, or if I have left out some significant contributions that you
would like to see there, please drop me a blurb and I'll update it.
JDH

There are some Gtk-specific functions at the bottom of mlab.py. It 
tries to "import gtk" and then will either define them or not.
On one of my recent profiling fests, I noticed that this "import gtk" 
(if gtk is installed, and running with some non-Gtk backend) accounts 
for between 1-2% of the total running time over all the examples in 
backend_driver.py. Not a huge amount, but it seems like it might be a 
good idea to avoid importing something as large as gtk unless we 
actually need to.
We can either import gtk "on demand", when one of these gtk-related 
functions is called, or perhaps better yet move this to another module, 
that would only be imported when needed. In light of recent 
pylab.py/mlab.py refactorings, are there any opinions on this?
Cheers,
Mike
-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

On Nov 29, 2007 2:14 PM, Michael Droettboom <md...@st...> wrote:
> This seems like a good idea to me, and it works for me (I can build mpl
> with the TkAgg extension from a console with X completely shutdown on
> RHEL4...).
I tested in my normal X11 enabled solaris build environment and in a
remote no X linux session and it went through fine. I can test on OX
X tonight.
> John -- does the timing for this work with you release plans?
Yep -- since we probably will be waiting until next week to get 91.1
out, this will give us ample opportunity to hit obvious problems with
this patch.
JDH

Ah! I see you just applied it yourself, John.
Consider the question answered!
(And again, thanks for the patch, Ludwig.)
Cheers,
Mike
Michael Droettboom wrote:
> This seems like a good idea to me, and it works for me (I can build mpl 
> with the TkAgg extension from a console with X completely shutdown on 
> RHEL4...).
> 
> John -- does the timing for this work with you release plans? (It 
> probably needs some testing in a bunch of different environments.) If 
> so, I'm happy to commit this.
> 
> Cheers,
> Mike
> 
> James Amundson wrote:
>> On 2007年11月29日 18:13:39 +0200
>> Ludwig Schwardt <lud...@gm...> wrote:
>>
>>> I've reworked the Tcl/Tk checking code in setupext.py (see attached
>>> patch). It is now possible to build matplotlib with Tk support without
>>> requiring a running X server. This is useful for doing autobuilds
>>> (e.g. as done by Debian) and for building a package on one machine to
>>> be installed and used on another.
>>>
>>> This seems to be an old issue... (see matplotlib-devel, "building
>>> matplotlib without X-server connection," 11 Nov 2004)
>> I would like to register my strong support for this patch. The bug in
>> question also prevents one from doing "python setup.py bdist_rpm" on
>> rpm-based systems. I recently tried to debug this myself and got as far
>> as identifying exactly the problem Ludwig describes, but ran out of
>> time/energy before coming up with the fix. I am very grateful to Ludwig
>> for having fixed the problem. I hope his patch will go in soon.
>>
>> --Jim Amundson
>>
>>
>> -------------------------------------------------------------------------
>> SF.Net email is sponsored by: The Future of Linux Business White Paper
>> from Novell. From the desktop to the data center, Linux is going
>> mainstream. Let it simplify your IT future.
>> http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
>> _______________________________________________
>> Matplotlib-devel mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
> 
-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

On Nov 29, 2007 10:13 AM, Ludwig Schwardt <lud...@gm...> wrote:
> Hi,
>
> Is there any chance that the patch I submitted on 31 October on this
> list could still make it into 0.91.0?
91.0 is already out, but I just committed it so it will make it in for 91.1.
> On an aside, what is the best way to submit patches to mpl? Via the
> bug-tracker, or as attached files or direct in-line in mpl-devel
> mails?
I advise posting here first -- more developers read this list so it is
more likely to get prompt attention. If it doesn't, put it up on the
sf site and send a reminder here with a link to the patch. That way
it will be there when some developer goes to catch up on patches.
Thanks,
JDH\

This seems like a good idea to me, and it works for me (I can build mpl 
with the TkAgg extension from a console with X completely shutdown on 
RHEL4...).
John -- does the timing for this work with you release plans? (It 
probably needs some testing in a bunch of different environments.) If 
so, I'm happy to commit this.
Cheers,
Mike
James Amundson wrote:
> On 2007年11月29日 18:13:39 +0200
> Ludwig Schwardt <lud...@gm...> wrote:
> 
>> I've reworked the Tcl/Tk checking code in setupext.py (see attached
>> patch). It is now possible to build matplotlib with Tk support without
>> requiring a running X server. This is useful for doing autobuilds
>> (e.g. as done by Debian) and for building a package on one machine to
>> be installed and used on another.
>>
>> This seems to be an old issue... (see matplotlib-devel, "building
>> matplotlib without X-server connection," 11 Nov 2004)
> 
> I would like to register my strong support for this patch. The bug in
> question also prevents one from doing "python setup.py bdist_rpm" on
> rpm-based systems. I recently tried to debug this myself and got as far
> as identifying exactly the problem Ludwig describes, but ran out of
> time/energy before coming up with the fix. I am very grateful to Ludwig
> for having fixed the problem. I hope his patch will go in soon.
> 
> --Jim Amundson
> 
> 
> -------------------------------------------------------------------------
> SF.Net email is sponsored by: The Future of Linux Business White Paper
> from Novell. From the desktop to the data center, Linux is going
> mainstream. Let it simplify your IT future.
> http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
-- 
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA

On 2007年11月29日 18:13:39 +0200
Ludwig Schwardt <lud...@gm...> wrote:
> I've reworked the Tcl/Tk checking code in setupext.py (see attached
> patch). It is now possible to build matplotlib with Tk support without
> requiring a running X server. This is useful for doing autobuilds
> (e.g. as done by Debian) and for building a package on one machine to
> be installed and used on another.
> 
> This seems to be an old issue... (see matplotlib-devel, "building
> matplotlib without X-server connection," 11 Nov 2004)
I would like to register my strong support for this patch. The bug in
question also prevents one from doing "python setup.py bdist_rpm" on
rpm-based systems. I recently tried to debug this myself and got as far
as identifying exactly the problem Ludwig describes, but ran out of
time/energy before coming up with the fix. I am very grateful to Ludwig
for having fixed the problem. I hope his patch will go in soon.
--Jim Amundson

matplotlib-0.91.0: Added additional include paths for aix5 to 
setupext.py.
freetype-2.3.5 because the system one was compiled old and compiled 
32-bit; did have to make one patch.
AIX 5.3
I'm not using gcc, because the project I'm trying to help uses an 
xlc-compiled Python. (I did read somewhere that someone was successful
compiling matplotlib using gcc.) 
xlc does not seem to give a version number readily.
Lila
On 2007年11月29日, John Hunter wrote:
> Does anyone have any thoughts on this? Lila, can you provide
> matplotlib, gcc and AIX version/platform info?
> 
> 
> ---------- Forwarded message ----------
> From: ch...@ll... <ch...@ol...>
> Date: Nov 29, 2007 11:49 AM
> Subject: problem compiling matplotlib on AIX
> To: jd...@gm...
> 
> 
> John,
> 
> I thought I'd go back and look at AIX again.
> Any hints on the following errors?
> 
> running build_ext
> building 'matplotlib.ft2font' extension
> xlc -q64 -DNDEBUG -O -I/usr/apps/link/include -I/usr/local/include
> -I/usr/include -I. -I/usr/apps/link/include/freetype2
> -I/usr/local/include/freetype2 -I/usr/include/freetype2 -I./freetype2
> -I/usr/gapps/python/aix_5_64_fed/opt-2.5/include/python2.5 -c
> src/ft2font.cpp -o build/temp.aix-5.3-2.5/src/ft2font.o
> "/usr/vacpp/include/xlocinfo.h", line 140.23: 1540-0040 (S) The text
> "_Ptr" is unexpected. "__xlocale_ptr" may be undeclared or ambiguous.
> "/usr/vacpp/include/xlocinfo.h", line 149.23: 1540-0040 (S) The text
> "_Ptr" is unexpected. "__xlocale_ptr" may be undeclared or ambiguous.
> "/usr/vacpp/include/xlocinfo.h", line 161.23: 1540-0040 (S) The text
> "_Ptr" is unexpected. "__xlocale_ptr" may be undeclared or ambiguous.
> "/usr/vacpp/include/xlocale", line 455.22: 1540-0138 (S) The undefined
> template "codecvt" must not be explicitly instantiated.
> "/usr/vacpp/include/xlocale", line 456.22: 1540-0138 (S) The undefined
> template "codecvt" must not be explicitly instantiated.
> "/usr/vacpp/include/xlocale", line 760.21: 1540-0138 (S) The undefined
> template "ctype" must not be explicitly instantiated.
> "/usr/vacpp/include/xlocale", line 768.14: 1540-0120 (S) The out-of-line
> member definition "std::ctype<wchar_t>" of an explicit specialization
> should not use a template prefix.
> "./CXX/Objects.hxx", line 1653.96: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
> require "typename" qualification.
> "./CXX/Objects.hxx", line 1654.96: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
> require "typename" qualification.
> "./CXX/Objects.hxx", line 1655.96: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
> require "typename" qualification.
> "./CXX/Objects.hxx", line 1656.96: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
> require "typename" qualification.
> "./CXX/Objects.hxx", line 1657.96: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
> require "typename" qualification.
> "./CXX/Objects.hxx", line 1658.96: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
> require "typename" qualification.
> "./CXX/Objects.hxx", line 1660.102: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
> may require "typename" qualification.
> "./CXX/Objects.hxx", line 1661.102: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
> may require "typename" qualification.
> "./CXX/Objects.hxx", line 1662.102: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
> may require "typename" qualification.
> "./CXX/Objects.hxx", line 1663.102: 1540-0140 (S) The text "&" is
> unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
> may require "typename" qualification.
> ...
> 
> "./CXX/Objects.hxx", line 2702.102: 1540-0140 (S) The text "&" is
> unexpected. "MapBase<T>::const_iterator" may be undeclared, ambiguous, or
> may require "typename" qualification.
> "src/ft2font.cpp", line 951.27: 1540-0274 (S) The name lookup for "__cos"
> did not find a declaration.
> "src/ft2font.cpp", line 952.27: 1540-0274 (S) The name lookup for "__sin"
> did not find a declaration.
> error: command 'xlc' failed with exit status 1
> 
> 
> Thanks,
> Lila
> 
> --
> Lila Chase
> lc...@ll...
> 925-422-4086
> 
-- 
Lila Chase
lc...@ll...
925-422-4086

Does anyone have any thoughts on this? Lila, can you provide
matplotlib, gcc and AIX version/platform info?
---------- Forwarded message ----------
From: ch...@ll... <ch...@ol...>
Date: Nov 29, 2007 11:49 AM
Subject: problem compiling matplotlib on AIX
To: jd...@gm...
John,
I thought I'd go back and look at AIX again.
Any hints on the following errors?
running build_ext
building 'matplotlib.ft2font' extension
xlc -q64 -DNDEBUG -O -I/usr/apps/link/include -I/usr/local/include
-I/usr/include -I. -I/usr/apps/link/include/freetype2
-I/usr/local/include/freetype2 -I/usr/include/freetype2 -I./freetype2
-I/usr/gapps/python/aix_5_64_fed/opt-2.5/include/python2.5 -c
src/ft2font.cpp -o build/temp.aix-5.3-2.5/src/ft2font.o
"/usr/vacpp/include/xlocinfo.h", line 140.23: 1540-0040 (S) The text
"_Ptr" is unexpected. "__xlocale_ptr" may be undeclared or ambiguous.
"/usr/vacpp/include/xlocinfo.h", line 149.23: 1540-0040 (S) The text
"_Ptr" is unexpected. "__xlocale_ptr" may be undeclared or ambiguous.
"/usr/vacpp/include/xlocinfo.h", line 161.23: 1540-0040 (S) The text
"_Ptr" is unexpected. "__xlocale_ptr" may be undeclared or ambiguous.
"/usr/vacpp/include/xlocale", line 455.22: 1540-0138 (S) The undefined
template "codecvt" must not be explicitly instantiated.
"/usr/vacpp/include/xlocale", line 456.22: 1540-0138 (S) The undefined
template "codecvt" must not be explicitly instantiated.
"/usr/vacpp/include/xlocale", line 760.21: 1540-0138 (S) The undefined
template "ctype" must not be explicitly instantiated.
"/usr/vacpp/include/xlocale", line 768.14: 1540-0120 (S) The out-of-line
member definition "std::ctype<wchar_t>" of an explicit specialization
should not use a template prefix.
"./CXX/Objects.hxx", line 1653.96: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.
"./CXX/Objects.hxx", line 1654.96: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.
"./CXX/Objects.hxx", line 1655.96: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.
"./CXX/Objects.hxx", line 1656.96: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.
"./CXX/Objects.hxx", line 1657.96: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.
"./CXX/Objects.hxx", line 1658.96: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.
"./CXX/Objects.hxx", line 1660.102: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
may require "typename" qualification.
"./CXX/Objects.hxx", line 1661.102: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
may require "typename" qualification.
"./CXX/Objects.hxx", line 1662.102: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
may require "typename" qualification.
"./CXX/Objects.hxx", line 1663.102: 1540-0140 (S) The text "&" is
unexpected. "SeqBase<T>::const_iterator" may be undeclared, ambiguous, or
may require "typename" qualification.
...
"./CXX/Objects.hxx", line 2702.102: 1540-0140 (S) The text "&" is
unexpected. "MapBase<T>::const_iterator" may be undeclared, ambiguous, or
may require "typename" qualification.
"src/ft2font.cpp", line 951.27: 1540-0274 (S) The name lookup for "__cos"
did not find a declaration.
"src/ft2font.cpp", line 952.27: 1540-0274 (S) The name lookup for "__sin"
did not find a declaration.
error: command 'xlc' failed with exit status 1
Thanks,
Lila
--
Lila Chase
lc...@ll...
925-422-4086

Hi,
Is there any chance that the patch I submitted on 31 October on this
list could still make it into 0.91.0?
On an aside, what is the best way to submit patches to mpl? Via the
bug-tracker, or as attached files or direct in-line in mpl-devel
mails?
To refresh your memory on the patch:
I've reworked the Tcl/Tk checking code in setupext.py (see attached
patch). It is now possible to build matplotlib with Tk support without
requiring a running X server. This is useful for doing autobuilds
(e.g. as done by Debian) and for building a package on one machine to
be installed and used on another.
This seems to be an old issue... (see matplotlib-devel, "building
matplotlib without X-server connection," 11 Nov 2004)
The patch also fixes a potential hang, which happens when I try to
build the latest matplotlib on Debian Etch with Python 2.4, on a
system with libtk8.4-dev installed but no running X server. The build
should have continued without TkAgg support, but it hangs. The hang
occurs because the Tcl/Tk checking code calls Tkinter.Tk() multiple
times, which is not a good idea. In fact, just running the following
snippet hangs the interpreter in this scenario:
import Tkinter
tk = Tkinter.Tk()
tk = Tkinter.Tk()
The patch applies to setupext.py in revision r4501. I checked it with
matplotlib r4064, using Python 2.4 on Debian Etch, and Python 2.5 on
Mac OS X 10.4.10. It hasn't been checked on Python 2.3 or on
Windows.
Regards,
Ludwig