What is the recommended way of Installing MPL 0.9.8 on Leopard?

I am a little confused. It seems the only ones having success installing MPL on Leopard are those building from
source. Is this the recommended way to install. I couldn't get the egg to install.

···

**********************************************************************
Samuel M. Smith Ph.D.
2966 Fort Hill Road
Eagle Mountain, Utah 84005-4108
801-768-2768 voice
801-768-2769 fax

**********************************************************************
"The greatest source of failure and unhappiness in the world is
giving up what we want most for what we want at the moment"
**********************************************************************

Samuel M. Smith wrote:

I am a little confused. It seems the only ones having success installing MPL on Leopard are those building from
source. Is this the recommended way to install. I couldn't get the egg to install.

Samuel: There was a recent message (pasted below) on the list from Vincent Noel that describes a workaround for the Leopard egg problem:

"If you rename matplotlib-0.98.0-py2.5-macosx-10.3-fat.egg to
matplotlib-0.98.0-py2.5.egg,
easy_install will install it from the disk.

I'm not sure why this happens. I've noticed lots of Mac OS X eggs with
cruft at the end of the
filename, which prevents their installation through easy_install.
Renaming is all it takes
to install them; still, it's annoying."

HTH,

-Jeff

···

--
Jeffrey S. Whitaker Phone : (303)497-6313
NOAA/OAR/CDC R/PSD1 FAX : (303)497-6449
325 Broadway Boulder, CO, USA 80305-3328

I renamed the egg on sf to “matplotlib-0.98.0-py2.5-macosx-10.3.egg”. For me, setuptools does not try to download the source after installing now. Unfortunately when I try, “easy_install matplotlib”, it still starts downloading the source instead of the osx binary. I am not having luck finding docs on this.

  • Charlie
···

On Sat, Jun 7, 2008 at 3:09 PM, Jeff Whitaker <jswhit@…146…> wrote:

Samuel M. Smith wrote:

I am a little confused. It seems the only ones having success

installing MPL on Leopard are those building from

source. Is this the recommended way to install. I couldn’t get the egg

to install.

Samuel: There was a recent message (pasted below) on the list from

Vincent Noel that describes a workaround for the Leopard egg problem:

"If you rename matplotlib-0.98.0-py2.5-macosx-10.3-fat.egg to

matplotlib-0.98.0-py2.5.egg,

easy_install will install it from the disk.

I’m not sure why this happens. I’ve noticed lots of Mac OS X eggs with

cruft at the end of the

filename, which prevents their installation through easy_install.

Renaming is all it takes

to install them; still, it’s annoying."

HTH,

-Jeff

Jeffrey S. Whitaker Phone : (303)497-6313

NOAA/OAR/CDC R/PSD1 FAX : (303)497-6449

325 Broadway Boulder, CO, USA 80305-3328


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

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Thanks, I got confused by later posts in the same thread that only said to delete the work fat not the whole -macosx-10.3-fat

That still didn't work completely as the permissions were wrong after the install
in /Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/matplotlib-0.98.0-py2.5.egg

drwx------ 10 root admin 340 Jun 7 14:20 matplotlib-0.98.0-py2.5.egg
after
  chmod ugo+rx matplotlib-0.98.0-py2.5.egg

then I could ipython --pylab successfully

Thanks

···

On 07 Jun 2008, at 15:09 , Jeff Whitaker wrote:

Samuel M. Smith wrote:

I am a little confused. It seems the only ones having success
installing MPL on Leopard are those building from
source. Is this the recommended way to install. I couldn't get the egg
to install.

Samuel: There was a recent message (pasted below) on the list from
Vincent Noel that describes a workaround for the Leopard egg problem:

"If you rename matplotlib-0.98.0-py2.5.egg to
matplotlib-0.98.0-py2.5.egg,
easy_install will install it from the disk.

I'm not sure why this happens. I've noticed lots of Mac OS X eggs with
cruft at the end of the
filename, which prevents their installation through easy_install.
Renaming is all it takes
to install them; still, it's annoying."

HTH,

-Jeff

--
Jeffrey S. Whitaker Phone : (303)497-6313
NOAA/OAR/CDC R/PSD1 FAX : (303)497-6449
325 Broadway Boulder, CO, USA 80305-3328

-------------------------------------------------------------------------
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
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

**********************************************************************
Samuel M. Smith Ph.D.
2966 Fort Hill Road
Eagle Mountain, Utah 84005-4108
801-768-2768 voice
801-768-2769 fax

**********************************************************************
"The greatest source of failure and unhappiness in the world is
giving up what we want most for what we want at the moment"
**********************************************************************

I installed numpy 1.10 from the package installer to OS X Leopard 10.5.3.
It installs fine but I get and error (see below) when running the tests.

Does anyone else get this error? Is it fixed in the current svn version?

>>> numpy.test()
Numpy is installed in /Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/numpy
Numpy version 1.1.0
Python version 2.5.2 (r252:60911, Feb 22 2008, 07:57:53) [GCC 4.0.1 (Apple Computer, Inc. build 5363)]

/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/numpy/core/ma.py:609: UserWarning: Cannot automatically convert masked array to numeric because data
     is masked in one or more locations.
   warnings.warn("Cannot automatically convert masked array to "\

FAIL: check_testUfuncRegression (numpy.core.tests.test_ma.TestUfuncs)

···

----------------------------------------------------------------------
Traceback (most recent call last):
   File "/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/numpy/core/tests/test_ma.py", line 692, in check_testUfuncRegression
     self.failUnless(eqmask(ur.mask, mr.mask))
AssertionError

----------------------------------------------------------------------
Ran 1319 tests in 2.475s

FAILED (failures=1)
<unittest._TextTestResult run=1319 errors=0 failures=1>

Numpy is installed in /Library/Frameworks/Python.framework/Versions/
2.5/lib/python2.5/site-packages/numpy
Numpy version 1.1.0

/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-
packages/numpy/core/ma.py:609: UserWarning: Cannot automatically
convert masked array to numeric because data
     is masked in one or more locations.
   warnings.warn("Cannot automatically convert masked array to "\

Samuel,
Something must be wrong with your numpy installation: did you forget to delete
any previous installation of numpy before installing 1.1.0 ? There shouldn't
be any ma.py module in numpy.core. The masked array module now resides in
numpy.ma

FAIL: check_testUfuncRegression (numpy.core.tests.test_ma.TestUfuncs)

Same here: you shouldn't have a numpy.core.tests.test_ma file, but
numpy.ma.tests.test_core.

Please try to reinstall numpy from scratch, or at least delete outdated files.

···

On Saturday 07 June 2008 16:04:13 Samuel M. Smith wrote:

Samuel,
Something must be wrong with your numpy installation: did you forget to delete
any previous installation of numpy before installing 1.1.0 ? There shouldn't
be any ma.py module in numpy.core. The masked array module now resides in
numpy.ma

That was it. I didn't realize I had to delete the old installation first.
After deleting everything with numpy init from /Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages
  and reinstalling, all the tests past cleanly.

Thanks

Charlie Moad wrote:

I am not having luck finding docs on this.

setuptools really don't seem to be living up to their (its?) promise.

Would it be hard to go back to *.mpkg binaries? We'd have to do one for Apple's python, and one for python.org's, but that should be pretty easy, at least for someone running Leopard.

Or is bdist_mpkg not working for MPL anymore?

Charlie, where/how did you get the Universal libs for the dependencies?

-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

Chris.Barker@...259...

I have had horrible luck with bdist_mpkg for the last several months. I did poke at the setuptools code and it looks like we just won’t be able to make it happy with a universal binary. The preferred extensions have ppc or i386 in them depending on your arch. I think the best way to move forward is to just post binaries for each architecture and not have a universal build.
I statically compile in universal builds of libpng and freetype so end-users don’t have to worry about it. I just build these by hand and the process is documented here: http://ipython.scipy.org/moin/MatplotlibOSXBuildNotes

  • Charlie
···

On Mon, Jun 9, 2008 at 1:03 PM, Christopher Barker <Chris.Barker@…259…> wrote:

Charlie Moad wrote:

I am not having luck finding docs on this.

setuptools really don’t seem to be living up to their (its?) promise.

Would it be hard to go back to *.mpkg binaries? We’d have to do one for

Apple’s python, and one for python.org’s, but that should be pretty

easy, at least for someone running Leopard.

Or is bdist_mpkg not working for MPL anymore?

Charlie, where/how did you get the Universal libs for the dependencies?

-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

Chris.Barker@…259…


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

Matplotlib-users@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-users