matplotlib-devel Mailing List for matplotlib (Page 3)

On 9/7/06, Andrew Straw <str...@as...> wrote:
> Hey Charlie, I totally appreciate the effort you put into making these
> releases, particularly on Windows, where I must admit, I have a faint
> heart...
>
> But I found a couple issues (neither require a re-release, but just to
> be aware of them next time):
>
> 1) there are several .pyc files left in the .tar.gz release
> 2) the setup.cfg file in the release specifies, in the egg_info section,
> "tag_svn_revision = 1", which makes any further attempts to do python
> setup.py sdist with setuptools result in a package versioned
> matplotlib-0.87.5-r0, which isn't right. (Basically, this setup.cfg file
> is telling setuptools that this is a development version and it should
> tag the svn version. Not being in a subversion tree, it finds the
> version as 0.) I made a note a few weeks ago in the DEVNOTES file
> (section "making a release") to remove the setup.cfg file, or at least
> this section, prior to making a release.
I agree with all your points and will look out for them in the future.
 I typically build the sdist without setuptools, but I guess some
lingering files are getting included.
- Charlie

Andrew Straw wrote:
> 1) there are several .pyc files left in the .tar.gz release
> 2) the setup.cfg file in the release specifies, in the egg_info section,
> "tag_svn_revision = 1", which makes any further attempts to do python
> setup.py sdist with setuptools result in a package versioned
> matplotlib-0.87.5-r0, which isn't right. (Basically, this setup.cfg file
> is telling setuptools that this is a development version and it should
> tag the svn version. Not being in a subversion tree, it finds the
> version as 0.) I made a note a few weeks ago in the DEVNOTES file
> (section "making a release") to remove the setup.cfg file, or at least
> this section, prior to making a release.
> 
3) in lib/matplotlib.egg-info/PKG-INFO, the version is specified as
0.87.5-r2761
Issues 2 and 3 appear to stem from matplotlib attempting to use
setuptools for those who would, but not requiring it. I think the
cleanest solution would be to make the actual sdist release using
setuptools -- this should bring these issues into the open rather than
lurking for someone like me to find them.

Hey Charlie, I totally appreciate the effort you put into making these
releases, particularly on Windows, where I must admit, I have a faint
heart...
But I found a couple issues (neither require a re-release, but just to
be aware of them next time):
1) there are several .pyc files left in the .tar.gz release
2) the setup.cfg file in the release specifies, in the egg_info section,
"tag_svn_revision = 1", which makes any further attempts to do python
setup.py sdist with setuptools result in a package versioned
matplotlib-0.87.5-r0, which isn't right. (Basically, this setup.cfg file
is telling setuptools that this is a development version and it should
tag the svn version. Not being in a subversion tree, it finds the
version as 0.) I made a note a few weeks ago in the DEVNOTES file
(section "making a release") to remove the setup.cfg file, or at least
this section, prior to making a release.
Cheers!
Andrew
Charlie Moad wrote:
> Sorry all for the error. I just uploaded "working" (at least for me)
> versions to sourceforge again. For those who are sicking of playing
> with mirrors here is a direct download. Don't expect this link to be
> good for a long time though.
>
> http://euclid.uits.iupui.edu/mplfiles/
>
> - Charlie
>
> On 9/6/06, Charlie Moad <cw...@gm...> wrote:
> 
>> The source error must of propagated to those builds. I will post new
>> ones shortly.
>>
>> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> 
>>> Well the thread on the devel list that I referred to explicitly has the
>>> win32 exe in its title ("Missing __init__.py in
>>> matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
>>> much says it all.
>>>
>>> There's also a recent post on the numpy list that sounds like it's maybe
>>> the same problem (quote):
>>> """
>>> Is there a compatible matplotlib as well? I was o.k. with mpl from
>>> enthought until I switched numerix to numpy. That made mpl unhappy.
>>> I downloaded 0.87.5 but I broke something in the process because now
>>> even switching back to Numeric doesn't make mpl happy.
>>> """
>>>
>>> Thanks,
>>> Sven
>>>
>>> Charlie Moad schrieb:
>>> 
>>>> That error was relating to the source release. Can you please post
>>>> your error for the binary?
>>>>
>>>> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>>>> 
>>>>> Charlie Moad schrieb:
>>>>> 
>>>>>> Minor rev bump for numpy 1.0b5 compatibility. This release should
>>>>>> remain compatible with future 1.0 releases of numpy.
>>>>>>
>>>>>> 
>>>>> I keep running into the ImportError problem described on the devel list,
>>>>> with the win32 2.4 binary (exe). Is that still just a case of waiting
>>>>> for the mirrors to update, or is there a deeper problem? Are there
>>>>> alternative download links? How to tell whether it's the "right" binary,
>>>>> since version numbers are the same?
>>>>>
>>>>> Thanks for your help,
>>>>> Sven
>>>>>
>>>>> 
>>> 
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
> 

Interesting ! The lack of proper 3D plotting in scipy is a major problem
in my eyes.
I would like to point out tvtk ( http://scipy.org/Cookbook/MayaVi/tvtk)
that is a great framework for developing 3D plotting tools in python. One
example is the new version of the famous Mayavi, but something "command
line driven", like pylab, can be easily implemented, as the mlab module (
http://scipy.org/Cookbook/MayaVi/mlab ) demonstrates. These rely on the
ETS (enthought tool suite) which is currently a very haevy dependency if
you are not using the enthought python distribution, but enthought is
working on repackaging this as eggs.
I think that using tvtk as a backend will allow faster development than
raw OpenGL, and code reuse with Mayavi2 and Co. A lot of very well
thought work has gone in tvtk and it needs little work to come out as a
useful plotting package. I coded such a package in a quick and dirty way
for myself and have always wanted to find the time to polish it and
propose its integration to tvtk. I send it as an example of what one can
do in an afternoon with tvtk. (see
https://mail.enthought.com/pipermail/enthought-dev/2006-July/002140.html
for a discussion of this module)
 Best regards,
Ga=EBl
************************************************
import os
os.environ['NUMERIX'] =3D 'numpy'
from scipy import arange, ravel, amax, amin, zeros, column_stack, sqrt, o=
nes
from enthought.tvtk.tools import mlab
class figure3d:
 """ A window for displaying 3D objects """
 title =3D ''
 xlabel =3D 'x'
 ylabel =3D 'y'
 zlabel =3D 'z'
 objects =3D []
 background =3D (1.0,1.0,1.0)
 foreground =3D (0.0,0.0,0.0)
 fig =3D None
 def update(self):
 """Updates the figure"""
 if self.fig:
 hasTitle=3D0
 # Update the title
 for object in self.fig.objects:
 if str(type(object))=3D=3D"<class 'enthought.tvtk.tools.m=
lab.Title'>":
 object.text =3D self.title
 object.text_actor.property.color=3Dself.foreground
 hasTitle=3D1
 if not self.title =3D=3D '' and not hasTitle:
 t=3Dmlab.Title()
 t.text=3Dself.title
 t.text_actor.property.color=3Dself.foreground
 self.fig.add(t)
 # Update the labels
 self.fig.objects[0].axis.x_label=3Dself.xlabel
 self.fig.objects[0].axis.y_label=3Dself.ylabel
 self.fig.objects[0].axis.z_label=3Dself.zlabel
 # Update the colors
 self.fig.renwin.background=3Dself.background
 for object in self.fig.objects[0].actors:
 object.property.color=3Dself.foreground
 self.fig.renwin.render()
 =20
 def show(self):
 """ Shows the figure, if not already displayed """
 if not self.fig:
 self.fig =3D mlab.figure(browser=3DFalse)
 self.fig.renwin.x_plus_view()
 self.fig.renwin.camera.azimuth(-62)
 self.fig.renwin.camera.elevation(19.5)
 self.update()
 def surf(self,z, transparency=3D0.0, transparent=3D0):
 """ Plot a 3d surface from a 2D array """
 y=3Darange(0,z.shape[0])
 x=3Darange(0,z.shape[1])
 # Flatten the matrix
 z=3Dravel(z)
 Zmax=3Damax(z)-amin(z)
 XYmax=3Damax((amax(y)-amin(y),amax(x)-amin(y)))
 Zscale=3DXYmax/float(Zmax)*0.5
 s=3Dmlab.SurfRegular(x,y,z,scale=3D[1.0,1.0,Zscale])
 s.scalar_bar.title=3D''
 s.show_scalar_bar=3DTrue
 s.scalar_bar.orientation=3D"horizontal"
 s.scalar_bar.property.color=3Dself.foreground
 # LUT means "Look-Up Table", it give the mapping between scalar v=
alue and
 # color
 s.lut_type=3D'blue-red'
 transparency=3D1.0-transparency
 s.lut.alpha=3Dtransparency
 s.lut.alpha_range=3D(transparency,transparency)
 if transparent:
 s.lut.alpha_range=3D(0.0,1.0)
 if not self.fig:
 self.show()
 # Scale properly the box outline
 self.fig.objects[0].axis.set_ranges((amin(x),amax(x),amin(y),amax=
(y),amin(z),amax(z)))
 self.fig.objects[0].axis.use_ranges=3D1
 self.fig.add(s)
 def plot3(self, x, y, z, color=3D(0,0.5,1.0), linewidth=3D1):
 """ Plot a 3D line from 3 sets of coordinates """
 assert len(x)=3D=3D len(y) and len(y)=3D=3Dlen(z), "x,y,z must ha=
ve same number of coordinates"
 x=3Dravel(x)
 y=3Dravel(y)
 z=3Dravel(z)
 size=3Dmax((amax(x)-amin(x),amax(y)-amin(y),amax(z)-amin(z)))
 pts =3D zeros((len(x), 3), 'd')
 pts[:,0], pts[:,1], pts[:,2] =3D x, y, z
 linewidth=3D0.05*linewidth/size
 l =3D mlab.Line3(pts,color=3Dcolor, radius=3Dlinewidth)
 if not self.fig:
 self.show()
 self.fig.add(l)
 def edit_traits(self):
 self.fig.edit_traits()
 def quiver3d(self, x, y, z, vx, vy, vz, scalars=3DNone, color=3DNone,=
 autoscale=3D1):
 """ Displays a plot of arrows located at x, y, z, and of
 coordinates vx, vy, vz"""
 assert len(x) =3D=3D len(y) and len(y) =3D=3D len(z) and len(z)=3D=
=3D len(vx) and len(vx)=3D=3Dlen(vy) and len(vy)=3D=3Dlen(vz), "coordinat=
e vectors must be of same length"
 x=3Dravel(x)
 y=3Dravel(y)
 z=3Dravel(z)
 vx=3Dravel(vx)
 vy=3Dravel(vy)
 vz=3Dravel(vz)
 positions =3D column_stack((x,y,z))
 norms =3D vx**2 + vy**2 + vz**2
 norms =3D sqrt(norms.astype(float))
 normsmax =3D amax(norms)
 if autoscale:
 vx=3Dvx.astype(float)/normsmax
 vy=3Dvy.astype(float)/normsmax
 vz=3Dvz.astype(float)/normsmax
 vectors =3D column_stack((vx,vy,vz))
 if scalars is not None:
 # scalars overrides color
 color =3D (0.0,0.5,1.0)
 scalars =3D ravel(scalars)
 scalars =3D scalars + amin(scalars)
 scalars =3D scalars.astype(float)/amax(scalars)
 elif not color:
 scalars =3D norms/normsmax
 color =3D (0.0,0.5,1.0)
 g =3D mlab.Arrows(positions,vectors=3Dvectors,scalars=3Dscalars,c=
olor=3Dcolor)
 g.glyph.scale_mode=3D'scale_by_vector'
 if not self.fig:
 self.show()
 self.fig.add(g)
 def scatter3d(self, x, y, z, scalars=3DNone, size=3DNone, color=3D(0.=
0,0.5,1.0), autoscale=3D1):
 """ Displays a bunch of spheres given by their 3d position.
 Optional scalar matrix defines their color and optional size
 matrix their radius"""
 assert len(x) =3D=3D len(y) and len(y) =3D=3D len(z), "coordinate=
 vectors must be of same length"
 x=3Dravel(x)
 y=3Dravel(y)
 z=3Dravel(z)
 positions =3D column_stack((x,y,z))
 if scalars is not None:
 scalars =3D ravel(scalars)
 scalars =3D scalars + amin(scalars)
 scalars =3D scalars.astype(float)/amax(scalars)
 if size is not None:
 size=3Dravel(size)
 size =3D size + amin(size)
 if autoscale:
 size =3D 1.5*size.astype(float)/amax(size)
 vectors =3D column_stack((size,size,size))
 else:
 vectors =3D None
 g =3D mlab.Spheres(positions,vectors=3Dvectors,scalars=3Dscalars,=
color=3Dcolor)
 if size is not None:
 g.glyph.scale_mode=3D'scale_by_vector'
 if not self.fig:
 self.show()
 self.fig.add(g)
 def savefig(self, filename, **kw_args):
 """Saves rendered scene to one of several image formats=20
 depending on the specified extension of the given filename.
 Any extra keyword arguments are passed along to the respective
 save method."""
 self.fig.renwin.save(filename, **kw_args)
def surf(z,transparency=3D0.0,transparent=3D0):
 """ Plot a 3d surface from a 2D array """
 f =3D figure3d()
 f.surf(z,transparency=3Dtransparency,transparent=3Dtransparent)
 return f
 =20
def plot3(x, y, z, color=3D(0,0.5,1.0), linewidth=3D0.2):
 """ Plot a 3D line from 3 sets of coordinates """
 f =3D figure3d()
 f.plot3(x,y,z,color=3Dcolor, linewidth=3Dlinewidth)
 return f
def scatter3d(x, y, z, scalars=3DNone, size=3DNone, color=3D(0.0,0.5,1.0)=
, autoscale=3D1):
 """ Displays a bunch of points given by their 3d position.
 Optional scalar matrix defines their color"""
 f =3D figure3d()
 f.scatter3d(x ,y ,z, scalars=3Dscalars, size=3Dsize, color=3Dcolor, a=
utoscale=3Dautoscale)
 return f
def quiver3d(x, y, z, vx, vy, vz, scalars=3DNone, color=3DNone, autoscale=
=3D1):
 """ Displays a plot of arrows located at x, y, z, and of
 coordinates vx, vy, vz"""
 f =3D figure3d()
 f.quiver3d(x ,y ,z, vx, vy, vz, scalars=3Dscalars, color=3Dcolor, aut=
oscale=3Dautoscale)
 return f

Hi, I've developed a simple 3D plotting module which u can find at =
http://www.scipy.org/WilnaDuToit . There's still alot of work to be =
done, so please feel free to play/use/extend/break it if u find it =
useful and send comments/enquiries/fixes etc to wi...@su....
=20
Hope someone can use it :)
=20
Wilna du Toit
=20

Hi list,
I downloaded the latest version from the link (see under). With 0.87.3 our
application worked fine, but with with 0.87.5 we get the following:
Traceback (most recent call last):
 File "C:\Python24\PanelCheck_V.1.1.0\PanelCheck.py", line 222, in ?
 import PanelCheck_GUI
 File "C:\Python24\PanelCheck_V.1.1.0\PanelCheck_GUI.py", line 10, in ?
 from Line_Plot import *
 File "C:\Python24\PanelCheck_V.1.1.0\Line_Plot.py", line 3, in ?
 from Plot_Setup import *
 File "C:\Python24\PanelCheck_V.1.1.0\Plot_Setup.py", line 9, in ?
 from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg as
FigureCanvas
 File
"C:\Python24\Lib\site-packages\matplotlib\backends\backend_wxagg.py", line
21, in ?
 from matplotlib.figure import Figure
 File "C:\Python24\Lib\site-packages\matplotlib\figure.py", line 5, in ?
 from artist import Artist
 File "C:\Python24\Lib\site-packages\matplotlib\artist.py", line 4, in ?
 from transforms import identity_transform
 File "C:\Python24\Lib\site-packages\matplotlib\transforms.py", line 223,
in ?
 from _transforms import Value, Point, Interval, Bbox, Affine
 File "C:\Python24\Lib\site-packages\matplotlib\_transforms.py", line 1,
in ?
 import sys, numerix
 File "C:\Python24\Lib\site-packages\matplotlib\numerix\__init__.py", line
145, in ?
 __import__('fft', g, l)
 File "C:\Python24\Lib\site-packages\matplotlib\numerix\fft\__init__.py",
line 11, in ?
 from numpy.dft.old import *
ImportError: No module named old
Thanks
Oliver
mat...@li... wrote on 06.09.2006
21:20:39:
> Sorry all for the error. I just uploaded "working" (at least for me)
> versions to sourceforge again. For those who are sicking of playing
> with mirrors here is a direct download. Don't expect this link to be
> good for a long time though.
>
> http://euclid.uits.iupui.edu/mplfiles/
>
> - Charlie
>
> On 9/6/06, Charlie Moad <cw...@gm...> wrote:
> > The source error must of propagated to those builds. I will post new
> > ones shortly.
> >
> > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> > > Well the thread on the devel list that I referred to explicitly has
the
> > > win32 exe in its title ("Missing __init__.py in
> > > matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there
pretty
> > > much says it all.
> > >
> > > There's also a recent post on the numpy list that sounds like it's
maybe
> > > the same problem (quote):
> > > """
> > > Is there a compatible matplotlib as well? I was o.k. with mpl from
> > > enthought until I switched numerix to numpy. That made mpl unhappy.
> > > I downloaded 0.87.5 but I broke something in the process because now
> > > even switching back to Numeric doesn't make mpl happy.
> > > """
> > >
> > > Thanks,
> > > Sven
> > >
> > > Charlie Moad schrieb:
> > > > That error was relating to the source release. Can you please post
> > > > your error for the binary?
> > > >
> > > > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> > > >> Charlie Moad schrieb:
> > > >> > Minor rev bump for numpy 1.0b5 compatibility. This release
should
> > > >> > remain compatible with future 1.0 releases of numpy.
> > > >> >
> > > >>
> > > >> I keep running into the ImportError problem described on the
> devel list,
> > > >> with the win32 2.4 binary (exe). Is that still just a case of
waiting
> > > >> for the mirrors to update, or is there a deeper problem? Are there
> > > >> alternative download links? How to tell whether it's the
> "right" binary,
> > > >> since version numbers are the same?
> > > >>
> > > >> Thanks for your help,
> > > >> Sven
> > > >>
> > > >
> > >
> > >
> >
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache
Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Yup, looks like the matplotlib-0.87.5.win32-py2.4.exe binary's been fixed.
Thanks!
Martin
Boyd Waters wrote:
> Try downloading again?
> 
> I have not tried the windows binary, but a source tarball from this 
> morning was strange.
> 
> I pulled a 0.87.5 tarball this morning that had lots of permissions 
> problems, an empty __init__.py, weird line terminators... permissions 
> looked like it had been filtered through a DOS filesystem. That tarball 
> had an MD5 checksum of 6cc874f1b4adf95c7aac5c39a38b5517.
> 
> I tried again this afternoon, I got a cleaner source tarball, MD5 
> af3015ceae28ab385dbe9f30c6d2f83e.
> 
> - boyd
> 

Thanks alot, everything seems to be ok now! (with numpy 1.0b5)
-sven
Charlie Moad schrieb:
> Sorry all for the error. I just uploaded "working" (at least for me)
> versions to sourceforge again. For those who are sicking of playing
> with mirrors here is a direct download. Don't expect this link to be
> good for a long time though.
> 
> http://euclid.uits.iupui.edu/mplfiles/
> 
> - Charlie
> 
> On 9/6/06, Charlie Moad <cw...@gm...> wrote:
>> The source error must of propagated to those builds. I will post new
>> ones shortly.
>>
>> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> > Well the thread on the devel list that I referred to explicitly has the
>> > win32 exe in its title ("Missing __init__.py in
>> > matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
>> > much says it all.
>> >
>> > There's also a recent post on the numpy list that sounds like it's
>> maybe
>> > the same problem (quote):
>> > """
>> > Is there a compatible matplotlib as well? I was o.k. with mpl from
>> > enthought until I switched numerix to numpy. That made mpl unhappy.
>> > I downloaded 0.87.5 but I broke something in the process because now
>> > even switching back to Numeric doesn't make mpl happy.
>> > """
>> >
>> > Thanks,
>> > Sven
>> >
>> > Charlie Moad schrieb:
>> > > That error was relating to the source release. Can you please post
>> > > your error for the binary?
>> > >
>> > > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> > >> Charlie Moad schrieb:
>> > >> > Minor rev bump for numpy 1.0b5 compatibility. This release should
>> > >> > remain compatible with future 1.0 releases of numpy.
>> > >> >
>> > >>
>> > >> I keep running into the ImportError problem described on the
>> devel list,
>> > >> with the win32 2.4 binary (exe). Is that still just a case of
>> waiting
>> > >> for the mirrors to update, or is there a deeper problem? Are there
>> > >> alternative download links? How to tell whether it's the "right"
>> binary,
>> > >> since version numbers are the same?
>> > >>
>> > >> Thanks for your help,
>> > >> Sven
>> > >>
>> > >
>> >
>> >
>>
> 

Sorry all for the error. I just uploaded "working" (at least for me)
versions to sourceforge again. For those who are sicking of playing
with mirrors here is a direct download. Don't expect this link to be
good for a long time though.
http://euclid.uits.iupui.edu/mplfiles/
- Charlie
On 9/6/06, Charlie Moad <cw...@gm...> wrote:
> The source error must of propagated to those builds. I will post new
> ones shortly.
>
> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> > Well the thread on the devel list that I referred to explicitly has the
> > win32 exe in its title ("Missing __init__.py in
> > matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
> > much says it all.
> >
> > There's also a recent post on the numpy list that sounds like it's maybe
> > the same problem (quote):
> > """
> > Is there a compatible matplotlib as well? I was o.k. with mpl from
> > enthought until I switched numerix to numpy. That made mpl unhappy.
> > I downloaded 0.87.5 but I broke something in the process because now
> > even switching back to Numeric doesn't make mpl happy.
> > """
> >
> > Thanks,
> > Sven
> >
> > Charlie Moad schrieb:
> > > That error was relating to the source release. Can you please post
> > > your error for the binary?
> > >
> > > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> > >> Charlie Moad schrieb:
> > >> > Minor rev bump for numpy 1.0b5 compatibility. This release should
> > >> > remain compatible with future 1.0 releases of numpy.
> > >> >
> > >>
> > >> I keep running into the ImportError problem described on the devel list,
> > >> with the win32 2.4 binary (exe). Is that still just a case of waiting
> > >> for the mirrors to update, or is there a deeper problem? Are there
> > >> alternative download links? How to tell whether it's the "right" binary,
> > >> since version numbers are the same?
> > >>
> > >> Thanks for your help,
> > >> Sven
> > >>
> > >
> >
> >
>

The source error must of propagated to those builds. I will post new
ones shortly.
On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> Well the thread on the devel list that I referred to explicitly has the
> win32 exe in its title ("Missing __init__.py in
> matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
> much says it all.
>
> There's also a recent post on the numpy list that sounds like it's maybe
> the same problem (quote):
> """
> Is there a compatible matplotlib as well? I was o.k. with mpl from
> enthought until I switched numerix to numpy. That made mpl unhappy.
> I downloaded 0.87.5 but I broke something in the process because now
> even switching back to Numeric doesn't make mpl happy.
> """
>
> Thanks,
> Sven
>
> Charlie Moad schrieb:
> > That error was relating to the source release. Can you please post
> > your error for the binary?
> >
> > On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> >> Charlie Moad schrieb:
> >> > Minor rev bump for numpy 1.0b5 compatibility. This release should
> >> > remain compatible with future 1.0 releases of numpy.
> >> >
> >>
> >> I keep running into the ImportError problem described on the devel list,
> >> with the win32 2.4 binary (exe). Is that still just a case of waiting
> >> for the mirrors to update, or is there a deeper problem? Are there
> >> alternative download links? How to tell whether it's the "right" binary,
> >> since version numbers are the same?
> >>
> >> Thanks for your help,
> >> Sven
> >>
> >
>
>

Well the thread on the devel list that I referred to explicitly has the
win32 exe in its title ("Missing __init__.py in
matplotlib-0.87.5.win32-py2.4.exe ?") . The starting post there pretty
much says it all.
There's also a recent post on the numpy list that sounds like it's maybe
the same problem (quote):
"""
Is there a compatible matplotlib as well? I was o.k. with mpl from
enthought until I switched numerix to numpy. That made mpl unhappy.
I downloaded 0.87.5 but I broke something in the process because now
even switching back to Numeric doesn't make mpl happy.
"""
Thanks,
Sven
Charlie Moad schrieb:
> That error was relating to the source release. Can you please post
> your error for the binary?
> 
> On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
>> Charlie Moad schrieb:
>> > Minor rev bump for numpy 1.0b5 compatibility. This release should
>> > remain compatible with future 1.0 releases of numpy.
>> >
>>
>> I keep running into the ImportError problem described on the devel list,
>> with the win32 2.4 binary (exe). Is that still just a case of waiting
>> for the mirrors to update, or is there a deeper problem? Are there
>> alternative download links? How to tell whether it's the "right" binary,
>> since version numbers are the same?
>>
>> Thanks for your help,
>> Sven
>>
> 

That error was relating to the source release. Can you please post
your error for the binary?
On 9/6/06, Sven Schreiber <sve...@gm...> wrote:
> Charlie Moad schrieb:
> > Minor rev bump for numpy 1.0b5 compatibility. This release should
> > remain compatible with future 1.0 releases of numpy.
> >
>
> I keep running into the ImportError problem described on the devel list,
> with the win32 2.4 binary (exe). Is that still just a case of waiting
> for the mirrors to update, or is there a deeper problem? Are there
> alternative download links? How to tell whether it's the "right" binary,
> since version numbers are the same?
>
> Thanks for your help,
> Sven
>

Charlie Moad schrieb:
> Minor rev bump for numpy 1.0b5 compatibility. This release should
> remain compatible with future 1.0 releases of numpy.
> 
I keep running into the ImportError problem described on the devel list,
with the win32 2.4 binary (exe). Is that still just a case of waiting
for the mirrors to update, or is there a deeper problem? Are there
alternative download links? How to tell whether it's the "right" binary,
since version numbers are the same?
Thanks for your help,
Sven

On 9/5/06, Boyd Waters <bw...@nr...> wrote:
> Try downloading again?
>
> I have not tried the windows binary, but a source tarball from this
> morning was strange.
>
> I pulled a 0.87.5 tarball this morning that had lots of permissions
> problems, an empty __init__.py, weird line terminators... permissions
> looked like it had been filtered through a DOS filesystem. That
> tarball had an MD5 checksum of 6cc874f1b4adf95c7aac5c39a38b5517.
>
> I tried again this afternoon, I got a cleaner source tarball, MD5
> af3015ceae28ab385dbe9f30c6d2f83e.
You caught me : ) . The first source had windows line feeds and linux
choked on it. I removed it and put up a fresh copy after a few hours.
 Some of the mirrors may of taken a while to update. If you have
problems please try downloading the source again.
- Charlie

Try downloading again?
I have not tried the windows binary, but a source tarball from this 
morning was strange.
I pulled a 0.87.5 tarball this morning that had lots of permissions 
problems, an empty __init__.py, weird line terminators... permissions 
looked like it had been filtered through a DOS filesystem. That 
tarball had an MD5 checksum of 6cc874f1b4adf95c7aac5c39a38b5517.
I tried again this afternoon, I got a cleaner source tarball, MD5 
af3015ceae28ab385dbe9f30c6d2f83e.
- boyd
Boyd Waters
Scientific Programmer
National Radio Astronomy Observatory
On Sep 5, 2006, at 8:13 PM, Martin Spacek wrote:
> Not sure if I'm crazy, but it seems the __init__.py file is missing 
> from
> my newly installed version 0.87.5. I installed from the
> matplotlib-0.87.5.win32-py2.4.exe binary. I now get:
>
>>>> import matplotlib
> Traceback (most recent call last):
> File "<stdin>", line 1, in ?
> ImportError: No module named matplotlib
>
> even though the folder's there in site-packages. Examining the binary
> shows that indeed, the file is missing in the PLATLIB/matplotlib 
> folder,
> yet it's there in the lib/matplotlib folder in the
> matplotlib-0.87.5.tar.gz source file.
>
> I now also have a new folder 'matplotlib-0.87.5-py2.4.egg-info' in my
> site-packages, as well as a file 'matplotlib-0.87.5-py2.4- 
> nspkg.pth' in
> the same folder.
>
> Can this be fixed?
>
> Martin
>
> ---------------------------------------------------------------------- 
> ---
> Using Tomcat but need to do more? Need to support web services, 
> security?
> Get stuff done quickly with pre-integrated technology to make your 
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache 
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

Not sure if I'm crazy, but it seems the __init__.py file is missing from 
my newly installed version 0.87.5. I installed from the 
matplotlib-0.87.5.win32-py2.4.exe binary. I now get:
 >>> import matplotlib
Traceback (most recent call last):
 File "<stdin>", line 1, in ?
ImportError: No module named matplotlib
even though the folder's there in site-packages. Examining the binary 
shows that indeed, the file is missing in the PLATLIB/matplotlib folder, 
yet it's there in the lib/matplotlib folder in the 
matplotlib-0.87.5.tar.gz source file.
I now also have a new folder 'matplotlib-0.87.5-py2.4.egg-info' in my 
site-packages, as well as a file 'matplotlib-0.87.5-py2.4-nspkg.pth' in 
the same folder.
Can this be fixed?
Martin

Minor rev bump for numpy 1.0b5 compatibility. This release should
remain compatible with future 1.0 releases of numpy.
http://cheeseshop.python.org/pypi/matplotlib/
http://sourceforge.net/project/showfiles.php?group_id=80706&package_id=82474
===============================================================
2006年09月05日 Released 0.87.5 at revision 2761
2006年09月04日 Added nxutils for some numeric add-on extension code --
		 specifically a better/more efficient inside polygon tester (see
		 unit/inside_poly_*.py) - JDH
2006年09月04日 Made bitstream fonts the rc default - JDH
2006年08月31日 Fixed alpha-handling bug in ColorConverter, affecting
 collections in general and contour/contourf in
 particular. - EF
2006年08月30日 ft2font.cpp: Added draw_rect_filled method (now used by mathtext2
 to draw the fraction bar) to FT2Font - ES
2006年08月29日 setupext.py: wrap calls to tk.getvar() with str(). On some
 systems, getvar returns a Tcl_Obj instead of a string - DSD
2006年08月28日 mathtext2.py: Sub/superscripts can now be complex (i.e.
 fractions etc.). The demo is also updated - ES
2006年08月28日 font_manager.py: Added /usr/local/share/fonts to list of
 X11 font directories - DSD
2006年08月28日 mahtext2.py: Initial support for complex fractions. Also,
 rendering is now completely separated from parsing. The
 sub/superscripts now work better.
 Updated the mathtext2_demo.py - ES
2006年08月27日 qt backends: don't create a QApplication when backend is
 imported, do it when the FigureCanvasQt is created. Simplifies
 applications where mpl is embedded in qt. Updated
 embedding_in_qt* examples - DSD
2006年08月27日 mahtext2.py: Now the fonts are searched in the OS font dir and
 in the mpl-data dir. Also env is not a dict anymore. - ES
2006年08月26日 minor changes to __init__.py, mathtex2_demo.py. Added matplotlibrc
 key "mathtext.mathtext2" (removed the key "mathtext2") - ES
2006年08月21日 mathtext2.py: Initial support for fractions
 Updated the mathtext2_demo.py
 _mathtext_data.py: removed "\" from the unicode dicts
 mathtext.py: Minor modification (because of _mathtext_data.py)- ES
2006年08月20日 Added mathtext2.py: Replacement for mathtext.py. Supports _ ^,
 \rm, \cal etc., \sin, \cos etc., unicode, recursive nestings,
 inline math mode. The only backend currently supported is Agg
 __init__.py: added new rc params for mathtext2
 added mathtext2_demo.py example - ES
2006年08月19日 Added embedding_in_qt4.py example - DSD
2006年08月11日 Added scale free Ellipse patch for Agg - CM
2006年08月10日 Added converters to and from julian dates to matplotlib.dates
 (num2julian and julian2num) - JDH
2006年08月08日 Fixed widget locking so multiple widgets could share the
 event handling - JDH
2006年08月07日 Added scale free Ellipse patch to SVG and PS - CM
2006年08月05日 Re-organized imports in numerix for numpy 1.0b2 -- TEO
2006年08月04日 Added draw_markers to PDF backend. - JKS
2006年08月01日 Fixed a bug in postscript's rendering of dashed lines - DSD
2006年08月01日 figure.py: savefig() update docstring to add support for 'format'
 argument.
 backend_cairo.py: print_figure() add support 'format' argument. - SC
2006年07月31日 Don't let postscript's xpdf distiller compress images - DSD
2006年07月31日 Added shallowcopy() methods to all Transformations;
 removed copy_bbox_transform and copy_bbox_transform_shallow
 from transforms.py;
 added offset_copy() function to transforms.py to
 facilitate positioning artists with offsets.
 See examples/transoffset.py. - EF
2006年07月31日 Don't let postscript's xpdf distiller compress images - DSD
2006年07月29日 Fixed numerix polygon bug reported by Nick Fotopoulos.
 Added inverse_numerix_xy() transform method.
 Made autoscale_view() preserve axis direction
 (e.g., increasing down).- EF
2006年07月28日 Added shallow bbox copy routine for transforms -- mainly
 useful for copying transforms to apply offset to. - JDH
2006年07月28日 Added resize method to FigureManager class
 for Qt and Gtk backend - CM
2006年07月28日 Added subplots_adjust button to Qt backend - CM
2006年07月26日 Use numerix more in collections.
 Quiver now handles masked arrays. - EF
2006年07月22日 Fixed bug #1209354 - DSD
2006年07月22日 make scatter() work with the kwarg "color". Closes bug
 1285750 - DSD
2006年07月20日 backend_cairo.py: require pycairo 1.2.0.
 print_figure() update to output SVG using cairo.
2006年07月19日 Added blitting for Qt4Agg - CM
2006年07月19日 Added lasso widget and example examples/lasso_demo.py - JDH
2006年07月18日 Added blitting for QtAgg backend - CM
2006年07月17日 Fixed bug #1523585: skip nans in semilog plots - DSD
2006年07月12日 Add support to render the scientific notation label
 over the right-side y-axis - DSD

>>>>> "Charlie" == Charlie Moad <cw...@gm...> writes:
 Charlie> I also get a compile error when trying to build against
 Charlie> the win32-py2.4 release. src\_na_nxutils.c(213) : error
 Charlie> C2275: 'PyObject' : illegal use of this type as an
 Charlie> expression c:\Python24\include\object.h(104) : see
It looks like I made the typical C++ programmer writing C mistake:
putting a type declaration in code where it is first used rather than
at the beginning of the function. I just committed a change that
should fix this.
JDH

Thanks for the patch. I am going to hold off until we hear from
Travis. I don't want to distribute anything built from a patched
numpy.
- Charlie
On 9/5/06, Boyd Waters <bw...@nr...> wrote:
> 1.0b5 compiles, with this patch:
>
>
>
>
> does that look right?
>
> - boyd
>
> Boyd Waters
> Scientific Programmer
> National Radio Astronomy Observatory
> http://www.aoc.nrao.edu
>
>
> On Sep 5, 2006, at 12:05 AM, Boyd Waters wrote:
>
> > Very sorry: here is the initial error:
> > numpy/core/src/arrayobject.c:564: error: 'op' undeclared (first use
> > in this function)
> >
> >
> > On Sep 5, 2006, at 12:01 AM, Boyd Waters wrote:
> >
> >> 1.0b5
> >> Fails to compile?
> >
> >
>
>
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>
>
>
>

I also get a compile error when trying to build against the win32-py2.4 release.
src\_na_nxutils.c(213) : error C2275: 'PyObject' : illegal use of this
type as an expression
c:\Python24\include\object.h(104) : see declaration of 'PyObject'
src\_na_nxutils.c(213) : error C2065: 'ret' : undeclared identifier
src\_na_nxutils.c(215) : warning C4047: 'return' : 'PyObject *'
differs in levels of indirection from 'int'
Do we need to modify our use of the c-api at all?
- Charlie
On 9/5/06, Boyd Waters <bw...@nr...> wrote:
> Very sorry: here is the initial error:
> numpy/core/src/arrayobject.c:564: error: 'op' undeclared (first use
> in this function)
>
>
> On Sep 5, 2006, at 12:01 AM, Boyd Waters wrote:
>
> > 1.0b5
> > Fails to compile?
>
>
>
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel
>
>
>
>

1.0b5 compiles, with this patch:

Very sorry: here is the initial error:
numpy/core/src/arrayobject.c:564: error: 'op' undeclared (first use 
in this function)
On Sep 5, 2006, at 12:01 AM, Boyd Waters wrote:
> 1.0b5
> Fails to compile?

1.0b5
Fails to compile?
numpy/core/src/arrayobject.c:2694: error: request for member 
'ob_type' in something not a structure or union
Mac OS X 10.4.7
i686-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc. build 
5363)
python 2.5c1
On Sep 4, 2006, at 7:00 PM, Charlie Moad wrote:
>
> ---------- Forwarded message ----------
> From: Travis Oliphant <oli...@ie...>
> Date: Sep 4, 2006 7:55 PM
> Subject: [Numpy-discussion] Going to cut NumPy 1.0b5 tonight
> To: Discussion of Numerical Python <numpy- 
> dis...@li...>
>
>
> This is a last reminder that I'm going to cut a release of NumPy 1.0b5
> tonight. Please have any fixes and/or problems worked out before 
> then.
>
> -Travis
>
>
>
> ---------------------------------------------------------------------- 
> ---
> Using Tomcat but need to do more? Need to support web services, 
> security?
> Get stuff done quickly with pre-integrated technology to make your 
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache 
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Numpy-discussion mailing list
> Num...@li...
> https://lists.sourceforge.net/lists/listinfo/numpy-discussion
>
> ---------------------------------------------------------------------- 
> ---
> Using Tomcat but need to do more? Need to support web services, 
> security?
> Get stuff done quickly with pre-integrated technology to make your 
> job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache 
> Geronimo
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-devel mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

With matplotlib 0.87.5 shortly after.....
---------- Forwarded message ----------
From: Travis Oliphant <oli...@ie...>
Date: Sep 4, 2006 7:55 PM
Subject: [Numpy-discussion] Going to cut NumPy 1.0b5 tonight
To: Discussion of Numerical Python <num...@li...>
This is a last reminder that I'm going to cut a release of NumPy 1.0b5
tonight. Please have any fixes and/or problems worked out before then.
-Travis
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Numpy-discussion mailing list
Num...@li...
https://lists.sourceforge.net/lists/listinfo/numpy-discussion

Hi all,
I implemented the following method in ft2font.cpp, and although it
works I don't know if it does the *right thing*.
char FT2Font::draw_rect_filled__doc__[] =
"draw_rect_filled(x0, y0, x1, y1)\n"
"\n"
"Draw a filled rect to the image. It is your responsibility to set the\n"
"dimensions of the image, eg, with set_bitmap_size\n"
"\n"
;
Py::Object
FT2Font::draw_rect_filled(const Py::Tuple & args) {
 _VERBOSE("FT2Font::draw_rect_filled");
 args.verify_length(4);
 long x0 = Py::Int(args[0]);
 long y0 = Py::Int(args[1]);
 long x1 = Py::Int(args[2]);
 long y1 = Py::Int(args[3]);
 FT_Int iwidth = (FT_Int)image.width;
 FT_Int iheight = (FT_Int)image.height;
 if ( x0<0 || y0<0 || x1<0 || y1<0 ||
 x0>iwidth || x1>iwidth ||
 y0>iheight || y1>iheight )
 throw Py::ValueError("Rect coords outside image bounds");
 for (long j=y0; j<y1; ++j) {
 for (long i=x0; i<x1+1; ++i) {
 image.buffer[i + j*iwidth] = 255;
 }
 }
 return Py::Object();
}
Basically, I copied the existing draw_rect method and changed the code
a bit. The above code draws a filled rectangle in the image buffer. I
use it mathtext for drawing a line (for a fraction etc.). What I'm
interested is the for loop:
for (long j=y0; j<y1; ++j) {
 for (long i=x0; i<x1+1; ++i) {
 image.buffer[i + j*iwidth] = 255;
 }
 }
I'm not sure that this is the right code - probably some pixel isn't
getting drawn. Anyone has some ideas?
Thanks,
Edin