matplotlib-users Mailing List for matplotlib

On 7/17/07, Chris Fonnesbeck <lis...@ma...> wrote:
> For some reason, builds from SVN dont install either pytz
> or dateutil (at least not in the right place). Importing pylab
> from these builds results in an import error.
>
> How can I build these so as to convince these modules to
> install correctly?
This is typically caused when the install process detects that pytz
and dateutil are already installed, and so doesn't overright them. My
guess is that they were available in your PYTHONPATH at install time
but not at run time. So at install time they are detected and not
installed, but at run time they cannot be found. If this is the
solution, you need to build and run in the same environment, or blow
away existing copies of pytz and dateutil whereever they are lurking
and then reinstall mpl. Use the __file__ module attr to poke around
and see if you can find them.
JDH

On Tuesday 17 July 2007 12:04:54 pm John Hunter wrote:
> > I was wondering, though, whether there'd be any support for some work
> > which tidied up the near-duplicate code in axes.py. I've been playing
>
> Certainly, but probably not using meta-classes.
>
> > around with an approach using python's metaclass support, which, for
> > example, would replace the definitions of the two near-identical
> > functions set_xscale() and set_yscale() with the one meta-definition:
> >
> > @MC_Traited.construct_traited_variants
> > def set__AXISLETTER_scale(self, value, base_AXISLETTER_ = 10,
> > subs_AXISLETTER_ = None): """
> > SET_%(axis_letter_UC)sSCALE(value, base%(axis_letter)s = 10,
> > subs%(axis_letter)s = None)
>
> I'm disinclined to use python black magic -- I find it makes the code
> harder to grok for the typical scientist, and these are our main
> developers. Most of these guys are still trying to figure out what a
> class is <wink>
It's true. That bit of code looks like Perl to me.
Darren

On 7/17/07, Ben North <be...@re...> wrote:
> There is a difference in the behaviour of Axes.hlines() vs
> Axes.vlines(), in that vlines() lets you supply scalars for ymin and
> ymax, whereas hlines() doesn't (for xmin and xmax). The patch below
> fixes that, and also what looks like a separate bug in vlines. There
> are also other differences, around units, but I haven't worked with that
> area of matplotlib so will leave that for others.
Thanks for catching that -- I recently introduced this bug when I
numpified the axes module. Fixed in svn
> I was wondering, though, whether there'd be any support for some work
> which tidied up the near-duplicate code in axes.py. I've been playing
Certainly, but probably not using meta-classes.
> around with an approach using python's metaclass support, which, for
> example, would replace the definitions of the two near-identical
> functions set_xscale() and set_yscale() with the one meta-definition:
>
> @MC_Traited.construct_traited_variants
> def set__AXISLETTER_scale(self, value, base_AXISLETTER_ = 10, subs_AXISLETTER_ = None):
> """
> SET_%(axis_letter_UC)sSCALE(value, base%(axis_letter)s = 10, subs%(axis_letter)s = None)
I'm disinclined to use python black magic -- I find it makes the code
harder to grok for the typical scientist, and these are our main
developers. Most of these guys are still trying to figure out what a
class is <wink>
But there are other ways to reduce code duplication that are not as clever
def set_xscale(self, value, basex = 10, subsx=None):
 self._set_scale(axis=self.xaxis,
 transfunc=self.transData.get_funcx(),
 limfunc=self.get_xlim,
 value=value,
 base=basex,
 subse=subsx)
def set_yscale(self, value, basex = 10, subsx=None):
 self._set_scale(axis=self.yaxis,
 transfunc=self.transData.get_funcy(),
 limfunc=self.get_ylim,
 value=value,
 base=basey,
 subse=subsy)
def _set_scale(self, axis, transfunc, limfunc, value, basex = 10, subsx=None):
 assert(value.lower() in ('log', 'linear', ))
 if value == 'log':
 axis.set_major_locator(mticker.LogLocator(base))
 axis.set_major_formatter(mticker.LogFormatterMathtext(base))
 axis.set_minor_locator(mticker.LogLocator(base,subsx))
 transfunc.set_type(mtrans.LOG10)
 minx, maxx = limfunc()
 if min(minx, maxx)<=0:
 self.autoscale_view()
 elif value == 'linear':
 axis.set_major_locator(mticker.AutoLocator())
 axis.set_major_formatter(mticker.ScalarFormatter())
 axis.set_minor_locator(mticker.NullLocator())
 axis.set_minor_formatter(mticker.NullFormatter())
 transfunc.set_type( mtrans.IDENTITY )
which I definitely encourage. There may be something more elegant
than this, but we only try to be a little clever. Get too clever, and
Darren and Eric will yell at you.

Hello,
I wrote a message some weeks ago about a problem I have with pylab,
but I think I probably did not explained it very well.
I am witting a small application that use tkFileDialog to prompt user
to select a file.
Then reads it and plot the data.
I want that the user could be able to press a key and read another
file, to plot it again together with the previous one(s). The problem
is that the thee graph is not drawn until I resize the window. Even
if I use the draw() method.
Any help will be appreciated.
Thanks in advance.
Angel.
code:
import sys
import os
import time
import tkFileDialog
import pylab
pylab.hold(1)
path='E:\\Nima'
n=0
def click(event):
 print 'click on: ', event.xdata, event.ydata
def keyp(event):
 global path
 if event.key.lower()=='o':
 fin = tkFileDialog.askopenfilename(initialdir=path)
 path=os.path.split(fin)[0]
 plot2(fin)
def plot2(fin):
 global myplot
 global n
 f=file(fin, 'r')
 lines=f.readlines()
 f.close()
 X=[]
 Y=[]
 for line in lines[1:]:
 line=line.replace(',', '.')
 X.append(float(line.split('\t')[2]))
 Y.append(float(line.split('\t')[5]))
 l,=myplot.plot(X, Y, label=os.path.split(fin)[1])
 myplot.legend()
 if n==0:
 print "showing..."
 n=1
 pylab.show()
 l.set_visible(1)
 myplot.draw()
if __name__=='__main__':
 myplot=pylab.subplot(111)
 myplot.grid()
 kid=pylab.connect('key_press_event', keyp)
 pylab.xlabel('Area [A2/molecule]')
 pylab.ylabel('SP [mN/m]')
 fin = tkFileDialog.askopenfilename(initialdir=path)
 path=os.path.split(fin)[0]
 plot2(fin)

For some reason, builds from SVN dont install either pytz
or dateutil (at least not in the right place). Importing pylab 
from these builds results in an import error.
How can I build these so as to convince these modules to
install correctly?
Thanks.

Hi,
There is a difference in the behaviour of Axes.hlines() vs
Axes.vlines(), in that vlines() lets you supply scalars for ymin and
ymax, whereas hlines() doesn't (for xmin and xmax). The patch below
fixes that, and also what looks like a separate bug in vlines. There
are also other differences, around units, but I haven't worked with that
area of matplotlib so will leave that for others.
I was wondering, though, whether there'd be any support for some work
which tidied up the near-duplicate code in axes.py. I've been playing
around with an approach using python's metaclass support, which, for
example, would replace the definitions of the two near-identical
functions set_xscale() and set_yscale() with the one meta-definition:
 @MC_Traited.construct_traited_variants
 def set__AXISLETTER_scale(self, value, base_AXISLETTER_ = 10, subs_AXISLETTER_ = None):
 """
 SET_%(axis_letter_UC)sSCALE(value, base%(axis_letter)s = 10, subs%(axis_letter)s = None)
 Set the %(axis_letter)s-scaling: 'log' or 'linear'
 If value is 'log', the additional kwargs have the following meaning
 * base%(axis_letter)s: base of the logarithm
 * subs%(axis_letter)s: a sequence of the location of the minor ticks;
 None defaults to autosubs, which depend on the number of
 decades in the plot. Eg for base 10, subs%(axis_letter)s = (1, 2, 5) will
 put minor ticks on 1, 2, 5, 11, 12, 15, 21, ....
 To turn off minor ticking, set subs%(axis_letter)s = []
 ACCEPTS: ['log' | 'linear']
 """
 assert(value.lower() in ('log', 'linear'))
 my_axis = _TRAITS_.my_axis(self)
 if value == 'log':
 my_axis.set_major_locator(LogLocator(base_AXISLETTER_))
 my_axis.set_major_formatter(LogFormatterMathtext(base_AXISLETTER_))
 my_axis.set_minor_locator(LogLocator(base_AXISLETTER_, subs_AXISLETTER_))
 _TRAITS_.get_my_func(self.transData).set_type(LOG10)
 minval, maxval = _TRAITS_.get_my_lim(self)
 if min(minval, maxval) <= 0:
 self.autoscale_view()
 elif value == 'linear':
 my_axis.set_major_locator(AutoLocator())
 my_axis.set_major_formatter(ScalarFormatter())
 my_axis.set_minor_locator(NullLocator())
 my_axis.set_minor_formatter(NullFormatter())
 _TRAITS_.get_my_func(self.transData).set_type(IDENTITY)
I haven't quite worked through the details, but it looks like it would
cut nearly 400 lines off axes.py (while adding c.200 lines of supporting
code), as well as avoid the type of inconsistency seen in hlines/vlines.
If there's interest, I'll post what I've got so far for comments.
Thanks,
Ben.
- - - - 8< - - - -
--- ORIG/axes.py 2007年07月17日 15:24:10.367402000 +0100
+++ NEW/axes.py 2007年07月17日 15:38:03.705394000 +0100
@@ -2325,6 +2325,11 @@
 xmin = asarray(xmin)
 xmax = asarray(xmax)
+ if len(xmin)==1:
+ xmin = xmin*ones(y.shape, typecode(y))
+ if len(ymax)==1:
+ xmax = xmax*ones(y.shape, typecode(y))
+
 if len(xmin)!=len(y):
 raise ValueError, 'xmin and y are unequal sized sequences'
@@ -2418,7 +2423,7 @@
 minx = nx.amin(x)
 maxx = nx.amax(x)
 miny = min(nx.amin(ymin), nx.amin(ymax))
- maxy = max(nx.amax(ymax), nx.amax(ymax))
+ maxy = max(nx.amax(ymin), nx.amax(ymax))
 minx, maxx = self.convert_xunits((minx, maxx))
 miny, maxy = self.convert_yunits((miny, maxy))
 corners = (minx, miny), (maxx, maxy)

On 7/17/07, Ben North <be...@re...> wrote:
> Hi,
>
> The normal rectangular Axes class and the derived PolarAxes class both
> have set_xlim() and set_ylim() functions, but the rectangular Axes class
> has a default value of False for the 'emit' argument, whereas the
> PolarAxes version has True.
> looked at all callers of set_xlim() so I could well be missing a good
> reason why this isn't currently done.)
This inconsistency was never intentional, and is fixed in svn.
JDH

I recently ran into a similar problem myself building stuff from source, 
but I'm not sure of the specifics with SuSE and their packages etc.
Python can be configured in two ways -- with two-byte (UCS2) or 
four-byte (UCS4) Unicode characters. Apparently the default for a 
source installation of Python is UCS2, but many (most) Linux 
distributions build it for UCS4. Python extensions built for one 
configuration can not be used with a Python built for the other 
configuration.
When Python extensions are built, if all goes well, they will match the 
configuration of the Python interpreter. It looks like somehow you have 
a mismatch between matplotlib and your Python interpreter.
If you installed everything from packages, I would expect them all to 
match (unless SuSE's quality control has really gone down as of late 
;). Perhaps something is still around from when you built things from 
source. Did you at any point build your own Python?
On a number of Linux distributions (probably including SuSE, but I don't 
know for sure), things installed from source are under the /usr/local 
tree. To diagnose this, you could see if anything is getting pulled in 
from there (rather than from the packaged stuff, which wouldn't be under 
/usr/local). For instance "whereis python", will tell you which python 
is being used. When you import a Python module, you can use __file__ to 
see where it was imported from. For example:
 >>> import pylab
 >>> pylab.__file__
Hope that at least offers some next steps for tracking this down.
Cheers,
Mike
mark starnes wrote:
> Hi everyone,
>
> I'm running Suse10.2 and installing packages using Yast (after much pain
> trying to install Numpy and Scipy without it!). After installing (and
> re-installing) Matplotlib in this way, I get the error,
>
> ImportError: matplotlib/ft2font.so: undefined symbol: PyUnicodeUCS4_GetSize
>
> when I attempt to import pylab.
>
> Can anybody help me fix this? I couldn't find any help on the
> matplotlib site and my .matplotlib directory is empty.
>
> Oh, I'm also a bit new to Linux - please be patient!
>
> Thanks in advance,
>
> Mark.
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 

Hi,
The normal rectangular Axes class and the derived PolarAxes class both
have set_xlim() and set_ylim() functions, but the rectangular Axes class
has a default value of False for the 'emit' argument, whereas the
PolarAxes version has True. I had a figure containing three Axes
instances, arranged as one 'main' plot and two subsidiary plots. I
wanted the subsidiary plots to have their x- and y-limits slaved to
changes in the main plot's, but with the current default value of
'emit', I wasn't getting callbacks when I used the navigation toolbar to
pan/zoom around the main plot. Would there be any disadvantage to
changing the default value to True, as in the patch below? (I haven't
looked at all callers of set_xlim() so I could well be missing a good
reason why this isn't currently done.)
Thanks,
Ben.
- - - - 8< - - - -
--- ORIG/axes.py 2007年07月17日 15:24:10.367402000 +0100
+++ NEW/axes.py 2007年07月17日 15:48:28.708471000 +0100
@@ -1449,7 +1449,7 @@
 return self.viewLim.intervalx().get_bounds()
- def set_xlim(self, xmin=None, xmax=None, emit=False, **kwargs):
+ def set_xlim(self, xmin=None, xmax=None, emit=True, **kwargs):
 """
 set_xlim(self, *args, **kwargs):
@@ -1573,7 +1573,7 @@
 'Get the y axis range [ymin, ymax]'
 return self.viewLim.intervaly().get_bounds()
- def set_ylim(self, ymin=None, ymax=None, emit=False, **kwargs):
+ def set_ylim(self, ymin=None, ymax=None, emit=True, **kwargs):
 """
 set_ylim(self, *args, **kwargs):

Hi everyone,
I'm running Suse10.2 and installing packages using Yast (after much pain
trying to install Numpy and Scipy without it!). After installing (and
re-installing) Matplotlib in this way, I get the error,
ImportError: matplotlib/ft2font.so: undefined symbol: PyUnicodeUCS4_GetSize
when I attempt to import pylab.
Can anybody help me fix this? I couldn't find any help on the
matplotlib site and my .matplotlib directory is empty.
Oh, I'm also a bit new to Linux - please be patient!
Thanks in advance,
Mark.

Hello,
I wrote a message some weeks ago about a problem I have with pylab,
but I think I probably did not explained it very well.
I am witting a small application that use tkFileDialog to prompt user
to select a file.
Then reads it and plot the data.
I want that the user could be able to press a key and read another
file, to plot it again together with the previous one(s). The problem
is that the thee graph is not drawn until I resize the window. Even
if I use the draw() method.
Any help will be appreciated.
Thanks in advance.
Angel.
code:
import sys
import os
import time
import tkFileDialog
import pylab
pylab.hold(1)
path='E:\\Nima'
n=0
def click(event):
 print 'click on: ', event.xdata, event.ydata
def keyp(event):
 global path
 if event.key.lower()=='o':
 fin = tkFileDialog.askopenfilename(initialdir=path)
 path=os.path.split(fin)[0]
 plot2(fin)
def plot2(fin):
 global myplot
 global n
 f=file(fin, 'r')
 lines=f.readlines()
 f.close()
 X=[]
 Y=[]
 for line in lines[1:]:
 line=line.replace(',', '.')
 X.append(float(line.split('\t')[2]))
 Y.append(float(line.split('\t')[5]))
 l,=myplot.plot(X, Y, label=os.path.split(fin)[1])
 myplot.legend()
 if n==0:
 print "showing..."
 n=1
 pylab.show()
 l.set_visible(1)
 myplot.draw()
if __name__=='__main__':
 myplot=pylab.subplot(111)
 myplot.grid()
 kid=pylab.connect('key_press_event', keyp)
 pylab.xlabel('Area [A2/molecule]')
 pylab.ylabel('SP [mN/m]')
 fin = tkFileDialog.askopenfilename(initialdir=path)
 path=os.path.split(fin)[0]
 plot2(fin)

Thanks for tracking this down, Ben. Applied in svn as r3547.
Ben North wrote:
> I've been using matplotlib for a little while and am finding it very
> useful. Yesterday, though, I hit a problem:

Hi,
I've been using matplotlib for a little while and am finding it very
useful. Yesterday, though, I hit a problem:
Because I didn't read the docs properly, I tried to use
 matplotlib.transforms.scale_transform
like this:
 t = scale_transform(1.0, 2.0)
but then I got a core-dump when trying to use it:
 from matplotlib.transforms import scale_transform
 t = scale_transform(1.0, 2.0)
 print t.xy_tup((1.0, 0.2))
What I should have done was this, which works:
 from matplotlib.transforms import scale_transform, Value
 t = scale_transform(Value(1.0), Value(2.0))
 print t.xy_tup((1.0, 0.2))
The function _transforms_module::new_affine() in _transforms.cpp does
contain checks that its args are LazyValue objects, but it seems to
ignore the results. The following patch (against 0.90.1) makes the
constructor throw a TypeError if it doesn't get what it wants. (It also
fixes a small typo in a separate error string.)
--- ORIG/_transforms.cpp 2007年07月17日 10:10:37.443202000 +0100
+++ NEW/_transforms.cpp 2007年07月17日 10:11:00.257365000 +0100
@@ -42,7 +42,7 @@
 int
 LazyValue::compare(const Py::Object &other) {
 if (!check(other))
- throw Py::TypeError("Can on compare LazyValues with LazyValues");
+ throw Py::TypeError("Can only compare LazyValues with LazyValues");
 LazyValue* pother = static_cast<LazyValue*>(other.ptr());
 double valself = val();
 double valother = pother->val();
@@ -2116,12 +2116,13 @@
 args.verify_length(6);
- LazyValue::check(args[0]);
- LazyValue::check(args[1]);
- LazyValue::check(args[2]);
- LazyValue::check(args[3]);
- LazyValue::check(args[4]);
- LazyValue::check(args[5]);
+ if (!LazyValue::check(args[0])
+ || !LazyValue::check(args[1])
+ || !LazyValue::check(args[2])
+ || !LazyValue::check(args[3])
+ || !LazyValue::check(args[4])
+ || !LazyValue::check(args[5]))
+ throw Py::TypeError("Affine(a, b, c, d, tx, ty) expected 6 LazyValue args");
 LazyValue* a = static_cast<LazyValue*>(args[0].ptr());
 LazyValue* b = static_cast<LazyValue*>(args[1].ptr());
Would this be worth applying? By the look of the code in transform.py,
translation_transform() and possibly others might be affected by this
too. Maybe a better solution might be to automatically construct Value
objects from Python floats where required, but that might need a bit
more thought.
Ben.

hi,
i observed a similar problem that persists in matplotlib 0.90.1 (python 2.5,
numpy 1.0.1, ipython 0.7.3).
the problem occurs when a window is closed and it seems to be specific to
the non-interactive mode using the GTK or GTKAgg backend. the following
short script runs ok once, but when i try to run it a second time, python
hangs:
# start script
import pylab as P
P.ioff()
P.figure()
P.close()
# stop script
this is probably a problem of matplotlib and we should continue this thread
on the matplotlib email list.
cheers,
 kilian
> From: "John Hunter" <jd...@gm...>
> > Date: July 13, 2007 1:10:35 PM PDT
> > To: "SciPy Users List" < sci...@sc...>
> > Subject: Re: [SciPy-user] (Mac) Close a plot window, crash IPython?
> > Reply-To: SciPy Users List < sci...@sc...>
> >
> > On 7/13/07, David Warde-Farley <dav...@ut...> wrote:
> >
> >> It does seem as though the *exact* same bug was reported today on
> >> matplotlib-devel (what are the odds?), I shall checkout the svn
> >> version and see if that fixes it, and make them aware of the other
> >> WX-
> >> related bug as well.
> >
> > I hope this does fix your problem, but the bug that was fixed was also
> > recently introduced (in a svn commit after the 0.90.1 release) so it
> > may not be your problem. But if you can update from svn and see if
> > the problem is still there, that would be a great start. See you on
> > the matplotlib-devel side :-)
> >
> > JDH
> > _______________________________________________
> > SciPy-user mailing list
> > Sci...@sc...
> > http://projects.scipy.org/mailman/listinfo/scipy-user
>
>

Stephen George wrote:
> 
> Hi,
> 
> Can MatPlotLib create a graph similar to the attached GnuPlot graph?
Steve,
That is a 3D plot, and matplotlib is fundamentally 2D. Some 3D plotting 
ability has been added, but it is mostly unmaintained and incomplete.
Eric

Hi,
Can MatPlotLib create a graph similar to the attached GnuPlot graph?
I have used contour before, .. but would like to also get the surface 
projection, .. above the contour
I'd prefer to use just the one plotting library for my work, not keen to 
install yet another plotting library.
Thanks
Steve