Jan Strube wrote: > Sorry, reposting, because I sent it from the wrong address. > > I'm having some problems understanding the difference between > pylab.xticks() and pylab.yticks() > Consider the following: > > import pylab as P > import numpy as N > > data = N.random.random((10, 10)) > P.matshow(data) > P.xticks([0, 1, 2], ['1', '2', '3']) > P.show() > > > Why does this work, but if I change P.xticks to P.yticks, it doesn't?
This is now fixed in svn 3904. The problem was not related to x versus y but was caused by incorrect handling of the reversed y axis (positive down instead of positive up). If the x axis were the reversed one, the same problem would occur. > > The error message that I get back doesn't give me any insight as to > what could cause the difference between xticks and yticks. Yes, it is one of those infamous error messages from the extension code in the transforms module. > > File > > "/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/matplotlib/figure.py", > line 612, in draw > for a in self.axes: a.draw(renderer) > File > > "/Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/matplotlib/axes.py", > line 1287, in draw > self.transData.freeze() # eval the lazy objects > ValueError: Domain error on eval_scalars in Transformation::freeze It is probably a divide-by-zero error; but the real bug was elsewhere in the extension code, where the y-data-interval was erroneously set to zero when you tried to set the ticks. > > > matplotlib.__version__ > Out[2]: '0.90.1' > > I am still in the learning phase, so any insight as to what's going > on is appreciated. > > Best, > Jan Thanks for the good bug report, or more precisely, the good report of a bad bug. I dimly recall some other email recently--maybe on matplotlib-devel--that may have been related. I think it was advocating a different method of keeping track of reversed axes. Maybe it was prompted by encountering the same problem you ran into. Eric ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Matplotlib-users mailing list Matplotlib-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-users
