matplotlib-users Mailing List for matplotlib

Hello,

I have further investigated problems with legend() when numpoints = 1.
The images show what happens when numpoints = 1 for a Line2D, such as
when calling plot(), for a LineCollection, and for a
RegularPolyCollection, such as when using scatter(). As can be seen in
the figures, calling legend() with numpoints = 1 results in either an
absence of a line or a misplacement of a symbol or colored region,
creating an ugly legend.

I have made the three figures using the script included below,
line_collection.py from the examples, and scatter_demo.py from the examples.

I will send a second message that includes a patch that attempts to fix
the problems with legend() when numpoints=1, and some figures showing
the improvement in the legends.

Paul

---
#!/usr/bin/env python

import matplotlib
matplotlib.use('GTKAgg')

from pylab import *

x = arange(0.0, 5.0)
y = 2.0 * x

figure(1)
plot(x, y, 'o', label='symbol')
plot(x, y, 'k-', label='line')

leg=legend(loc='best',numpoints=1)

show()

Hello,

I have included below a patch that attempts to fix the problems with
legend() when numpoints = 1. When numpoints = 1 and a line is plotted,
_marker == 'None', and a line segment of length handlelen is placed in
the legend; when symbols are being plotted, a single symbol is centered
in the handlelen region and placed in the legend. For LineCollection and
RegularPolyCollection, the legend when numpoints = 1 looks like the
legend when numpoints > 1; the legend always has a line or colored
region and the line or colored region and text do not overlap. The
results of the patch can be seen in the included figures.

There are some problems with the patch.
1) The variable left has to be introduced into the function
_get_handles() to properly place the legend markers.
2) Legends which contain a Patch are not handled, because I do not know
which kind of plot would result in a Patch in the legend, so I could not
test the effect of changing the legend.
3) The patch was created against 0.91.1; however, I do not think that
there is any difference in legend.py between 0.91.1 and SVN.

I would appreciate any comments or improvements on the patch.

Thanks,

Paul


diff -u a/lib/matplotlib/legend.py b/lib/matplotlib/legend.py
--- a/lib/matplotlib/legend.py	2008-01-02 16:33:47.000000000 -0600
+++ b/lib/matplotlib/legend.py	2008-01-03 08:56:01.000000000 -0600
@@ -273,6 +273,7 @@

         def _get_handles(self, handles, texts):
             HEIGHT = self._approx_text_height()
+        left = 0.5

             ret = []   # the returned legend lines

@@ -280,6 +281,10 @@
                 x, y = label.get_position()
                 x -= self.handlelen + self.handletextsep
                 if isinstance(handle, Line2D):
+                if self.numpoints == 1 and handle._marker == 'None':
+                    self._xdata = npy.linspace(left, left +
self.handlelen, 2)
+                elif self.numpoints == 1:
+                    self._xdata = npy.array([left + self.handlelen*0.5])
                     ydata = (y-HEIGHT/2)*npy.ones(self._xdata.shape, float)
                     legline = Line2D(self._xdata, ydata)
                     legline.update_from(handle)
@@ -298,6 +303,8 @@
                     p.set_clip_box(None)
                     ret.append(p)
                 elif isinstance(handle, LineCollection):
+                if self.numpoints == 1:
+                    self._xdata = npy.linspace(left, left +
self.handlelen, 2)
                     ydata = (y-HEIGHT/2)*npy.ones(self._xdata.shape, float)
                     legline = Line2D(self._xdata, ydata)
                     self._set_artist_props(legline)
@@ -311,6 +318,8 @@
                     ret.append(legline)

                 elif isinstance(handle, RegularPolyCollection):
+                if self.numpoints == 1:
+                    self._xdata = npy.array([left])
                     p = Rectangle(xy=(min(self._xdata), y-3/4*HEIGHT),
                                   width = self.handlelen, height=HEIGHT/2,
                                   )

Hi,

I have some two sets of data with their x- and y-coordinates, (x1, y1) and
(x2, y2). I want to plot these curves as a parametric plot:
#########
from numpy import *
from pylab import *

t = arange (0.0 , pi/2 , 0.05)
x1= cos(t)
y1 = -sin(t)
x2 = -1+2/pi*t
y2 = -2/pi*t

# plotted as a normal plot
plot( t , x1 , t , y1 ,t , x2,  t , y2)
legend(('x1','y1','x2','y2'))
show()
#########

I have read that this is done with the countour function, but I cannot
figure out how. Can somebody help me.

Also I have tried to search for information in this forum using the search
tool at the top of the page, but I could not get any hit even for obvious
queries such as "output". Is there something wrong?

Thank you in advance
Aitor