possible bug with scatter

Hi,

I'm trying to use the scatter method, making use of the option to specify
the marker as a tuple. From the documentation, it would seem that specifying
marker=(0,3,0) should draw a circle. However, this is not the case. If you
consider the following code:

import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as mpl

fig = mpl.figure()
ax = fig.add_subplot(111)
#ax.scatter([x],[y],c='red',marker=(0,3,0.))
ax.scatter([10.],[10.],c='red',marker=(3,3,0.))
ax.scatter([11.],[10.],c='red',marker=(6,3,0.))
ax.set_xlim(5.,15.)
fig.savefig('scatter.png')

The first ax.scatter causes an error, the second plots a triangle, and the
third a hexagon. However, the documentation states that (a) setting the
second element to '3' should plot a circle, and (b) the other arguments
should be ignored, so the first ax.scatter should not cause an error.

Is this a bug, or am I misunderstanding the documentation?

Thanks,

Thomas

···

--
View this message in context: http://www.nabble.com/possible-bug-with-scatter-tp24136534p24136534.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Thanks for the report.
And, this turned out to be a bug. The symbol style code was simply
ignored when its value is 3.

While the bug should now be fixed (both in the trunk and the maint.
branch), you may use marker style like (20,0,0) (or increase the first
number when symbol is large) for a workaround.

Regards,

-JJ

···

On Sun, Jun 21, 2009 at 1:09 PM, Thomas Robitaille<thomas.robitaille@...287...> wrote:

Hi,

I'm trying to use the scatter method, making use of the option to specify
the marker as a tuple. From the documentation, it would seem that specifying
marker=(0,3,0) should draw a circle. However, this is not the case. If you
consider the following code:

import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as mpl

fig = mpl.figure()
ax = fig.add_subplot(111)
#ax.scatter([x],[y],c='red',marker=(0,3,0.))
ax.scatter([10.],[10.],c='red',marker=(3,3,0.))
ax.scatter([11.],[10.],c='red',marker=(6,3,0.))
ax.set_xlim(5.,15.)
fig.savefig('scatter.png')

The first ax.scatter causes an error, the second plots a triangle, and the
third a hexagon. However, the documentation states that (a) setting the
second element to '3' should plot a circle, and (b) the other arguments
should be ignored, so the first ax.scatter should not cause an error.

Is this a bug, or am I misunderstanding the documentation?

Thanks,

Thomas
--
View this message in context: http://www.nabble.com/possible-bug-with-scatter-tp24136534p24136534.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Are you an open source citizen? Join us for the Open Source Bridge conference!
Portland, OR, June 17-19. Two days of sessions, one day of unconference: $250.
Need another reason to go? 24-hour hacker lounge. Register today!
http://ad.doubleclick.net/clk;215844324;13503038;v?http://opensourcebridge.org
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hi Jae-Joon,

Thanks for the quick fix! Just looked in the svn browser, and noticed you changed line 5290 of axes.py to

'o' : (0,0,3),

Should this not be

'o' : (0,3,0),

?

Thanks,

Tom

···

On Jun 21, 2009, at 2:59 PM, Jae-Joon Lee wrote:

Thanks for the report.
And, this turned out to be a bug. The symbol style code was simply
ignored when its value is 3.

While the bug should now be fixed (both in the trunk and the maint.
branch), you may use marker style like (20,0,0) (or increase the first
number when symbol is large) for a workaround.

Regards,

-JJ

On Sun, Jun 21, 2009 at 1:09 PM, Thomas > Robitaille<thomas.robitaille@...287...> wrote:

Hi,

I'm trying to use the scatter method, making use of the option to specify
the marker as a tuple. From the documentation, it would seem that specifying
marker=(0,3,0) should draw a circle. However, this is not the case. If you
consider the following code:

import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as mpl

fig = mpl.figure()
ax = fig.add_subplot(111)
#ax.scatter([x],[y],c='red',marker=(0,3,0.))
ax.scatter([10.],[10.],c='red',marker=(3,3,0.))
ax.scatter([11.],[10.],c='red',marker=(6,3,0.))
ax.set_xlim(5.,15.)
fig.savefig('scatter.png')

The first ax.scatter causes an error, the second plots a triangle, and the
third a hexagon. However, the documentation states that (a) setting the
second element to '3' should plot a circle, and (b) the other arguments
should be ignored, so the first ax.scatter should not cause an error.

Is this a bug, or am I misunderstanding the documentation?

Thanks,

Thomas
--
View this message in context: http://www.nabble.com/possible-bug-with-scatter-tp24136534p24136534.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Are you an open source citizen? Join us for the Open Source Bridge conference!
Portland, OR, June 17-19. Two days of sessions, one day of unconference: $250.
Need another reason to go? 24-hour hacker lounge. Register today!
http://ad.doubleclick.net/clk;215844324;13503038;v?http://opensourcebridge.org
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hi Jae-Joon,

Thanks for the quick fix! Just looked in the svn browser, and noticed you
changed line 5290 of axes.py to

'o' : (0,0,3),

Should this not be

'o' : (0,3,0),

?

No, somehow the meaning of each item is different there. The last item
is the symbol style, unlike the input parameter for scatter where the
second one is the symbol style. For example,

          's' : (4,math.pi/4.0,0), # square

Regards,

-JJ

···

On Sun, Jun 21, 2009 at 4:42 PM, Thomas Robitaille<thomas.robitaille@...287...> wrote:

Thanks,

Tom

On Jun 21, 2009, at 2:59 PM, Jae-Joon Lee wrote:

Thanks for the report.
And, this turned out to be a bug. The symbol style code was simply
ignored when its value is 3.

While the bug should now be fixed (both in the trunk and the maint.
branch), you may use marker style like (20,0,0) (or increase the first
number when symbol is large) for a workaround.

Regards,

-JJ

On Sun, Jun 21, 2009 at 1:09 PM, Thomas >> Robitaille<thomas.robitaille@...287...> wrote:

Hi,

I'm trying to use the scatter method, making use of the option to specify
the marker as a tuple. From the documentation, it would seem that
specifying
marker=(0,3,0) should draw a circle. However, this is not the case. If
you
consider the following code:

import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as mpl

fig = mpl.figure()
ax = fig.add_subplot(111)
#ax.scatter([x],[y],c='red',marker=(0,3,0.))
ax.scatter([10.],[10.],c='red',marker=(3,3,0.))
ax.scatter([11.],[10.],c='red',marker=(6,3,0.))
ax.set_xlim(5.,15.)
fig.savefig('scatter.png')

The first ax.scatter causes an error, the second plots a triangle, and
the
third a hexagon. However, the documentation states that (a) setting the
second element to '3' should plot a circle, and (b) the other arguments
should be ignored, so the first ax.scatter should not cause an error.

Is this a bug, or am I misunderstanding the documentation?

Thanks,

Thomas
--
View this message in context:
http://www.nabble.com/possible-bug-with-scatter-tp24136534p24136534.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Are you an open source citizen? Join us for the Open Source Bridge
conference!
Portland, OR, June 17-19. Two days of sessions, one day of unconference:
$250.
Need another reason to go? 24-hour hacker lounge. Register today!

http://ad.doubleclick.net/clk;215844324;13503038;v?http://opensourcebridge.org
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users