Hi. I just installed the newest version of matplotlib (0.82), and discovered all my subplots were ending on top of each other. After going through my code and verifying everything looked okay, I tried a simple:
subplot(2, 1, 1); plot(range(0, 10)); subplot(2, 2, 2); plot(range(0, 10))
and still only got one subplot. I went to the matplotlib code, and after some prodding, discovered that all the keys for the _seen map were exactly the same. Going to the _make_key function revealed that my use of subplot (i.e. three seperate arguments), was not working. Here is a fix that seems to work for me:
(int _make_key function):
if iterable(args[0]):
key = tuple(args[0]), tuple( fixitems(kwargs.items()))
## NEW
elif len(args) > 1:
key = args, tuple( fixitems(kwargs.items()))
## \NEW
else:
key = args[0], tuple(fixitems( kwargs.items()))
Abe
Abe,
That bug has been fixed in CVS; like you, I tripped over it in 0.82 and tracked it down--but someone else had already fixed it. A workaround, if you don't want to install from CVS, is to use the old subplot(211) form instead of subplot(2,1,1).
Eric
Abraham Schneider wrote:
···
Hi. I just installed the newest version of matplotlib (0.82), and discovered all my subplots were ending on top of each other. After going through my code and verifying everything looked okay, I tried a simple:
subplot(2, 1, 1); plot(range(0, 10)); subplot(2, 2, 2); plot(range(0, 10))
and still only got one subplot. I went to the matplotlib code, and after some prodding, discovered that all the keys for the _seen map were exactly the same. Going to the _make_key function revealed that my use of subplot (i.e. three seperate arguments), was not working. Here is a fix that seems to work for me:
(int _make_key function):
if iterable(args[0]):
key = tuple(args[0]), tuple( fixitems(kwargs.items()))
## NEW
elif len(args) > 1:
key = args, tuple( fixitems(kwargs.items()))
## \NEW
else:
key = args[0], tuple(fixitems( kwargs.items()))
Abe
-------------------------------------------------------
This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening
July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
core and dual graphics technology at this free one hour event hosted by HP,
AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
matplotlib-devel List Signup and Options
That's good to know. I tried googling it, but I guess I should've also checked CVS as well. I'm sure it's better than my hack anyways -- just good to have my scripts working again. I tend to prefer the subplot(w, h, n) since most of my plotting is automated, and it's more of a pain to have to convert the parameters to a string first, etc.
A
Eric Firing wrote:
···
Abe,
That bug has been fixed in CVS; like you, I tripped over it in 0.82 and tracked it down--but someone else had already fixed it. A workaround, if you don't want to install from CVS, is to use the old subplot(211) form instead of subplot(2,1,1).
Eric
Abraham Schneider wrote:
Hi. I just installed the newest version of matplotlib (0.82), and discovered all my subplots were ending on top of each other. After going through my code and verifying everything looked okay, I tried a simple:
subplot(2, 1, 1); plot(range(0, 10)); subplot(2, 2, 2); plot(range(0, 10))
and still only got one subplot. I went to the matplotlib code, and after some prodding, discovered that all the keys for the _seen map were exactly the same. Going to the _make_key function revealed that my use of subplot (i.e. three seperate arguments), was not working. Here is a fix that seems to work for me:
(int _make_key function):
if iterable(args[0]):
key = tuple(args[0]), tuple( fixitems(kwargs.items()))
## NEW
elif len(args) > 1:
key = args, tuple( fixitems(kwargs.items()))
## \NEW
else:
key = args[0], tuple(fixitems( kwargs.items()))
Abe
-------------------------------------------------------
This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening
July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
core and dual graphics technology at this free one hour event hosted by HP,
AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
matplotlib-devel List Signup and Options
-------------------------------------------------------
This SF.Net email is sponsored by the 'Do More With Dual!' webinar happening
July 14 at 8am PDT/11am EDT. We invite you to explore the latest in dual
core and dual graphics technology at this free one hour event hosted by HP,
AMD, and NVIDIA. To register visit http://www.hp.com/go/dualwebinar
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
matplotlib-devel List Signup and Options