Re: Customize SHADOW
Michael Droettboom <
mdroe@...>
2009-10-01 12:37:09 GMT
Yeah, alpha handling is a bit of a mess -- it should probably be
revamped in light of the fact that most places now accept rgba. We just
need to decide if there is a good solution that doesn't break backward
compatibility, or whether we should just break compatibility (e.g.
remove set/get_alpha) moving forward.
As for the current problem, I'll add a set_alpha call to the Shadow class.
Mike
Jae-Joon Lee wrote:
> Mike,
>
> I think this maybe related with some changes in how alpha is set (this
> happened sometime early this year I guess).
>
> I think the issue here is, when the shadow patch is created, it sets
> its facecolor with alpha=0.5., i.e., its _facecolor is something like
> (r, g, b, 0.5). But, shadow._alpha = 1 still. And later when the
> shadow is drawn, the alpha of the facecolor is simply overridden by
> _alpha. Given that alpha=0.5 is intended, I think this is a bug. But
> I'm not sure what is the preferred way to fix this.
>
> I think this is a general issue of Patch classes. While the alpha
> values can be set with facecolor and edgecolor, they are simply
> overridden by _alpha. If this behavior is necessary and intended, we
> should change the Shadow class to set its alpha correctly.
>
> I, personally, want to have different alphas for the facecolor and
> edgecolor, which cannot be done with the current approach. However, I
> believe the current backend api itself (draw_path) does not allow
> different alphas for edgecolor and facecolor, so it may best stick to
> the current behavior.
>
> Regards,
>
> -JJ
>
>
> On Wed, Sep 30, 2009 at 4:51 PM, Michael Droettboom <mdroe@...> wrote:
>
>> I'm still not seeing a difference between 0.98.5 and 0.99.1 here. I
>> further investigation of the code shows that there were no changes in
>> how the shadow color is computed between these versions. Is it possible
>> you're using an even earlier version? You can determine it using:
>>
>> >>> import matplotlib
>> >>> matplotlib.__version__
>>
>> Are there any other differences between the two installations, such as
>> backend?
>>
>> Cheers,
>> Mike
>>
>> Gewton Jhames wrote:
>>
>>> sorry, this is the script:
>>>
>>> #!/usr/bin/python
>>> # -*- coding: utf-8 -*-
>>> from pylab import *
>>>
>>> from matplotlib.pyplot import figure, show
>>> from matplotlib.patches import Ellipse
>>> import numpy as np
>>>
>>> figure(1, figsize=(6,6), facecolor='#ffffff')
>>> ax = axes([0.1, 0.1, 0.8, 0.8])
>>>
>>> labels = 'label1', 'label2'
>>> fracs = [40, 60]
>>> colors = ['#E3AB9C', '#C6E9F8']
>>>
>>> explode=(0, 0.05)
>>>
>>> plots = pie(fracs, explode=explode, labels=labels, colors=colors,
>>> autopct='%0.2f%%', shadow=True)
>>> plots[0][0].set_edgecolor('#E4471A')
>>> plots[0][1].set_edgecolor('#1AA8E4')
>>> title('Raining Hogs and Dogs', bbox={'facecolor':'0.8', 'pad':5})
>>>
>>>
>>> show()
>>>
>>>
>>> On Wed, Sep 30, 2009 at 5:19 PM, Michael Droettboom <mdroe@...
>>> <mailto:mdroe@...>> wrote:
>>>
>>> Can you provide the script that produces these graphs? I don't
>>> see any difference between 0.98.5 and 0.99.1 on the included
>>> pie_demo.py example. Which backend are you using?
>>>
>>> Mike
>>>
>>> Gewton Jhames wrote:
>>>
>>> Hello, I'm having two different results in the shadow of a
>>> graph. I develop the graph in a system with matplotlib 0.98.5.
>>> When I put the code in other machine, with the same version of
>>> libpng, but with matplotlib 0.99.0, the shadow of the graph
>>> has changed to a real bad one.
>>>
>>> In attachment, the two generated graphs. I wish that the
>>> graph's shadown generated in 0.99 be the same of 0.98.5.
>>>
>>> Well, see the attachment and you'll understand.
>>> thanks.
>>>
>>> ------------------------------------------------------------------------
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> ------------------------------------------------------------------------
>>>
>>> ------------------------------------------------------------------------------
>>> Come build with us! The BlackBerry® Developer Conference
>>> in SF, CA
>>> is the only developer event you need to attend this year.
>>> Jumpstart your
>>> developing skills, take BlackBerry mobile applications to
>>> market and stay ahead of the curve. Join us from November
>>> 9-12, 2009. Register now!
>>> http://p.sf.net/sfu/devconf
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Matplotlib-users@...
>>> <mailto:Matplotlib-users@...>
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>
>>>
>>>
>>> --
>>> Michael Droettboom
>>> Science Software Branch
>>> Operations and Engineering Division
>>> Space Telescope Science Institute
>>> Operated by AURA for NASA
>>>
>>>
>>>
>> --
>> Michael Droettboom
>> Science Software Branch
>> Operations and Engineering Division
>> Space Telescope Science Institute
>> Operated by AURA for NASA
>>
>>
>> ------------------------------------------------------------------------------
>> Come build with us! The BlackBerry® Developer Conference in SF, CA
>> is the only developer event you need to attend this year. Jumpstart your
>> developing skills, take BlackBerry mobile applications to market and stay
>> ahead of the curve. Join us from November 9-12, 2009. Register now!
>> http://p.sf.net/sfu/devconf
>> _______________________________________________
>> Matplotlib-users mailing list
>> Matplotlib-users@...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>>
--
--
Michael Droettboom
Science Software Branch
Operations and Engineering Division
Space Telescope Science Institute
Operated by AURA for NASA
------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf
RSS Feed