---------- Forwarded message ----------
From: Phil Rosenberg <p.d.rosenb...@gmail.com>
Date: 26 January 2015 at 22:14
Subject: Re: Another patch for plbuf
To: Jim Dishaw <j...@dishaw.org>


Hi Jim
I've moved this discussion to the developer list.

Yes I think that suggestion sounds correct. If rather than waiting for
some feature e.g. colour map, to be required, finding that it is
uninitialized and then setting some default vales I think that as you
suggest it would be better if on initialization plplot set these
defaults explicitly by calling the appropriate functions. This way all
defaults will end up in the buffer and we shouldn't get any overspill
from subsequent pages, plus any other initialisation code (array
allocation etc) can be kept together.

Regarding Qt, I think you are probably correct. From memory when the
window is resize the plot remains a fixed size so it almost certainly
just blits the image from a bimap on redraws rather than actually
redrawing the vector graphics from the plplot buffer.

Phil

On 26 January 2015 at 19:31, Jim Dishaw <j...@dishaw.org> wrote:
> I discovered the source of the greyscale on resize.  The colormap was changed 
> when the Lena image was plotted.  When the plot buffer is replayed, the 
> default colormap was not restored.  I have to think the best way to handle 
> this--I'm thinking of having the plplot initialization routine stuff the 
> default configurations (e.g. colormap) into the plot buffer.
>
> Font changes would have a similar problem.
>
>
> On Jan 26, 2015, at 1:13 PM, Jim Dishaw <j...@dishaw.org> wrote:
>
>> Example 20 has some interesting behaviors.  The xwin driver has the same 
>> problem as wxWidgets.  The Qt Widget works, but I have a feeling it is not 
>> using the plplot redraw (I just ran it and have not looked at the code).  
>> The xcairo clips the plot in a weird way when resizing.
>>
>> I will fix this as a separate commit from the plmeta fixing.
>>
>> On Jan 26, 2015, at 5:28 AM, Phil Rosenberg <p.d.rosenb...@gmail.com> wrote:
>>
>>> No worries and no rush. Like I said in one of my on list emails, I
>>> would consider getting the metafile output a more pressing priority to
>>> start with then we can begin systematically hunting down these buffer
>>> problems, which I suspect will be numerous.
>>>
>>> Phil
>>>
>>> On 25 January 2015 at 20:00, Jim Dishaw <j...@dishaw.org> wrote:
>>>> I will track down the bug and revise.
>>>>
>>>> On Jan 25, 2015, at 10:35 AM, Phil Rosenberg <p.d.rosenb...@gmail.com> 
>>>> wrote:
>>>>
>>>>> Hi Jim
>>>>>
>>>>> There are still some problems in the buffer with colour maps. For
>>>>> example if you run example 20 with the wxWidgets driver it will show
>>>>> the first page as a red/purple plot, but resizing it causing it to be
>>>>> redrawn from the buffer results in a greyscale plot.
>>>>>
>>>>> I think it would be best to deal with this all in one chunk commited
>>>>> to master together, mostly because when I don't know what the bug is,
>>>>> I don't know what changes (however innocuous)  might change this from
>>>>> a mis-colouring to a segfault. Also I can't test the patches to make
>>>>> sure things still work, when things don't work at all. So for this
>>>>> reason, for now I haven't applied the patches you sent.
>>>>>
>>>>> Does this sound sensible to you?
>>>>>
>>>>> Phil
>>>>>
>>>>> On 25 January 2015 at 00:11, Jim Dishaw <j...@dishaw.org> wrote:
>>>>>> There was some duplicative code in the PLSTATE_CMAP0 and PLSTATE_CMAP1 
>>>>>> blocks.
>>>>>>
>>>>
>>
>

------------------------------------------------------------------------------
Dive into the World of Parallel Programming. The Go Parallel Website,
sponsored by Intel and developed in partnership with Slashdot Media, is your
hub for all things parallel software development, from weekly thought
leadership blogs to news, videos, case studies, tutorials and more. Take a
look and join the conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel

Reply via email to