pre rfc draft on rendering

[Steve Lime]

I figured that ultimately with some of the vector output formats we might take
a look at Cairo as an alternative. It will be much easier to do so with the architecture
Thomas is proposing in place.

Steve

>>> On 11/14/2007 at 5:45 AM, in message
<d2b988930711140345l3d394f98l7f2b6b69841560ab at mail.gmail.com>, thomas bonfort
<thomas.bonfort at GMAIL.COM> wrote:
> On Nov 13, 2007 8:05 PM, Yewondwossen Assefa <yassefa at dmsolutions.ca> wrote:
>>   simple case is for example swf and pdf drivers use the GD driver to
>> draw the rasters into a temporary image and integrate them in their
>> "document". Would that be possible in this architecture? Mainly I think
>> would the different renderers be able to redefine "high level" functions
>> such ad DrawRasterLayer ?
> 
> What I've essentially been looking into is the lower level rendering
> of vector primitives. I don't think there will be a need to
> significantly modify the functions that are above the msDrawXXXSymbol
> functions in mapdraw.c , so my impression here is that the current
> behaviour for these renderers would go unchanged.
> 
> As for raster layers, I still can't come up with an elegant way to
> treat them elegantly *and* efficiently.
> 
>> I have started taking notes on how the different interface functions
>> proposed would fit with the swf, pdf and svg and I am willing to put
>> some time toward this if we generally agree to go forward with the
>> proposition.
> 
> that didn't fall on deaf ears ;) given the task at hand and my limited
> knowledge on the svg/pdf/swf side, I was going for unifying the agg
> and gd functions as a first step. If we can get the others done at the
> same time it would be really much better, so I'm really looking
> forward to working this out with you if we go onto this.
> 
> thomas