Re: [Pnet-developers] GraphicsPath

[Rich Baumann]

On Mon, 2005-01-24 at 23:43 +0100, mboss wrote:
> At the first glance, the strategy of PathObject helper classes looks 
> appealing. When you come to describe the "points" making up a graphic path, 
> things become difficult. 
> 
> How does Microsoft handle it? every shape (yes, even strings) is mapped 
> internally to a set of points with corresponding point type ( a PathData 
> structure ).
> Now of course you will ask me what the hell are you talking about MS, this is 
> dotGNU here!

Actually we're working towards being as compatible as possible. I'm
already aware of how MS does things, but the current implementation is
as it is due to limitations in the toolkits (backends) for our current
System.Drawing implementation. I'm in the process of redesigning the
toolkits so that functionality such as this is possible.

> Well, the advantages of using points instead of PathObjects would be:
>  a compact representation
>  straitforward implementation for most of the missing methods
>  fast and easy matrix transformation possible
>  straitforward implementation for GraphicsPathIterator
>  no need for the draw methods of (closed) curves in ToolkitGraphicsBase 

A complete and compatible implementation of GraphicsPath should be
available in pnetlib cvs within the next few weeks. If you would like to
help out with getting our System.Drawing complete and compatible,
figuring out the RegionData format would be _very_ helpful. It's
completely undocumented by MS and I've been unable to find any
information on it from other sources as well. I've spent some time
trying to figure out the format myself but I haven't made much progress
yet and I've got a lot of other work to do on our drawing code.


Rich