Re: cairo_gstate_t?
Fred Kiefer <[email protected]>
| Newsgroups | gmane.comp.lib.cairo |
|---|---|
| Message-ID | <[email protected]> |
On 10.10.2011 12:31, Chris Wilson wrote: > On Sat, 8 Oct 2011 21:40:33 +0200, Fred Kiefer<[email protected]> > wrote: >> I would highly welcome that addition to cairo. This is the >> interface we need for GNUstep to easily implement the display >> Postscript interface we need to provide. Up to now we needed to >> work around this restrictin by adding a rather ugly and inefficent >> layer ourselves. > > I am a little wary of exposing what I consider an internal > implementation detail of Cairo. More so now that we have > opportunities for semi-native backends that may not be using > cairo_gstate_t in their fast paths. > > Can we have a bit of discussion on the concrete use-cases people > have for a named gstate? In GNUstep we use cairo in one of our different drawing backends to implement a drawing concept similar to display Postscript. One requirement there is to be able to store a gstate (for example in NSWindow) to use it later on. That is not just to push and pop gstates, which is available in cairo easily, but to capture a gstate and reuse it later on. As far as I know this isn't currently possible that easy with cairo and we have some horrible code to extract all internal state from a gstate and apply it later on. Fred -- cairo mailing list [email protected] http://lists.cairographics.org/mailman/listinfo/cairo