[cairo] API Shakeup: cairo_begin_group, cairo_end_group, cairo_get_group

[Carl Worth]

On Wed, 16 Feb 2005 10:46:42 -0500, Owen Taylor wrote:
> These are pretty clumsy names, especially
> cairo_create_group_begin_format().

Well, cairo_create_group_begin_with_format, but yes, this is getting
almost totally unintelligible.

> Also, introducing "create" in an unnatural place to indicate memory
> management is going to make  language bindings ugly for no good reason.
> 
> I think that the knowledge that end_group() returns a new pattern
> is a small piece of information that a) most people are going to guess
> right initially and b) we can reasonably expect the rest to learn.

OK. I'm convinced on this point.

I'm having a very enlightening discussion with Ian McIntosh right now,
(Gnome booth at LWE). His background is as a user of the API who has
already written a significant application[*] by just figuring out our
poorly documented stuff.

He's given me a lot of useful feedback on the new API changes. Most
things are going pretty well, (though explaining the rendering
equation took a while --- I think we'll just need a good picture).

He doesn't find the name "group" at all intuitive for what the
operation is here. Are we doing something intermediate? offscreen?
should the name be tied to the "pattern" we're making?

-Carl

[*] Roadster is a really slick mapping application that reads Tiger
data and displays road name labels along the roads:

	http://linuxadvocate.org/projects/roadster/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/cairo/attachments/20050216/49620c4d/attachment.pgp