[cairo] openvg still used ?

[Mike Gran]

> On Wednesday, August 17, 2016 9:27 PM, Dave Yeo <dave.r.yeo at gmail.com> wrote:
> > On 08/16/16 10:25 PM, Bryce Harrington wrote:
>>  On Fri, Aug 12, 2016 at 11:12:56PM -0700, Dave Yeo wrote:
>>>  On 08/12/16 10:41 PM, Enrico Weigelt, metux IT consult wrote:
>>>>  On 12.08.2016 22:59, Bryce Harrington wrote:
>>>> 
>>>>>  I don't know much about OpenVG but you're right it 
> doesn't appear to
>>>>>  have really caught on generally; I'm not spotting any users 
> of the Cairo
>>>>>  OpenVG backend either.  I wouldn't mind seeing us drop it 
> and give more
>>>>>  focus to OpenGLES...
>>>> 
>>>>  Maybe we should move them to separate packages.

Here's the Cairo discussion when they first talked about
making an OpenVG backend, back in 2005.

https://lists.freedesktop.org/archives/cairo/2005-August/004787.html

It appears that Cairo wraps OpenVG calls, instead of the
other way around, so splitting it into a separate package might
require some work on a "backend-writing API" that
lets one map Cairo actions to backend actions, but, allowing
Cairo rendering for operations that don't map.

As an aside, how does one test OpenVG?  There appear to be
OpenVG software renderers and OpenVG to GL converters. There
still are chips being built with OpenVG pipelines, but,
I don't see any cheap developer-friendly hardware options that
could be used to test it, without descending into buying
ARM embedded dev kits.

Splitting it out into a separate package might be an interesting 
exercise, just to see how easy they are to separate. 


[...]


>> 
>>  Also, are you favoring only the OpenVG deletion or both it and the Skia
>>  backend?
>> 
> 
> To be honest, I don't have an opinion on actual deletion besides hating 
> to see deletion but realizing unmaintained code rots and should be pruned.
> I use the OS/2 backend and lurk here in case a discussion such as this 
> comes up. It's a shame that there aren't any OpenVG or Skia people here 
> to put in there 2 cents
>