[PATCH wayland 2/3] scanner: Allow adding a prefix to exported symbols

[Pekka Paalanen]

On Mon, 22 Jan 2018 14:46:15 +0000
Emil Velikov <emil.l.velikov at gmail.com> wrote:

> On 19 January 2018 at 15:49, Derek Foreman <derekf at osg.samsung.com> wrote:
> > On 2018-01-19 01:22 AM, Jonas Ådahl wrote:  
> >>
> >> On Thu, Jan 18, 2018 at 10:48:14AM +0200, Pekka Paalanen wrote:  
> >>>

> >>>
> >>> the original proposal was to prefix ABI symbols, and leave everything
> >>> else as is. Maybe writing down again how exactly that is supposed to be
> >>> used and what problems it solves would crystallize the idea. Perhaps in
> >>> the form of wayland-scanner user instructions which the original patch
> >>> seems to be lacking?
> >>>
> >>> Earlier I didn't like prefixing only some bits of the API, but on
> >>> second thought, if that's all what's needed, maybe it isn't that bad
> >>> from hand-written code readability point of view?
> >>>
> >>> The discussion showed that any other solution becomes hard and/or messy
> >>> compared to that. This is not a library ABI/API change either, just a
> >>> new operation mode in wayland-scanner.
> >>>
> >>> I join you with the question about use cases and how badly do we need
> >>> this.  
> >>
> >>
> >> I don't see the point in keeping xdg-shell unstable v5 support in
> >> weston, so from that point of view, *weston* I'd say don't need any
> >> symbol prefix feature.
> >>  
> >
> > Ok, I guess at this point we should drop it from consideration for the next
> > release.
> >
> > Pekka's synopsis is a really good status report for anyone interested in
> > picking this up later without reading all the previous threads.
> >  
> Something I mentioned over IRC, but forgot to add it here.
> QT5 has been using unstable protocols for a while now. Combine that with:
>  - interface symbols are exported by default
>  - binary-only applications ship with their own version of QT (some
> even have wayland-client.so and wayland-egl.so...)
>  - some of those won't be updated to use the stable interface
> 
> This brings us to the original topic - symbol collusion. Jonas' idea
> sounds reasonable to me.

Hi Emil,

all those issues sounds like they would be solved by not exporting
symbols and ensuring to use only local symbols for resolving, which are
both quite orthogonal. One cannot change a prefix every time when
building a different version of a protocol definition. Where not
exporting and local resolution would not be available, arguably neither
would prefixing.

Therefore I think we should concentrate on the no-export feature rather
than prefixing.

Let's try hard to drop xdg-shell v5 completely, just like originally
intended. I want to keep the distiction between unstable and stable.


Thanks,
pq
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.freedesktop.org/archives/wayland-devel/attachments/20180122/c531ff59/attachment.sig>