On Thu, Aug 10, 2017 at 10:44:30AM +0200, Marco Martin wrote:
> On Thu, Aug 10, 2017 at 3:47 AM, Jonas Ådahl <jad...@gmail.com> wrote:
> > On Wed, Aug 09, 2017 at 01:42:51PM +0200, Marco Martin wrote:
> >> On Wednesday 27 July 2016 15:54:58 Jonas Ĺdahl wrote:
> >> > xdg-foreign is a protocol meant to enable setting up inter surface
> >> > relationships across clients. Potential use cases are out-of-process
> >> > dialogs, such as file dialogs, meant to be used by sandboxed processes
> >> > that may not have the access it needs to implement such dialogs.
> >> a quick feedback while trying on implementing it in kde side.
> >> since we use c++, the file generated by wayland-scanner, won't compile due
> >> to
> >> the request called "export" which is a reserved keyword in c++11.
> >> could the request be renamed to something else, even just a bit more
> >> redundant
> >> as export_surface which would be safer as compilers are concerned?
> > Ah. Would make sense with a test case for this in wayland-scanner I'd
> > say, so we don't add other things that would make it not compile with a
> > c++ compiler.
> > Anyhow, "export_surface" or maybe even "export_toplevel" (as that is the
> > only
> > thing we allow exporting anyway) seems fine to me. The "import" request
> > should be renamed in a similar manner as well then.
> yeah, export_toplevel/import_toplevel sounds good.
> if there is a continuous integration somewhere, could be nice just
> make it try compiling the file resulting from wayland-scanner with g++
> -std=c++14 or something like that
There is a very tiny test suite in wayland-protocols. It could be
extended to test build for various variants of C and C++. Currently it
just tests that wayland-scanner doesn't complain.
> Marco Martin
wayland-devel mailing list