On Friday 05 August 2011 09:31:59 [email protected] wrote: > On 5.8.2011 1.28, "ext Hugo Lima" <[email protected]> wrote: > >Matti, I think we don't need a PSEP, PSEPs are for PySide things, this > >is specific to Shiboken based bindings, PySide is included as well but > >it isn't the focus, I mean, this module is not made to access PySide > >internal stuff, but Shiboken internal stuff, it's highly tied to > >Shiboken, if we create a PSEP and change the PySide backend from > >Shiboken to something else in an unknown future the PSEP could turn > >into a non sense PSEP. > > OK, but Shiboken is developer under the PySide umbrella, and hence our > commitments to an open-governance process and the PSEP design approach > apply here too. The proposed module would be a non-trivial Python API > closely related to PySide, and therefore would still require an open > design process, IMO. > > Also, the PSEP also works as a specification and documentation for new > featuers, and they would be required in any case for a new module, even if > the PSEP wasn't done. > > There are no plans to replace Shiboken in PySide, so that's a moot point. > But even if that would happen, there would be absolutely no harm done. The > PSEP would just become obsolete then.
I'm still not convinced about the need of a PSEP for this but I can (will) write one if it is required. PSEP means open design process, but the absence of an PSEP doesn't mean a closed design process, it can be open as it is at the moment, just with less bureaucracy. I'm also *not* against the PSEP idea, it's a good idea, I just have the opinion that PSEPs should take care of PySide Python API only, not the internal ecosystem surrounding the bindings generation/runtime. > >This also explain why to name the module "shiboken" instead of put it > >into PySide module, there's no technical reasons to make this module > >depend on PySide and being a separate module all Shiboken based > >bindings can have gain. > > Point taken. This I can live with. > > >P.S.: Suggestions about the module implementation, what functions it > >need to have, etc could be made here or on the bug report[1], I'll > >read both anyway. > > > >[1] http://bugs.pyside.org/show_bug.cgi?id=902 > > No need for commenting on the bug - we have the well-defined PSEP process > for defining implementations anyway. ;-) > > ma. -- Hugo Parente Lima INdT - Instituto Nokia de Tecnologia
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ PySide mailing list [email protected] http://lists.pyside.org/listinfo/pyside
