John, as the current maintainer, would you be willing to start the conversation on the pyside-dev list? I am in for being a contributor and it was my intention to start working through the typesystem definition files for Qt5.
Steve > On Dec 11, 2014, at 11:13 AM, John Ehresman <[email protected]> wrote: > > On 12/11/14, 10:49 AM, Todd Rme wrote: >> Taking a bit of a step back, what is currently holding Qt 5.x back >> from allowing completely dynamic bindings? Has anyone talked to the >> Qt devs to see if these pieces could be implemented? > > Qt 5.x support is mostly a matter of making changes to compile against > Qt5, do the split of the QtGui, go through deprecated and new classes & > methods, etc. It's a substantial amount of work but should be > relatively straightforward. I don't know what, if anything, could be > changed in Qt to make things easier given the need to maintain backward > compatibility in Qt. > > Related but orthogonal is fixing a number of serious bugs that stem from > assumptions that shiboken makes about object lifetime. This is what > motivates the desire to either fix shiboken or move to another binding tool. > > I think what PySide needs is a small group of developers to work on it. > There is a developer mailing list at [email protected] and I > think we should discuss how development should move forward. Note that > to contribute to the PySide core, a knowledge of C++ and the Python C > api or the willingness to learn them is required. > > John > _______________________________________________ > PySide mailing list > [email protected] > http://lists.qt-project.org/mailman/listinfo/pyside _______________________________________________ PySide mailing list [email protected] http://lists.qt-project.org/mailman/listinfo/pyside
