Hi there, At Qt CS I hosted a session about the clang based tooling I wrote to automate the boring parts of porting from Qt4+Qt3Support to Qt5.
http://www.kdab.com/automated-porting-from-qt-4-to-qt-5/ We discussed: * The scope and limitations of the tool - Works best for boring stuff, not something that requires complex porting because for example, entirely different APIs have to be used in some cases when methods get deprecated. * How the tool works - I showed and explained the code and the result after running it in kdelibs. * Whether anyone else is interested in using or contributing to it - At least Ben Meyer was interested in trying it out and contributing a few patches. * Where it should live - As it depends on non-public API of clang, the clang source tree is a good candidate, however, that may inconvenient for many Qt contributors. There is a possibility of getting some or most of it into clang itself (http://thread.gmane.org/gmane.comp.compilers.clang.devel/22253/focus=22261) where a contributor community for simliar refactoring fragments is starting to emerge. The Qt-Creator repo, which might also introduce a clang dependency could also be a candidate. * How it should be used - It has potentially long term use-cases for KDAB at least as we'll be doing ports from Qt 3 and 4 to 5 for some time, but also could be used to help porting away from API deprecated in Qt 5.x potentially. Feel free to ask anything else about the tooling here. Thanks, -- Stephen Kelly <stephen.ke...@kdab.com> | Software Engineer KDAB (Deutschland) GmbH & Co.KG, a KDAB Group Company www.kdab.com || Germany +49-30-521325470 || Sweden (HQ) +46-563-540090 KDAB - Qt Experts - Platform-Independent Software Solutions
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ Development mailing list Development@qt-project.org http://lists.qt-project.org/mailman/listinfo/development
