On Tuesday, 18 February 2020 04:11:01 PST Christian Kandeler wrote: > In qbs, we use it to call setpgid() with the id of the newly created > process. But I don't understand why that should matter at all: Whether we > inject the code via an overriden virtual or a std::function is purely a > question of implementation technique, is it not? How can there possibly be > a functional difference?
The difference is how you communicate to QProcess that you want to run extra code. Right now, you do this by deriving from QProcess and overriding setupChildProcess. That means QProcessPrivate::start has no proper way of knowing when you did so and must assume that you do want that, preventing us from having some optimisations. See the two commits I've just uploaded to QProcess 5.15: they are not allowed if user code is running. I'm asking for a way that we can be certain that you want to run code. It will clean up your code and QProcess's. And if we like it, we can have a flag saying "my function is async-signal-safe too", which will allow the optimisations and the use code. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel System Software Products _______________________________________________ Development mailing list Development@qt-project.org https://lists.qt-project.org/listinfo/development
