Hi @wolthera
I agree that using C++ for this plugin would be great for performances (and for Krita also, to have this natively in filters
), but for now I’m not just able to do it.
I already tried to understand code from krita/plugins/filters/halftone/kis_halftone_filter.cpp file because it’s not working properly when called from PyKrita API, but for now even if I’m able to understand the code in his globality, as I already told to @tiar I think I’ve some step to pass before starting to work on Krita’s source code:
- I’m relatively a newbie with C++ language
- I need to understand Krita’s dedicated Kis* classes and architecture
- I need to learn Qt a little bit more (with this plugin I think I now have some basis, but not enough, and working on python plugin with PyQt is easier for me to learn Qt concept than trying to learn Qt and C++ at the same time)
- I need to learn how to use git (ok I’ve already started this point
) - I need to be able to compile Krita (next step after git learning, should not be so difficult, but it’s another step to learn)
- I need time (I currently have 2 more plugins in my mind and I want to be able to work on them before really starting C++)
So for now we will have do with the python plugin, waiting to have something better in Krita’s native filters, I think it’s a nice workaround ![]()
And in few month I’ll try to work on a C++ version of the plugin, but before I think I need to try to fix one or two bug that @tiar listed as they’re defined as ‘relatively easy’.
But as you can see the plugin provides many options and it won’t be just trying to fix problem with current halftone filter, it make sense to code the plugin with C++ only if all current plugin functionalities and options are reported in C++
In any case, the fact you asked me to rewrite the plugin as a native C++ to be part of Krita touches me and comforts me on the quality of the plugin ![]()
Grum999