[Bf-committers] GSoC 2018: Questions on Layer Painting

Cheryl Chen chnchryl at gmail.com
Sun Mar 11 13:43:12 CET 2018

*Hi there,I've been looking into the layer painting project for GSoC this
year, and there are a couple things I’m wondering about.The project
description mentions that it is about adding multi-layer image support to
Blender to bring the painting pipeline closer to Blender, so it seems like
this will be focusing on texture paint mode to extend the current paint
slot system to support sets of layers.What is the extent of the project,
and there are particular features that need to be implemented? Along with
multi-layer support, is it just to add basic operations such as reordering,
visibility, and export of the layers as a single image, or also to add more
advanced features such as masking, blend modes, multilayer painting, and
multilayered exports?How should layers be treated as a data structure, and
how should this be integrated into the current codebase? Would it be
reasonable to represent a layer as a struct which holds an image and some
layer data (such as the ordering and visibility), and extend TexPaintSlot
to hold a list of those layer structs? I haven’t really had experience with
this, so I’m not sure if my approach is too simple and naive.Since the
project involves introducing a new system, I’ll need to make changes and
possibly add new features to the UI. Could I get some guidance on how to
approach this? For example, I want to change the appearance and behaviour
of  list elements in the list of ‘Available Paint Slots’ to increase the
size of the image preview, add a sublist feature, and allow it to be
dragged and dropped. Would this involve creating a new kind of data
structure in C, or combining already existing UI elements in a python
script?Thank you for your time!Best,Cheryl Chen*

More information about the Bf-committers mailing list