How to support development?
Hi @RupertAvery, am I correct to say that you are the only one closing issues? I beleive it would benefit for DiffusionToolkit development to have in README some short instructions on how to help you with closing issues, cause the more people use it - the more issues will come. And I feel AI image generation is not going anywhere and people will use it more and more 😁
Also to add @RupertAvery, I wanted to send a PR with some improvements to the tag parsing for ComfyUI! This would help by parsing the metadata from other common custom nodes like KSampler Advanced, etc.
Is something like this desired? Can I just send a PR to the repo or is there some rule to follow? Thanks!
Hi. If you have forked this repo to your own github account, you can just create a pull request into this repo and it will appear here.
The thing about ComfyUI is that the node will only work for certain workflows. Any slight difference and it won't really work. I've been planning to add a way to let users map their own workflows, just haven't gotten around to it.
I see! I plan on trying to add some kind of abstract / concrete / interface class thing - lets say we have the base Clip Encoder that is standard from Comfy and a couple other classes that implement parsing for other encoders. During parsing it would take the first one that is found (searching from the base to the custom nodes), but I agree some way to map a "preferred node" for each category would be great!
I will work on the PR and send it your way then. Thanks!