huxingyi

This framework has the potential to become another Qt if UI is been added. The coding style and document[0] are solid and similar with Qt[1]. Giving the Qt is heading in an unpopular way[2], I am looking forward to this framework.

[0] https://plywood.arc80.com/docs/modules/runtime/api/string/St... [1] https://doc.qt.io/qt-5/qstring.html [2] https://www.qt.io/blog/qt-offering-changes-2020

Neat! I checked the async.h, looks like it doesn't support nested calls? I did a similar toy many years ago, which could be used with other async libraries such as libuv: https://github.com/huxingyi/c-block

Thanks, that's good!

Thanks, to conclude from your detailed answer: I can still license the code other than CGAL to MIT, but license the whole project to GPL, and in the future if I removed CGAL, I can license the whole project back to MIT. And the plugin API is not related with CGAL at all, so I can let the user choose whatever the license.

Thanks, I checked that link. The "CGAL/Polygon Mesh Processing" package is licensed on GPL: https://doc.cgal.org/latest/Manual/packages.html

And the mentioned other libraries doesn't have the features I want.

Thanks, so that means the user have to open source their script also.

Thanks, I mainly use the CGAL/Polygon_mesh_processing to do mesh boolean. I saw some thread say CGAL commercial license cost $12K, it's not affordable for a side project yet.

Thanks, > 1. MIT can easily be re-licensed as GPL. You can keep a branch of it right before the introduction of CGAL, so if someone really wanted to, they could replace it with something else themselves.

It's too late, I introduced CGAL very early.

> 2. This is where it gets tricky. Yes, though you'll need the permission of all your contributors to the GPL version of your project, and replace all the GPL code ith similar functioning code. You should avoid just line-by-line rewriting of the GPL code though, you'll need to have a unique implementation.

So, I guess it's a Yes. If I removed CGAL in the future.

> 3. Again, tricky, but MIT would probably be OK, because MIT is compatible with GPL, if it's within the GPL'd project.

Basically, if I have to license my project on GPL because of CGAL, I would like the scripts wrote by user could license on whatever user like. But I am not sure if is possible.

The actual effort of writing one line of code in the specific language doesn’t mention in the article. Write one line of rust is extremely difficult than one line of Python, or even C++.

Looks like a really good start. One suggestion: add mesh boolean algorithm, currently, only CGAL do this well, if your library could supports this on the MIT license, I would definitely try it. Currently, I am using libigl to do parameterization and CGAL to do mesh boolean in my 3D modeling software[1]

[1] https://github.com/huxingyi/dust3d

Inspired by Protothreads, I implemented a similar framework 5 years ago, well, not exactly a framework, just several lines of macros, which could be used in general purpose not just embedded system (I am not sure if Protothreads can be used in general purpose). I used it in product environment with libuv, here is the github link: https://github.com/huxingyi/c-block

Thanks. Yes, it’s painful of manage cpp with different compilers, the cross platform of Dust3D mainly benefit from Qt.

Thanks for asking, there is a target in my mind that generate highly detailed mesh but maintained the easy usability. I don't know how far the target is away. The near roadmap is making this software could be used in the game I am developing exclusively(Without use other modeler), if something is missing, I would add it.

Thanks for the info, may have a try to see how Dust3D could benefit from this.

Input by direct type number of position and radius? Not yet.

Blue smudges! my god, LOL, That is the drawing of the front view of the mosquito! The blue lines dragged down in the GIMP is the reference line. Usually, when make reference sheet for Dust3D, the same position in the two profiles(Front/Side) should sit in the same height(Y coord). This video is a demonstration of how bad at drawing can still make good 3D model with Dust3D.

> What do you think of CGAL? I am using some function of CGAL in Dust3D, it's a great work. However, it doesn't work with iOS and it's license is more strict than MIT, may remove it in the future.

> Are you using a scene graph for the display? I am using Qt/QOpenGLWidget for display.

Thanks for verifying :-)

Thanks, the underlying mesh render use the Qt/QOpenGLWidget, haven't try it with OpenGL ES2.0 and OpenGL 2.x

should be ok with 3D printing, because the outcome of Dust3D generated is a watertight mesh. However, I don't have a 3D printer, cannot try yet.

Hi, thanks, currently support glb, fbx obj export, but without import supports, may be add in the future.

Thanks, I am really happy to hear that it can work with VR.

Thanks, but it's a little too late to rename, too many things need to sync, such as domain.

Hi, some reported several hours ago, and I have released 1.0.0-beta.4 to fix it, could you help to verify if it still crash please? Download link for windows: https://github.com/huxingyi/dust3d/releases/download/1.0.0-b...

What interaction patterns? You mean the weird combine two profiles in one window :-) ? That is my idea. The sphere guided mesh generation is based on a paper: "B-Mesh: A Fast Modeling System for Base Meshes of 3D Articulated Shapes"

Edit: I am not familiar with STL format, may have a look at it, thanks.

Dust3D use right-handed coordinate space, X - front view, Z - side view, Y - up. I am not quite sure what exactly coordinate space other software use, but I make sure the exported fbx looks as the same in Unity, and the exported glb looks as the same in Godot.

Oh, no, that guitar is my favorite :-( The main focus of this software is the modeling speed, for example, that guitar only took less than five minutes. So I can make more game asset as an indie. Actually, the model made recently is already looks far better than several months ago I made, I will keep improving the mesh generating quality of the algorithm. Thanks!