35 releases
0.10.2 | Mar 2, 2021 |
---|---|
0.8.9 | Nov 16, 2020 |
0.8.1 | Jul 12, 2020 |
0.5.0 | Mar 13, 2020 |
0.0.1 | Sep 7, 2018 |
#41 in Data formats
79 downloads per month
Used in webblender
5MB
4K
SLoC
Landon
A collection of tools, data structures and methods for exporting Blender data (such as meshes and armatures) and preparing it for your rendering pipeline.
Initial Background / Motivation
Before this module I would export blender mesh / armature data to COLLADA using blender's collada exporter, and then parse that COLLADA into JSON.
This worked mostly well - but here and there I'd run into a model that didn't export quite right and I'd have to dig around to figure out why.
After a year or two of this occasionally happening.. I finally decided to invest some time in writing something myself, knowing that I'd still run into issues here and there, but they'd be issues that I'd know how to address.
The goal of landon
is to be a minimal suite of heavily tested, well documented tooling
for getting data out of Blender and a set of functions for pre-processing that data so that you can
make use of it in your rendering pipeline.
From the beginning landon
will be targeted towards my needs for my game Akigi, but please
feel very free to open issues / PRs with questions / thoughts / functionality that you think might fit into landon
.
The goal is that getting data out of Blender and into your rendering pipeline becomes easy as pie.
Getting Started
The Landon Book is a work in progress guide with examples on how to use the libraries
in landon
.
Take a look at the mesh-visualizer directory to see a full working example of implementing skeletal
animation with models that were exported using landon
.
Quick Start
Here's an example where we'll download a Blender file, export the meshes to JSON and then extract each bounding box from the JSON using the jq CLI.
# Install landon
cargo install -f landon
landon install --mesh-to-json --armature-to-json
# Download a Blender file to try landon with
BLEND_FILE='https://github.com/chinedufn/landon/blob/master/crates/blender-export-test/src/tests/multiple_meshes.blend?raw=true'
curl -L $BLEND_FILE > /tmp/multiple-meshes.blend
# Export meshes and armatures to stdout and redirect stdout to a file
landon export -f /tmp/multiple-meshes.blend > exported.json
# List all of the mesh names and bounding boxes
cat exported.json | jq -r '.meshes | to_entries[] | .value | to_entries[] | "\(.key), \(.value | .bounding_box)"'
# Second_Mesh, {"min_corner":[-1.3121787,0.44901967,0.67399526],"max_corner":[0.7619256,2.523124,2.7480996]}
# AMesh, {"min_corner":[-3.2487504,-3.3098261,1.2566323],"max_corner":[-1.2487504,-1.3098261,3.2566323]}
# Mesh3, {"min_corner":[-1.2058887,-2.4149196,-1.8447866],"max_corner":[0.86821556,-0.3408153,0.22931767]}
To Install
We currently support Blender 2.80
cargo install -f landon
landon install --mesh-to-json --armature-to-json
# FIXME: landon install --ik-to-fk
npm install -g ik2fk && ik2fk --install
# More info
landon install --help
API
Landon provides a Rust API for exporting data programatically.
In the future we will also provide C and Wasm APIs as light wrappers around the Rust API in order to enable interop with other languages.
CLI Usage
# Help on all of the subcommands
landon --help
Running the mesh visualizer locally
TODO: Rewrite this example and remove watchexec as a dependency.
# Install a static server that sets the application/wasm mime type
npm install -g http-server
# Watcher
cargo install watchexec
git clone https://github.com/chinedufn/landon
watchexec -r -w mesh-visualizer --ignore mesh-visualizer/out ./mesh-visualizer/build.sh
http-server ./mesh-visualizer/out --open
Your web browser should open up with an application that allows you to visualize all of the model's in our test suite.
Contributing
Please open issues explaining your intended use case and let's see if we should or shouldn't make landon
support it.
Also feel free to open issues with any questions / thoughts that you have!
To test
cargo test --all
TODO
- BlenderMesh's triangulate function can deal with ngons. Right now only handles 3 or 4 faces
See Also
License
MIT
Dependencies
~8MB
~152K SLoC