# Build

## Using local VTK WASM build tree

This example needs VTK.wasm compiled with `VTK_WEBASSEMBLY_THREADS=ON`. If you're compiling
with vtk-wasm-sdk image from dockerhub, make sure to use the threads variant.

Compared to `AsyncClipper`, the main difference in this example is that there is no `main(int, char**)`
function and the entire application is driven from javascript through C++ functions wrapped with embind.

This example shows in detail how to create a pthread and transfer the
canvas offscreen so that the new web worker can render into it. In the
`AsyncClipper`, we used `-sPROXY_TO_PTHREAD` link flag which did all the
tasks necessary to run our `main(int, char**)` on a new web worker.

```
emcmake cmake \
  -G Ninja \
  -S /path/to/vtk/Examples/Emscripten/Cxx/Cone \
  -B out/build \
  -DVTK_DIR=/path/to/where/vtk/wasm/was/built

cmake --build out/build
```
## Using vtk-wasm-sdk

### Linux/macOS
```sh
# Configure
docker run --rm -it \
  -v"$PWD":/work kitware/vtk-wasm-sdk:latest \
  emcmake cmake -GNinja -S /work -B /work/build -DVTK_DIR=/VTK-install/Release/wasm32-threads/lib/cmake/vtk
# Build
docker run --rm -it \
  -v"$PWD":/work kitware/vtk-wasm-sdk:latest \
  cmake --build /work/build
```

### Windows
```sh
docker run --rm -it `
  -v"$PWD":/work kitware/vtk-wasm-sdk:latest `
  emcmake cmake -GNinja -S /work -B /work/build -DVTK_DIR=/VTK-install/Release/wasm32-threads/lib/cmake/vtk
# Build
docker run --rm -it `
  -v"$PWD":/work kitware/vtk-wasm-sdk:latest `
  cmake --build /work/build
```

# Serve and test generated code

This example uses wasm in web workers and needs `SharedArrayBuffer`.
For that reason, use the `server.py` from this directory which configures the required
HTTP headers for browsers to allow `SharedArrayBuffer`.

```
cd out/build
python3 ./server.py
```

Open your browser to http://localhost:8000