Scripts for building and downloading pre-built versions of SpiderMonkey, compiled to wasm32-wasi as a static library.

Building from upstream source

The build-engine.sh script can be used to build either release or debug builds. It's recommended that this script only be used in CI environments, as it bootstraps a build environment for SpiderMonkey on each invokation.

The source is retrieved from the repository in gecko-repository at the revision in gecko-revision.

Building for release or debug

The script can compile both release and debug builds, with release being the default. To compile a debug build, pass debug as the first argument:

sh build-engine.sh debug

Build output

Running the build script will result in three different subdirectories in the current directory:

• include, containing the public header files
• lib, containing the object files and static libraries needed to statically link SpiderMonkey
• obj, the output directory the SpiderMonkey build system creates. This can be ignored or deleted

Downloading pre-built releases

The download-engine.sh script can be used to download pre-built object files and headers for static linking for either release or debug builds. The object files are compiled to wasm32-wasi, and can be used on any platform.

Running download-engine.sh [debug] will download a tarball for a SpiderMonkey build (release by default, debug if the debug argument is passed), and extract it. When successful, it'll result in the subdirectories lib and include in the current working directory.

Note: download-engine.sh can be called from anywhere, but must itself reside in a git checkout whose origin is the github repository to download the release from. It expects that repository to have a release whose name is derived from the current git revision of the containing checkout.