SL4F

The core of the project is an HTTP server called Scripting Layer for Fuchsia (SL4F), which runs on Fuchsia devices. The server processes JSON RPC requests from remote host driven test frameworks in the form of core service commands, parses these requests, and fulfills them with the appropriate Fuchsia equivalent (FIDL) commands.

Getting Started

SL4F Includes

To include SL4F by default to the build include it with your build: --with garnet/bin/sl4f:bin

Pushing incremental changes

1. Make sure SL4F is running.
2. Build and push with this command fx build-push sl4f.
3. Before changes will take effect make sure there are no actively running SL4F instances. Within fx shell run killall sl4f.cmx.
4. Run SL4F within an fx shell instance: run fuchsia-pkg://fuchsia.com/sl4f#meta/sl4f.cmx. Note that you can tune your test framework to open an ssh connection to run SL4F in the background.

Facades

Facades is a term used in SL4F to represent the actual wrapper for FIDL apis and maintain statefulness of the context. For example if there is an asyncronous event the Facade will keep track of anything important to verify for testing at a later time.

Logging

In garnet/bin/sl4f/server/sl4f.rs there is a macro called with_line. Use this with your tag in your log lines.

Example:

200: let tag = "GenericFacade::func";
201: fx_log_info!(tag: &with_line!(tag), "{:?}", "Really important log.");

This outputs to

GenericFacade::func:201 Really important log.