Struct ElmRoot

pub struct ElmRoot { /* private fields */ }

The main entrypoint for this crate.

Represents a directory with Elm files inside of it.

Implementations

impl ElmRoot

pub fn new<P>(path: P) -> Result<Self>

where PathBuf: From<P>,

Create an ElmRoot for a given directory. The directory should NOT be the one with the elm.json file in it, but the directory with the .elm files in it (with the default elm.json file, this is usually ./src).

The reason for this is: You may choose any source-directories in your elm.json you like, so we would need to parse elm.json and check all directories for the specified module.

pub fn debug(self) -> Self

Set the ElmRoot to debug mode.

This has two effects:

• Some println! logs to describe what the crate is doing
• The temporarily created files are not deleted

pub fn prepare<I, O>( &self, fully_qualified_function: &str, ) -> Result<ElmFunctionHandle<I, O>>

where I: DeserializeOwned, O: DeserializeOwned,

Prepare an Elm function for execution.

The given function name should be in the same form as you would call it in your Elm project when not using import aliases or unqualified imports.

E.g. if your function myFun is defined in the module MyModule.Submodule, you would pass in MyModule.Submodule.myFun.

The input and output types intended to be passed in subsequent calls have to be known at this point, either by type inference or by explitely specifying them. The reason for this is that we generate a wrapper application module for the requested function which needs type annotations (at least the type annotation for the port cannot be inferred).