elm-static-html-lib

Generate static html by passing an optional json object to your Elm views.

Library version of elm-static-html.

Install

npm install --save elm-static-html-lib

Usage

with an argument

In this example, we use decodeModel to turn the passed JSON into a model that our view can use.

import elmStaticHtml from "elm-static-html-lib";


const model = { name: "Noah", age : 24 };
const options = { model : model, decoder: "MyModule.decodeModel" };

elmStaticHtml("/absolute/path/to/elm-package.json", "MyModule.view", options)
.then((generatedHtml) => {
    fs.writeFileSync("output.html", generatedHtml);
});

without an argument

In this case, our view has the type Html msg.

import elmStaticHtml from "elm-static-html-lib";


const options = { };

elmStaticHtml("/absolute/path/to/elm-package.json", "MyModule.view", options)
.then((generatedHtml) => {
    fs.writeFileSync("output.html", generatedHtml);
});

With no indent

In order to truly match what Elm generates at runtime, you may not want to have spaces or indent inserted. You can do this by setting the newLines and indent options like so:

import elmStaticHtml from "elm-static-html-lib";


const model = { name: "Noah", age : 24 };
const options = { model : model, decoder: "MyModule.decodeModel", newLines: false, indent: 0 };

elmStaticHtml("/absolute/path/to/elm-package.json", "MyModule.view", options)
.then((generatedHtml) => {
    fs.writeFileSync("output.html", generatedHtml);
});

multiple at once

When you want to render many views - particularly when they share dependencies - it is faster to use the multiple function.

const configs = [ 
    { viewFunction: "MyModule.view", model, decoder: "MyModule.decodeModel", fileOutputName: "grouped1.html" }, 
    { viewFunction: "MyModule.lazyView", model, decoder: "MyModule.decodeModel", fileOutputName: "grouped2.html" }, 
    ];

elmStaticHtml.multiple("/absolute/path/to/elm-package.json", configs)
.then((generatedHtmls) => {
    generatedHtmls
        .map((output) => fs.writeFileSync(output.fileOutputName, output.generatedHtml));
});

API description

elmStaticHtml(packagePath, viewFunction, options)

• packagePath (String): An absolute path to the elm-package.json of your project.
• viewFunction (String): Qualified name to the view function. Format <ModuleName>.<functionName>
• options (object): A map of options. Can be either empty or contain a model and a qualified decoder name. See above for usage details.

elmStaticHtml.multiple(packagePath, configs)

• packagePath (String): An absolute path to the elm-package.json of your project.
• configs ViewFunctionConfig[]: An array of configurations, see below.
• alreadyRun? (Boolean): When true, doesn't generate boilerplate again. Useful if only your models have changed, and not your elm code.
• elmMakePath? (String): Specify the path to elm-make.
• installMethod (String): Specify a custom package installation command.

returns Output: An object containing the generated html and the outputFileName.

export interface ViewFunctionConfig {
    viewFunction: string;
    fileOutputName: string;
    model?: any;
    decoder?: string;
    indent?: number;
    newLines?: boolean;
}

• viewFunction: Qualified name to the view function. Format <ModuleName>.<functionName>
• fileOutputName: File name. This value is not touched and given back to Output, to make saving the generated html easier
• model?: Optional object that is given as the model to your view function
• decoder?: Qualified name to the decoder for model. Format <ModuleName>.<decoderName>
• indent?: Optional formatting flag. Sets whether the generated html should be indented (default true)
• newLines?: Optional formatting flag. Sets whether new tags should start on a new line when (default true)

export interface Output {
    generatedHtml: string;
    fileOutputName: string;
}

• generatedHtml: The html that your view has produced
• fileOutputName: A file name that is threaded through for convenience

More examples

Check out the example folder for a more in-depth example.

Production

If you are running this in production, you may want to only generate the boilerplate files once. You can do that by setting the option alreadyRun to true. When alreadyRun is true, the Elm app is only started -- no boilerplate is generated.