Wayang is an engine for interactive storytelling, currently it focuses on enabling its user to prototype interactive story published in the form of website. Under the hood, Wayang uses Rust to parse story into sets of web pages, ready to be deployed anywhere a static website is available. Wayang is traditional Indonesian puppet show, it's a well known artform.
Story in Wayang is written using DSL (Domain Specific Language) also called Wayang that takes the .wyg file extension. Wayang engine will then parse this file and generates a website that consists of multiple interconnected web pages.
Wayang DSL is deesigned to enable rapid development of story. I omit usage of JSON, to reduce the probablity of syntax mistype. Currently, the syntax is as follows:
[scene 1]
Scene:
What, do you think? The water is clear. The sky is blue. The air is fresh.
Event:
The pool to your left and door to your right. Which one will you choose?
Choice:
- Go to left. -> [scene 2]
- Go to right. -> [scene 3]
This file should generate an HTML file roughly as follows
<html>
<head>
<title>scene 1</title>
</head>
<body>
<div class="scene">
<div class="sceneContent">
What, do you think? The water is clear. The sky is blue. The air is fresh.
</div>
</div>
<div class="event">
<div class="eventContent">
The pool to your left and door to your right. Which one will you choose?
</div>
</div>
<div class="choice">
<div class="choiceContent">
<a href="scene 2.html">Go to left.</a>
<a href="scene 3.html">Go to right.</a>
</div>
</div>
</body>
</html>
How this HTML is presented is then left to the implementation phase of the file, as it's just a matter of styling HTML and CSS.
At the moment the biggest milestone is getting the DSL properly parsed into said HTML format. After managed to parse it by hand, I realize that using parser Crate such as pest is a better option.
As an effort to parse .wyg file using pest, a grammar needs to be established first. The grammar.pest file is an effort to do so. The current grammar file is enough to parse the specified .wyg file to grab required data. If you're testing the grammar, pest.rs homepage has a nice parser playground for experimenting.
Clone this project and simply run cargo run in the cloned directory. It will do 2 things:
- Parse
.wygfile stored instorydirectory - Render
.htmlfile stored inresultdirectory, based on the parsing result and the.hbstemplate stored in/templatesdirectory.
Under the hood, what the parser does is parsing the .wyg file and store them as story struct. Afterwards html_template will read story and populate the .hbs template based on the values stored.
Important: please ensure that the line ending is LF and not CRLF. Check your IDE for detail on how to set that up.
I'm open for contribution! Just submit a PR that solve documented issue or suggesting new changes. To contribute:
- Fork this project
- Create a new branch name in the forked repo, you can name it anything, but it's better to name a branch to a certain issue or feature, so it's easy to review
- Create changes in that branch
- Submit the PR :D
Also, feel free to submit issue as well. I highly appreciate any form of contribution to this small project. 🙌