Introducing Grammar DB

I have been thinking and tinkering with this idea for a long time. It came from working with NodeJS and Rails spin-off languages, like coffeescript, babel, pug, slim, etc. I liked those languages, but using them often means you have to sell out to whatever platform wrote them, or look for some variant in your platform.

I wanted to be able to say, “I want to use a pug template”, (indent-oriented HTML markup), but not have to install the NodeJS compiler which generates code in Javascript. I might want to use it from Python or Go or some other language. There are ports to different languages, but they don’t always work as expected. I know building parsers takes a lot of time and effort to work correctly. Then I realized that language definitions are pretty language-independent, so I should be able to generate code for a parser in each language based on that definition.

I’ve tinkered around with architecture and am getting close to something that works. I thought of writing the language specifications into a data representation that can deal with inheritance and has enough information to generate working parsers. I also thought this could be used to generate specifications for languages for editors, like Atom, Visual Studio, Unstructured, or even generate Javascript web page editors for that language.