My NANY contribution for this year is a simple description language for drawings and diagrams (which I've named
Drawskrit). My main goal is to be able to describe diagrams with text. The description language should be natural and simple (I don't want a diagram programming language).
Its current state is experimental. I have a JavaScript implementation that demonstrates the current specification. I'm afraid it's not very useful yet though. I'm not sure I'll ever get it to a state where it'll be powerful enough to specify real-world non-toy diagrams without becoming a complex monster language.
A few of the current obvious shortcomings:
- needs more shapes (e.g. arrows)
- current text alignment sucks (specification bug)
- dotted/dashed looks horrible with thick lines (implementation bug)
- needs more documentation (e.g. layers)
- JS implementation needs to be decoupled from the HTML demo page
If I ever get it to a state where it's actually useful, I'm planning to implement it as a cross-platform tool which can be integrated into an automated build process.
A non-formal specification (i.e. a description) and the JavaScript implementation can be found here:
https://github.com/phitsc/drawskritDrawskrit is open source.
If someone wants to play around with it please direct your browser here:
http://rawgit.com/ph...er/js/drawskrit.htmlAs a first example, paste the following text into the big text box on the left of the above web page:
orange background filled shapes
_
small blue circle small green circle
tiny purple square big red triangle tiny purple square
small olive smile
==
fat navy huge empty dashed circle