rbop

rbop (Rust Binary Operations) is a framework for implementing intuitive mathematical expression editors.

rbop is no_std, so you can use it pretty much anywhere. To create an editor for your particular use-case, all you need to do is provide simple method implementations to draw core mathematical glyphs. rbop will use these to calculate a two-dimensional layout and draw to your chosen canvas!

Try it out

rbop comes with a simple ASCII renderer, which is used in an example named ascii_calc. If you cargo build --examples --features examples, you'll be able to run this example and get a feel for how natural rbop's editor feels!

Documentation

There isn't too much proper documentation yet. The two examples ascii_calc and window_calc are heavily commented, and designed to be read (in that order) to see rbop's usage in action.

Implementing a renderer

Refer to AsciiRenderer for a pretty good example of this. You'll need to implement the Renderer trait, which will allow rbop to:

• Reset and prepare your graphics surface with init
• Determine the size which a particular glyph will be when drawn to your canvas, using size, to calculate a layout
• Draw glyphs to your canvas at a particular location with draw

Do not override the default implementations of any other methods in Renderer.

Node trees

There are two available node types.

An UnstructuredNode tree is really easy to build through user inputs. Horizontal inputs are left as token streams, so operator precedence does not need to be considered.

1+2*3             Fraction
-----             /      \
  5           1,+,2,*,3   5

A StructuredNode tree contains no raw token streams, and the tree structure fully expresses the correct operator precedence.

1+2*3             Fraction
-----             /      \
  5             Add       5
               /   \
              1    Mult
                  /    \
                 2      3

An unstructured node tree can be converted to a structured node tree by upgrading it. Some functionality, such as evaluation, can only be performed on a structured node tree.