A Javascript implementation of the Poincaré disk model of the hyperbolic plane, on an HTML canvas.
Usage examples can be found on the project site.
This library prioritizes the visualization of hyperbolic geometry over precise mathematical calculation. Due to the less-than-infinite precision of floating-point numbers, and because certain trigonometric functions are ill-conditioned, these goals are often at odds.
The arbitrary constants HyperbolicCanvas.INFINITY and HyperbolicCanvas.ZERO have been defined for use in internal comparisons in place of Infinity and 0, respectively. Their values may be overridden, but increased accuracy will tend to lead to more unpredictable behavior.
This library uses Jasmine specs to validate the code and prevent regressions.
The specs have been written to use random input values. While this approach is unconventional, it provides more convidence than would an attempt to test an effectively infinite number of edge cases. Some specs do occasionally fail; the frequency at which this occurs is determined by the accuracy of the constants HyperbolicCanvas.INFINITY and HyperbolicCanvas.ZERO.
The Jasmine library itself has been modified to run each spec multiple times, and a random number seed is used so that errors may be reproduced. The seed and the spec run count can be set in the options menu on the SpecRunner page.
Certain browsers do not provide support for the hyperbolic trigonometric functions. Polyfills are available.
<script type="application/javascript" src="https://github.com/ItsNickBarry/hyperbolic-canvas/raw/v1.0.1/dist/hyperbolic_canvas.js"></script>
npm install --save hyperbolic-canvas
Pass a unique selector of a div element, to the function HyperbolicCanvas.create. Nonzero width and height styling must be specified. Absolute px values in a 1:1 ratio are recommended:
let canvas = HyperbolicCanvas.create('#hyperbolic-canvas');
See API.md for a list of functions and their descriptions.
$ claude mcp add hyperbolic-canvas \
-- python -m otcore.mcp_server <graph>