Skip to content

Latest commit

 

History

History
56 lines (47 loc) · 1.82 KB

README.md

File metadata and controls

56 lines (47 loc) · 1.82 KB
Raw

Graph

This is a basic plotting library, written in nim.
The end goal is to have a tiny plotting lib to use with jupyternim
Outputs .png or .svg files, or a string that contains the png as binary data or the svg as string.

For what I want to achieve and where I'm at, see target

Some examples are in examples:

Note: text labels are WIP, only for svg atm

Example

current

import graph, math, arraymancer
let 
  x  = arange(0.0'f64, 10,0.1)
let 
  y  = sin(x)
  y2 = cos(x)
var srf = plot(x.data,y.data)
srf.plot(x.data, y2.data)
srf.grid
# Save to file
srf.saveTo("currentpng.png")
srf.saveTo("currentsvg.svg")

Current structure

  • graph: exposes everything ( basic functionality )

Inside graph there are specific apis:

  • color: exposes various colours and the proc color(r,g,b,a)
  • plot: initializing the plot, adding plots
  • surface: the implementation of Surface and Axis
  • backend<x>: handles rendering the plot

TODO:

  • matplotlib defaults
    • figure size is 6.4x4.8"
    • dpi is 100
  • target style
  • plotProc should lazily evaluate the proc?
  • better integration with Arraymancer (a Concept that matches if .data and [] ?)
  • integrate chroma? (need to contribute blend?)
  • separate drawing layers for the plot and the background/names/etc so that lines aren't overwritten
  • can I use Arraymancer's tensor without blas? Would they work in js?
  • documentation
  • looks like matplotlib does some spline/approximation stuff to get that smooth
    • nope, checked the svg and it's plain lines => I need a better line algo
  • distinguish margin and padding
  • calculate max/min y value str len and adjust eg left margin to fit labels
  • svg backend: less strings, more nodes