You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

129 lines
4.0 KiB

# frac
Rational approximation to a floating point number with bounded denominator.
Uses the [Mediant Method](https://en.wikipedia.org/wiki/Mediant_method).
This module also provides an implementation of the continued fraction method as
described by Aberth in "A method for exact computation with rational numbers".
The algorithm is used in <a href="http://sheetjs.com">SheetJS Libraries</a> to
replicate fraction formats.
## Installation
### JS
With [`npm`](https://www.npmjs.org/package/frac):
```bash
$ npm install frac
```
In the browser:
```html
<script src="frac.js"></script>
```
The script will manipulate `module.exports` if available . This is not always
desirable. To prevent the behavior, define `DO_NOT_EXPORT_FRAC`
### Python
From [`PyPI`](https://pypi.python.org/pypi/frac):
```bash
$ pip install frac
```
## Usage
In all cases, the relevant function takes 3 arguments:
- `x` the number we wish to approximate
- `D` the maximum denominator
- `mixed` if true, return a mixed fraction; if false, improper
The return value is an array of the form `[quot, num, den]` where `quot==0`
for improper fractions. `quot <= x` for mixed fractions, which may lead to some
unexpected results when rendering negative numbers.
### JS
The exported `frac` function implements the Mediant method.
`frac.cont` implements the Aberth algorithm
For example:
```js
> // var frac = require('frac'); // uncomment this line if in node
> frac(1.3, 9); // [ 0, 9, 7 ] // 1.3 ~ 9/7
> frac(1.3, 9, true); // [ 1, 2, 7 ] // 1.3 ~ 1 + 2/7
> frac(-1.3, 9); // [ 0, -9, 7 ] // -1.3 ~ -9/7
> frac(-1.3, 9, true); // [ -2, 5, 7 ] // -1.3 ~ -2 + 5/7
> frac.cont(1.3, 9); // [ 0, 4, 3 ] // 1.3 ~ 4/3
> frac.cont(1.3, 9, true); // [ 1, 1, 3 ] // 1.3 ~ 1 + 1/3
> frac.cont(-1.3, 9); // [ 0, -4, 3 ] // -1.3 ~ -4/3
> frac.cont(-1.3, 9, true); // [ -2, 2, 3 ] // -1.3 ~ -2 + 2/3
```
### Python
`frac.med` implements Mediant method.
`frac.cont` implements Aberth algorithm.
For example:
```py
>>> import frac
>>> frac.med(1.3, 9) ## [ 0, 9, 7 ] ## 1.3 ~ 9/7
>>> frac.med(1.3, 9, True) ## [ 1, 2, 7 ] ## 1.3 ~ 1 + 2/7
>>> frac.med(-1.3, 9) ## [ 0, -9, 7 ] ## -1.3 ~ -9/7
>>> frac.med(-1.3, 9, True) ## [ -2, 5, 7 ] ## -1.3 ~ -2 + 5/7
>>> frac.cont(1.3, 9) ## [ 0, 4, 3 ] ## 1.3 ~ 4/3
>>> frac.cont(1.3, 9, True) ## [ 1, 1, 3 ] ## 1.3 ~ 1 + 1/3
>>> frac.cont(-1.3, 9) ## [ 0, -4, 3 ] ## -1.3 ~ -4/3
>>> frac.cont(-1.3, 9, True) ## [ -2, 2, 3 ] ## -1.3 ~ -2 + 2/3
```
## Testing
The test TSV baselines in the `test_files` directory have four columns:
- Column A contains the raw values
- Column B format "Up to one digit (1/4)" (`denominator = 9`)
- Column C format "Up to two digits (21/25)" (`denominator = 99`)
- Column D format "Up to three digits (312/943)" (`denominator = 999`)
`make test` will run the node-based tests.
`make pytest` will run the python tests against the system Python version.
`make pypytest` will run the python tests against `pypy` if installed
## License
Please consult the attached LICENSE file for details. All rights not explicitly
granted by the Apache 2.0 License are reserved by the Original Author.
## Badges
[![Build Status](https://saucelabs.com/browser-matrix/frac.svg)](https://saucelabs.com/u/frac)
[![Build Status](https://travis-ci.org/SheetJS/frac.svg?branch=master)](https://travis-ci.org/SheetJS/frac)
[![Coverage Status](http://img.shields.io/coveralls/SheetJS/frac/master.svg)](https://coveralls.io/r/SheetJS/frac?branch=master)
[![NPM Downloads](https://img.shields.io/npm/dt/frac.svg)](https://npmjs.org/package/frac)
[![Dependencies Status](https://david-dm.org/sheetjs/frac/status.svg)](https://david-dm.org/sheetjs/frac)
[![ghit.me](https://ghit.me/badge.svg?repo=sheetjs/js-xlsx)](https://ghit.me/repo/sheetjs/js-xlsx)
[![Analytics](https://ga-beacon.appspot.com/UA-36810333-1/SheetJS/frac?pixel)](https://github.com/SheetJS/frac)