string-mismatch

All mismatch between two strings

Usage no npm install needed!

<script type="module">
  import stringMismatch from 'https://cdn.skypack.dev/string-mismatch';
</script>

README

string-mismatch

Build Status codecov.io contributions welcome

Algorithms for compare strings. Currently the list of available algorithms are:

Greedy Levenshtein Dice Coefficient
Complexity O(n*k) (k precision) O(n^2) O(nlog n)
Good Fast algorithm Always the optimal solution Is based in probabilities and is a really fast algorithm
Bad The solution is not the optimal Complexity is O(n^2) Impossible to see the differences between the strings
Use n^2 memory
Methods difference difference distance
distance distance
Operations for transform the string insertion insertion not apply
deletion deletion
substitution
Class name Greedy Levenshtein DiceCoefficient

Why use string-mismatch:

  • Ease to install and start using it
  • Modular library (use only what you want to use).
  • Support for browser and node applications.
  • Compatible with es5
  • Not external dependencies.
  • Completely documented.
  • Coverage over 95%.

Install

npm install --save string-mismatch

Getting started

### Nodejs application example

How to use the library and see the differences between two strings:

const sm = require("string-mismatch");
const greedyInstance = new sm.Greedy();

var start = 'This is a test for see how work the library',
    end   = 'This is a test for know how work the new library';

console.log(greedyInstance.differences(start, end));

The result is an object array with the mismatch result. Each object with the next structure:

{
  type: string, // type of sub-string:
                //   'sub' -> substitution
                //   'ins' -> insertion
                //   'del' -> deletion
                //   'eql' -> equal
  value: string // value of the current sub-string
}

The resulting string can be concatenated like the next example:

const sm = require("string-mismatch");
const greedyInstance = new sm.Greedy();

var start = 'This is a test for see how work the library',
    end   = 'This is a test for know how work the new library';

function showResult(diffs) {
    return diffs.reduce(function (text, value) {
        switch (value.type) {
            case 'del':
                return text + '(-' + value.value + ')';
            case 'ins':
                return text + '(+' + value.value + ')';
            case 'sub':
                return text + '(-+' + value.value + ')';
            case 'eql':
                return text + value.value;
        }
    }, '');
}

console.log(showResult(greedyInstance.differences(start, end)));
/*
result:
This is a test for (-see)(+know) how work the (+new )library
*/

This code can be tested in the project's examples. To run the examples use the next command:

npm start

Web application example

Import the library

<!--Greedy algorithm-->
<script src="lib/greedy.min.js" type="application/javascript"></script>
<!--Levenshtein algorithm-->
<script src="lib/levenshtein.min.js" type="application/javascript"></script>

Example with greedy algorithm:

<script type="application/javascript">
    var start = 'This is a test for see how work the library';
    var end = 'This is a test for know how work the new library';
    var alg = new Greedy(options);
    var diffs = alg.differences(start, end);
    console.log(diffs);
</script>

Example with the levenshtein algorithm:

<script type="application/javascript">
    var start = 'This is a test for see how work the library';
    var end = 'This is a test for know how work the new library';
    var alg = new Levenshtein(options);
    var diffs = alg.differences(start, end);
    console.log(diffs);
</script>

Testing code

npm test

Built With

  • webpack - For build the project
  • npm - Dependency Management
  • jest - Jest framework for test

Contributing

All contributions are welcome.

Versioning

We use SemVer for versioning. For the versions available, see the tags on this repository.

Authors

  • Guillermo González - Initial work - wil92

License

This project is licensed under the MIT License - see the LICENSE.md file for details