unilight

1.0.0 • Public • Published

Unilight

Unilight is a unicode string highlighter. It uses unicode variations to highlight raw strings for almoast any programming language. The output can be used anywhere that accepts unicode chars. Unilight has no dependencies, e.g.

/**
  * 𝕋𝕙𝕚𝕤 𝕚𝕤 𝕒 𝕡𝕦𝕣𝕖 𝕤𝕥𝕣𝕚𝕟𝕘 𝕠𝕦𝕥𝕡𝕦𝕥, 𝕪𝕠𝕦 𝕔𝕒𝕟 𝕡𝕒𝕤𝕥𝕖 𝕚𝕥 𝕒𝕟𝕪𝕨𝕙𝕖𝕣𝕖
  * 𝕥𝕙𝕒𝕥 𝕤𝕦𝕡𝕡𝕠𝕣𝕥𝕤 𝕦𝕟𝕚𝕔𝕠𝕕𝕖 𝕒𝕟𝕕 𝕥𝕙𝕖 𝕙𝕚𝕘𝕙𝕝𝕚𝕘𝕙𝕥𝕤 𝕨𝕚𝕝𝕝 𝕓𝕖 𝕡𝕣𝕖𝕤𝕖𝕣𝕧𝕖𝕕
  **/
𝗳𝘂𝗻𝗰𝘁𝗶𝗼𝗻 example() {
  𝗰𝗼𝗻𝘀𝘁 a = "fizz";
  𝗰𝗼𝗻𝘀𝘁 b = "buzz";

  // 𝕋𝕙𝕚𝕤 𝕚𝕤 𝕒𝕟 𝕚𝕟𝕝𝕚𝕟𝕖 𝕔𝕠𝕞𝕞𝕖𝕟𝕥
  𝗿𝗲𝘁𝘂𝗿𝗻 `${a}${b}`;
}

Basic Usage

If you want to simply highlight a code string to use it anywhere, you can use the GUI tool available in this link: https://felippe-regazio.github.io/unilight/

If you want to highlight strings programatically, create your own highlighter, parse a source code string and modify its tokens, derive a tool from this library or anything else, keep reading this documentation.

Using Programatically

First install the unilight as a dependency

npm install unilight

Now require or import unilight (or just the highlight method). Then call the highlight method passing a collection of keywords to be highlighted. Styles for comments, strings and ponctuation will be automatically infered:

const { highlight } = require('unilight');

const code = `
  // This is an example

  function example() {
    const a = "fizz";
    const b = "buzz";
    return a+b;
  }
`;

const uTextHighlighted = highlight(code, [
  'function',
  'const',
  'return',
  'var',
  'let',
  // ...
]);

The output will be:

  // 𝕋𝕙𝕚𝕤 𝕚𝕤 𝕒𝕟 𝕖𝕩𝕒𝕞𝕡𝕝𝕖

  𝗳𝘂𝗻𝗰𝘁𝗶𝗼𝗻 example() {
    𝗰𝗼𝗻𝘀𝘁 a = "fizz";
    𝗰𝗼𝗻𝘀𝘁 b = "buzz";
    𝗿𝗲𝘁𝘂𝗿𝗻 a+b;
  }

Structure

Unilight is structured in 3 different modules: the parser, the formatter and unilight itself:

The Parser

The parser can parse almost any source code in any language. It receives a string input, a keyword list and some formatting callbackss. It will parse the source code string and run a callback to every diffent kind of token found. For each token types there will be a formatting callback which you can use to modify the token output. Unilight will identify 5 different token types:

  • Neutral (no formatting)
  • Keyword
  • Ponctuation
  • String
  • Comment

You can import the parser from unilight and use like this:

const { parse } = require('unilight');

const code = `
  // THIS IS AN EXAMPLE

  function example() {
    const a = "fizz";
    const b = "buzz";
    return a+b;
  }
`;

parse(code, {
  keywords: [ 'function', 'const', 'return' ],

  formatKeyword(t) {
    // this callback will be called for all the keywords
    // defined in { keywords }. you can return a modified
    // version of the string found to modify the output.
    // this is the general rule for all the other callbacks.
    // for example, lets output all keywords in uppercase:

    return t.toUpperCase();
  },

  formatPonctuation(t) {
    // no formatting
    return t;
  },

  formatString(t) {
    // no formatting
    return t;
  },

  formatComment(t) {
    // comments will be lowercased
    return t.toLowerCase();
  }
});

The output would be:

// this is an example

FUNCTION example() {
  CONST a = "fizz";
  CONST b = "buzz";
  RETURN a+b;
}

The parser was orinally developed by Helios (ASVD) for microlight lib and slightly modified to be used on this library.

The Formatter

The formatter is a module that can convert a normal unicode string to some variation like bold, italic, outlined and others. It was originally developed by David Konrad as toUnicodeVariant lib. Unilight uses a modified version on this library. You can check the original source of the formatter and its usage here: https://github.com/davidkonrad/toUnicodeVariant

To use toUnicodeVariant on unilight you must do:

const { format } = require('unilight');

const result = format('This is an example', 'd');

console.log(result);

The output will be:

𝕋𝕙𝕚𝕤 𝕚𝕤 𝕒𝕟 𝕖𝕩𝕒𝕞𝕡𝕝𝕖

On the code above, d tells the formatter which variation to use while formating the string. The available variations are:

Variant Alias Description Example
monospace m Monospace 𝚖𝚘𝚗𝚘𝚜𝚙𝚊𝚌𝚎
bold b Bold text 𝐛𝐨𝐥𝐝
italic i Italic text 𝑖𝑡𝑎𝑙𝑖𝑐
bold italic bi bold+italic text 𝒃𝒐𝒍𝒅 𝒊𝒕𝒂𝒍𝒊𝒄
script c Handwriting style 𝓈𝒸𝓇𝒾𝓅𝓉
bold script bc Bolder handwriting 𝓫𝓸𝓵𝓭 𝓼𝓬𝓻𝓲𝓹𝓽
gothic g Gothic (fraktur) 𝔤𝔬𝔱𝔥𝔦𝔠
gothic bold gb Gothic in bold 𝖌𝖔𝖙𝖍𝖎𝖈 𝖇𝖔𝖑𝖉
doublestruck d Outlined text 𝕕𝕠𝕦𝕓𝕝𝕖𝕤𝕥𝕣𝕦𝕔𝕜
𝗌𝖺𝗇𝗌 s Sans-serif style 𝗌𝖺𝗇𝗌
bold 𝗌𝖺𝗇𝗌 bs Bold sans-serif 𝗯𝗼𝗹𝗱 𝘀𝗮𝗻𝘀
italic 𝗌𝖺𝗇𝗌 is Italic sans-serif 𝘪𝘵𝘢𝘭𝘪𝘤 𝘴𝘢𝘯𝘴
bold italic sans bis Bold italic sans-serif 𝙗𝙤𝙡𝙙 𝙞𝙩𝙖𝙡𝙞𝙘 𝙨𝙖𝙣𝙨
circled o Letters within circles ⓒⓘⓡⓒⓛⓔⓓ
circled negative on -- negative 🅒🅘🅡🅒🅛🅔🅓
squared q Letters within squares 🅂🅀🅄🄰🅁🄴🄳
squared negative qn -- negative 🆂🆀🆄🅰🆁🅴🅳
paranthesis p Letters within paranthesis ⒫⒜⒭⒠⒩⒯⒣⒠⒮⒤
fullwidth w Wider monospace font fullwidth

The Highlighter

The unilight highlighter is just a pre-configured parser and formatter that you can use to highlight raw strings. You can use the highlighter by import the highlight function from unilight.

const { highlight } = require('unilight');

const output = highlight(code, [ 'function', 'const', 'var', /** goes on... **/ ]);

THe unicode formatting for keywords and comments will be automatically applied. You can also create your own highlighter.

Creating your own Highlighter

Use the snippet below to create your own Highlighter. You dont have to rely on unicode variations, you can modify the tokens as you prefer. This would be a Rich Code Highlighter to be used on an HTML page for example:

const { parse, format } = require('unilight');

function customHighlighter(text) {
  return parse(text, {
    keywords: [ 'function', 'return' ],

    formatKeyword(t) {
      return `<span class="highlight-keyword">${t}</span>`;
    },

    formatPonctuation(t) {
      return `<span class="highlight-ponctuation">${t}</span>`;
    },

    formatString(t) {
      return `<span class="highlight-string">${t}</span>`;
    },

    formatComment(t) {
      return `<span class="highlight-comment">${t}</span>`;
    }
  });
}

// usage

const output = customHighlighter(`
  function example() {
    console.log("This is just an example");

    return true;
  }
`);

The code above will find tokens and sorround them with a span identified by a custom class depending on the token type. You can use this classes style the output on an HTML page.

Developing

Unilight is a colletion of modules that builds to an UMD under the global name unilight using parcel bundler.

Dev dependencies

Before start, install dev dependencies

npm install

Build

npm run build

Test

npm run test

Running GUI editor locally

npm run start

The keywords.js file

This file is used by the GUI Editor (index.html) to serve different collections of keywords for different programming languages.

Adding new languagues to the GUI Editor

Just open the keywords.js file on the project root, add a new property with the language name you want to add, and an array of strings containing each keyword you want to be highlighted.

Cavets

This is a toy project and it was made for aesthetics only. Some of the cavets of a unicode highlighter are:

  1. Environments that does not support unicode can show a broken text
  2. Since unicode variations are basically hidden chars, the String may be bigger then it looks
  3. Interpreters Wont run the code because of the keywords will not be recognized
  4. The unicode caracters may vary from one system to another

Credits

Unilight was written by Felippe Regazio.

Special thanks to David Konrad, which created the module Unicode Variations that can modify a unicode string and apply new variations. This is used by Unilight to highlight the string.

Special thanks to Helios (ASVD) that created Microlight Highlighter. His parsing techiniques where a strong inspiration and direct reference to Unilight parser, which uses part of his code.

Package Sidebar

Install

npm i unilight

Weekly Downloads

1

Version

1.0.0

License

MIT

Unpacked Size

60.7 kB

Total Files

11

Last publish

Collaborators

  • felipperegazio