chbrown / unidata

Javascript interface to the Unicode Character Database

unidata

JavaScript interface to the Unicode Character Database. The package version is intended to match the Unicode version.

npm install unidata --save

API

Using ES6 syntax:

import {getBlocks, getCharacters} from 'unidata'

There are a few other exports (and a couple useful interfaces, if you're using TypeScript), but these are the main methods.

They both simply call require() to load the preprocessed Unicode data from a JSON file.

const blocks = getBlocks()
const characters = getCharacters()

These are both just arrays.

• blocks is an array of Block
• characters is an array of Character

Blocks are simple; every item in blocks has these three fields:

> blocks.length
308
> blocks.slice(0, 5)
[ { startCode:   0, endCode: 127, blockName: 'Basic Latin'        },
  { startCode: 128, endCode: 255, blockName: 'Latin-1 Supplement' },
  { startCode: 256, endCode: 383, blockName: 'Latin Extended-A'   },
  { startCode: 384, endCode: 591, blockName: 'Latin Extended-B'   },
  { startCode: 592, endCode: 687, blockName: 'IPA Extensions'     } ]

Characters are richer, but the representation is parsimonious: if a value is not available or not applicable for a given character, that key will be omitted.

> characters.length
33797
> characters.slice(32, 40)
[ { code: 32, name: 'SPACE',            cat: 'Zs', bidi: 'WS' },
  { code: 33, name: 'EXCLAMATION MARK', cat: 'Po', bidi: 'ON' },
  { code: 34, name: 'QUOTATION MARK',   cat: 'Po', bidi: 'ON' },
  { code: 35, name: 'NUMBER SIGN',      cat: 'Po', bidi: 'ET' },
  { code: 36, name: 'DOLLAR SIGN',      cat: 'Sc', bidi: 'ET' },
  { code: 37, name: 'PERCENT SIGN',     cat: 'Po', bidi: 'ET' },
  { code: 38, name: 'AMPERSAND',        cat: 'Po', bidi: 'ON' },
  { code: 39, name: 'APOSTROPHE',       cat: 'Po', bidi: 'ON', oldName: 'APOSTROPHE-QUOTE' } ]

The first three fields, code, name, and cat, are always present. The other ten are optional. (For details on the optional fields, and what values to assume when they are omitted, see the comments on the Block interface.)

License

Copyright 2015–2020 Christopher Brown. MIT Licensed.

unidata's Issues

Load initial set of characters faster

Split up UnicodeData.js into two files, one for the BMP and one for the astral planes. Add something like unidata.getCharacters but which only require()'s the BMP characters.