Unicode version diff for ECMAScript

.gitignore

node_modules
package-lock.json

aliases.js

const aliasesToProperties = require('unicode-property-aliases');
const propertiesToAliases = new Map();
for (const [property, alias] of aliasesToProperties) {
	if (propertiesToAliases.has(property)) {
		propertiesToAliases.get(property).push(alias);
	} else {
		propertiesToAliases.set(property, [alias]);
	}
}

const aliasesToValuesPerProperty = require('unicode-property-value-aliases');
const valuesToAliasesPerProperty = new Map();
for (const [property, aliasesToValues] of aliasesToValuesPerProperty) {
	const valuesToAliases = new Map();
	for (const [alias, value] of aliasesToValues) {
		if (valuesToAliases.has(value)) {
			valuesToAliases.get(value).push(alias);
		} else {
			valuesToAliases.set(value, [alias]);
		}
	}
	valuesToAliasesPerProperty.set(property, valuesToAliases);
}

module.exports = {
	propertiesToAliases,
	valuesToAliasesPerProperty,
};

compare.js

const NEW_VERSION = '16.0.0';
const PREVIOUS_VERSION = '15.1.0';

const {
	propertiesToAliases,
	valuesToAliasesPerProperty
} = require('./aliases.js');

const formatter = new Intl.NumberFormat();
const formatNumber = (number) => formatter.format(number);

console.log(`Comparing Unicode v${ NEW_VERSION } to Unicode v${ PREVIOUS_VERSION }…`);

const diffProperties = (name) => {
	const a = require(`@unicode/unicode-${ PREVIOUS_VERSION }`)[name];
	const b = new Set(require(`@unicode/unicode-${ NEW_VERSION }`)[name]);
	const diff = new Set(a.filter(x => !b.has(x)));
	if (diff.size) {
		console.log(diff);
		console.log(`Note: didn’t check for new ${ name } values`); // TODO
	} else {
		const delta = b.size - a.length;
		console.log(`${ name }: ${ formatNumber(delta) } new values`);
		if (delta) {
			const props = [...b].filter(x => !a.includes(x));
			for (const prop of props) {
				const aliases = name === 'Binary_Property'
					? propertiesToAliases.get(prop)
					: valuesToAliasesPerProperty.get(name).get(prop);
				const aliasText = aliases ? ` (${ aliases.join(', ') })` : '';
				console.log(`- ${ prop }${ aliasText }`);
			}
		}
	}
};

diffProperties('Binary_Property');
diffProperties('General_Category');
diffProperties('Script');
diffProperties('Script_Extensions');
diffProperties('Sequence_Property');

const properties = [
	'Binary_Property/ID_Start',
	'Binary_Property/ID_Continue',
	'General_Category/Space_Separator',
];
for (const property of properties) {
	const a = require(`@unicode/unicode-${ PREVIOUS_VERSION }/${ property }/code-points.js`);
	const b = new Set(require(`@unicode/unicode-${ NEW_VERSION }/${ property }/code-points.js`));
	const diff = new Set(a.filter(x => !b.has(x)));
	console.log(`${ property }: ${ formatNumber(diff.size) } removals`);
	if (diff.size) {
		console.log(diff);
		console.log('Note: didn’t check for new code points'); // TODO
	} else {
		const delta = b.size - a.length;
		console.log(`${ property }: ${ formatNumber(delta) } new code points`);
	}
}

const caseFoldings = ['S', 'F'];
for (const form of caseFoldings) {
	const a = Object.keys(require(`@unicode/unicode-${ PREVIOUS_VERSION }/Case_Folding/${ form }/code-points.js`));
	const b = new Set(Object.keys(require(`@unicode/unicode-${ NEW_VERSION }/Case_Folding/${ form }/code-points.js`)));
	const removals = new Set(a.filter(x => !b.has(x)));
	console.log(`Case_Folding=${ form }: ${ formatNumber(removals.size) } removals`);
	if (removals.size) {
		console.log(removals);
		console.log('Note: didn’t check for new code points'); // TODO
	} else {
		const delta = b.size - a.length;
		console.log(`Case_Folding=${ form }: ${ formatNumber(delta) } new code points`);
	}
}

package.json

{
	"private": true,
	"dependencies": {
		"@unicode/unicode-15.1.0": "*",
		"@unicode/unicode-16.0.0": "*",
		"unicode-property-aliases": "^2.0.0",
		"unicode-property-value-aliases": "^3.8.0"
	}
}