Add removePath and transformPath option

Closes #188

Author: sindresorhus

index.d.ts

@@ -278,6 +278,45 @@ export type Options = {
 	```
 	*/
 	readonly sortQueryParameters?: boolean;
+
+	/**
+	Removes the entire URL path, leaving only the domain.
+
+	@default false
+
+	@example
+	```
+	normalizeUrl('https://example.com/path/to/page', {
+		removePath: true
+	});
+	//=> 'https://example.com'
+	```
+	*/
+	readonly removePath?: boolean;
+
+	/**
+	Custom function to transform the URL path components.
+
+	The function receives an array of non-empty path components and should return a modified array.
+
+	@default false
+
+	@example
+	```
+	// Keep only the first path component
+	normalizeUrl('https://example.com/api/v1/users', {
+		transformPath: (pathComponents) => pathComponents.slice(0, 1)
+	});
+	//=> 'https://example.com/api'
+
+	// Remove specific components
+	normalizeUrl('https://example.com/admin/users', {
+		transformPath: (pathComponents) => pathComponents.filter(c => c !== 'admin')
+	});
+	//=> 'https://example.com/users'
+	```
+	*/
+	readonly transformPath?: (pathComponents: string[]) => string[];
 };
 
 /**

index.js

@@ -89,6 +89,8 @@ export default function normalizeUrl(urlString, options) {
 		removeDirectoryIndex: false,
 		removeExplicitPort: false,
 		sortQueryParameters: true,
+		removePath: false,
+		transformPath: false,
 		...options,
 	};
 
@@ -200,6 +202,18 @@ export default function normalizeUrl(urlString, options) {
 		}
 	}
 
+	// Remove path
+	if (options.removePath) {
+		urlObject.pathname = '/';
+	}
+
+	// Transform path components
+	if (options.transformPath && typeof options.transformPath === 'function') {
+		const pathComponents = urlObject.pathname.split('/').filter(Boolean);
+		const newComponents = options.transformPath(pathComponents);
+		urlObject.pathname = newComponents?.length > 0 ? `/${newComponents.join('/')}` : '/';
+	}
+
 	if (urlObject.hostname) {
 		// Remove trailing dot
 		urlObject.hostname = urlObject.hostname.replace(/\.$/, '');

readme.md

@@ -307,6 +307,41 @@ normalizeUrl('www.sindresorhus.com?b=two&a=one&c=three', {
 //=> 'http://sindresorhus.com/?b=two&a=one&c=three'
 ```
 
+##### removePath
+
+Type: `boolean`\
+Default: `false`
+
+Removes the entire URL path, leaving only the domain.
+
+```js
+normalizeUrl('https://example.com/path/to/page', {
+	removePath: true
+});
+//=> 'https://example.com'
+```
+
+##### transformPath
+
+Type: `Function`\
+Default: `false`
+
+Custom function to transform the URL path components. The function receives an array of non-empty path components and should return a modified array.
+
+```js
+// Keep only the first path component
+normalizeUrl('https://example.com/api/v1/users', {
+	transformPath: (pathComponents) => pathComponents.slice(0, 1)
+});
+//=> 'https://example.com/api'
+
+// Remove specific components
+normalizeUrl('https://example.com/admin/users', {
+	transformPath: (pathComponents) => pathComponents.filter(c => c !== 'admin')
+});
+//=> 'https://example.com/users'
+```
+
 ## Related
 
 - [compare-urls](https://github.com/sindresorhus/compare-urls) - Compare URLs by first normalizing them

test.js

@@ -439,6 +439,53 @@ test('encoded backslashes do not get decoded', t => {
 	t.is(normalizeUrl('https://foo.com/something\\else/great'), 'https://foo.com/something/else/great');
 });
 
+test('removePath option', t => {
+	// Boolean: Remove entire path
+	t.is(normalizeUrl('https://example.com/path/to/page', {removePath: true}), 'https://example.com');
+	t.is(normalizeUrl('https://example.com/path/to/page?query=1', {removePath: true}), 'https://example.com/?query=1');
+	t.is(normalizeUrl('https://example.com/path/to/page#hash', {removePath: true}), 'https://example.com/#hash');
+	t.is(normalizeUrl('https://example.com/', {removePath: true}), 'https://example.com');
+	t.is(normalizeUrl('https://example.com', {removePath: true}), 'https://example.com');
+	
+	// With other options
+	t.is(normalizeUrl('https://example.com/path/', {removePath: true, removeTrailingSlash: true}), 'https://example.com');
+	t.is(normalizeUrl('https://www.example.com/path', {removePath: true, stripWWW: true}), 'https://example.com');
+});
+
+test('transformPath option', t => {
+	// Function: Keep only first path component
+	const keepFirst = pathComponents => pathComponents.slice(0, 1);
+	t.is(normalizeUrl('https://example.com/api/v1/users', {transformPath: keepFirst}), 'https://example.com/api');
+	t.is(normalizeUrl('https://example.com/path/to/page', {transformPath: keepFirst}), 'https://example.com/path');
+	t.is(normalizeUrl('https://example.com/', {transformPath: keepFirst}), 'https://example.com');
+	
+	// Function: Remove specific component
+	const removeAdmin = pathComponents => pathComponents.filter(c => c !== 'admin');
+	t.is(normalizeUrl('https://example.com/admin/users', {transformPath: removeAdmin}), 'https://example.com/users');
+	t.is(normalizeUrl('https://example.com/path/admin/page', {transformPath: removeAdmin}), 'https://example.com/path/page');
+	
+	// Function: Custom logic
+	const customLogic = pathComponents => {
+		if (pathComponents[0] === 'api') {
+			return pathComponents.slice(0, 1); // Keep /api only
+		}
+		return []; // Remove everything else
+	};
+	t.is(normalizeUrl('https://example.com/api/v1/users', {transformPath: customLogic}), 'https://example.com/api');
+	t.is(normalizeUrl('https://example.com/other/path', {transformPath: customLogic}), 'https://example.com');
+	
+	// Edge cases
+	t.is(normalizeUrl('https://example.com/path', {transformPath: () => []}), 'https://example.com');
+	t.is(normalizeUrl('https://example.com/path', {transformPath: () => null}), 'https://example.com');
+	t.is(normalizeUrl('https://example.com/path', {transformPath: () => undefined}), 'https://example.com');
+	
+	// Combining with removePath (removePath should take precedence)
+	t.is(normalizeUrl('https://example.com/path/to/page', {
+		removePath: true,
+		transformPath: pathComponents => pathComponents.slice(0, 2)
+	}), 'https://example.com');
+});
+
 test('path-like query strings without equals signs are preserved', t => {
 	// Issue #193 - Path-like query strings should not get '=' appended
 	t.is(normalizeUrl('https://example.com/index.php?/Some/Route/To/Path/12345'), 'https://example.com/index.php?/Some/Route/To/Path/12345');