build(deps): bump undici from 6.21.0 to 7.2.0 (#4309)

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Felix <[email protected]>

package.json

@@ -124,7 +124,7 @@
     "parse5": "^7.2.1",
     "parse5-htmlparser2-tree-adapter": "^7.1.0",
     "parse5-parser-stream": "^7.1.2",
-    "undici": "^6.21.0",
+    "undici": "^7.2.0",
     "whatwg-mimetype": "^4.0.0"
   },
   "devDependencies": {

src/index.spec.ts

@@ -1,7 +1,7 @@
 import { describe, it, expect, afterEach } from 'vitest';
 import * as cheerio from './index.js';
 import { Writable } from 'node:stream';
-import { createServer, type Server } from 'node:http';
+import { createServer, type Server, type RequestListener } from 'node:http';
 
 function noop() {
   // Ignore
@@ -119,12 +119,13 @@ describe('fromURL', () => {
   function createTestServer(
     contentType: string,
     body: string | Buffer,
+    handler: RequestListener = (_req, res) => {
+      res.writeHead(200, { 'Content-Type': contentType });
+      res.end(body);
+    },
   ): Promise<number> {
     return new Promise((resolve, reject) => {
-      server = createServer((_req, res) => {
-        res.writeHead(200, { 'Content-Type': contentType });
-        res.end(body);
-      });
+      server = createServer(handler);
 
       server.listen(0, () => {
         const address = server?.address();
@@ -177,4 +178,22 @@ describe('fromURL', () => {
 
     expect($.html()).toBe(TEST_HTML);
   });
+
+  it('should throw on non-HTML/XML Content-Type', async () => {
+    const port = await createTestServer('text/plain', TEST_HTML);
+    await expect(cheerio.fromURL(`http://localhost:${port}`)).rejects.toThrow(
+      'The content-type "text/plain" is neither HTML nor XML.',
+    );
+  });
+
+  it('should throw on non-2xx responses', async () => {
+    const port = await createTestServer('text/html', TEST_HTML, (_, res) => {
+      res.writeHead(500);
+      res.end();
+    });
+
+    await expect(cheerio.fromURL(`http://localhost:${port}`)).rejects.toThrow(
+      'Response Error',
+    );
+  });
 });

src/index.ts

@@ -186,8 +186,6 @@ const defaultRequestOptions: UndiciStreamOptions = {
   method: 'GET',
   // Allow redirects by default
   maxRedirections: 5,
-  // NOTE: `throwOnError` currently doesn't work https://github.com/nodejs/undici/issues/1753
-  throwOnError: true,
   // Set an Accept header
   headers: {
     accept: 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
@@ -217,26 +215,32 @@ export async function fromURL(
   options: CheerioRequestOptions = {},
 ): Promise<CheerioAPI> {
   const {
-    requestOptions = defaultRequestOptions,
+    requestOptions = {
+      ...defaultRequestOptions,
+      dispatcher: undici
+        .getGlobalDispatcher()
+        .compose(undici.interceptors.responseError()),
+    },
     encoding = {},
     ...cheerioOptions
   } = options;
-  let undiciStream: Promise<undici.Dispatcher.StreamData> | undefined;
+  let undiciStream: Promise<undici.Dispatcher.StreamData<unknown>> | undefined;
 
   // Add headers if none were supplied.
   requestOptions.headers ??= defaultRequestOptions.headers;
 
   const promise = new Promise<CheerioAPI>((resolve, reject) => {
     undiciStream = undici.stream(url, requestOptions, (res) => {
       const contentTypeHeader = res.headers['content-type'] ?? 'text/html';
-      const contentType = Array.isArray(contentTypeHeader)
-        ? contentTypeHeader[0]
-        : contentTypeHeader;
-      const mimeType = new MIMEType(contentType);
+      const mimeType = new MIMEType(
+        Array.isArray(contentTypeHeader)
+          ? contentTypeHeader[0]
+          : contentTypeHeader,
+      );
 
       if (!mimeType.isHTML() && !mimeType.isXML()) {
         throw new RangeError(
-          `The content-type "${contentType}" is neither HTML nor XML.`,
+          `The content-type "${mimeType.essence}" is neither HTML nor XML.`,
         );
       }