@crawlee/http 4.0.0-beta.176 → 4.0.0-beta.178

package/internals/dom-crawler.js

@@ -86,7 +86,11 @@ export class DOMCrawler extends HttpCrawler {
                 });
             },
             async parseWithCheerio(selector, _timeoutMs = 5_000) {
-                const $ = (await parser.toCheerio?.(context)) ?? (await import('cheerio')).load(context.body);
+                // Import full cheerio (not cheerio/slim) to be browser-compliant for DOM Crawlers (not CheerioCrawler).
+                const $ = (await parser.toCheerio?.(context)) ??
+                    (await import('cheerio')).load(context.body, {
+                        xmlMode: context.contentType.type.includes('xml'),
+                    });
                 if (selector && $(selector).get().length === 0) {
                     throw new Error(`Selector '${selector}' not found.`);
                 }

package/internals/http-crawler.js

@@ -286,16 +286,18 @@ export class HttpCrawler extends BasicCrawler {
         tryCancel();
         const response = parsed.response;
         const contentType = parsed.contentType;
+        const loadBody = async () => {
+            const { load } = await import('cheerio/slim');
+            return load(parsed.body.toString(), { xmlMode: contentType.type.includes('xml') });
+        };
         const waitForSelector = async (selector, _timeoutMs) => {
-            const cheerio = await import('cheerio');
-            const $ = cheerio.load(parsed.body.toString());
+            const $ = await loadBody();
             if ($(selector).get().length === 0) {
                 throw new Error(`Selector '${selector}' not found.`);
             }
         };
         const parseWithCheerio = async (selector, timeoutMs) => {
-            const cheerio = await import('cheerio');
-            const $ = cheerio.load(parsed.body.toString());
+            const $ = await loadBody();
             if (selector) {
                 await crawlingContext.waitForSelector(selector, timeoutMs);
             }

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@crawlee/http",
-    "version": "4.0.0-beta.176",
+    "version": "4.0.0-beta.178",
     "description": "The scalable web crawling and scraping library for JavaScript/Node.js. Enables development of data extraction and web automation jobs (not only) with headless Chrome and Puppeteer.",
     "engines": {
         "node": ">=22.0.0"
@@ -49,11 +49,11 @@
     "dependencies": {
         "@apify/timeout": "^1.0.1",
         "@apify/utilities": "^3.0.1",
-        "@crawlee/basic": "4.0.0-beta.176",
-        "@crawlee/core": "4.0.0-beta.176",
-        "@crawlee/http-client": "4.0.0-beta.176",
-        "@crawlee/types": "4.0.0-beta.176",
-        "@crawlee/utils": "4.0.0-beta.176",
+        "@crawlee/basic": "4.0.0-beta.178",
+        "@crawlee/core": "4.0.0-beta.178",
+        "@crawlee/http-client": "4.0.0-beta.178",
+        "@crawlee/types": "4.0.0-beta.178",
+        "@crawlee/utils": "4.0.0-beta.178",
         "@types/content-type": "^1.1.8",
         "cheerio": "^1.0.0",
         "content-type": "^1.0.5",
@@ -70,5 +70,5 @@
             }
         }
     },
-    "gitHead": "3e9e27d524dff8fe72aaf337e3d24fc2232eb241"
+    "gitHead": "7beea593ac56d23b678466bc28390d527dbc3d8b"
 }