From 35d1d85978ef840c0bba05f5c0201c0e2ac1d17e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?M=C3=B3ricz=20Gerg=C5=91?= <mo.geryy@gmail.com>
Date: Tue, 7 Jan 2025 09:29:58 +0100
Subject: [PATCH] fix(crawler): also take the hostname of the base url when
 determining isInternalLink

---
 apps/api/src/scraper/WebScraper/crawler.ts | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/apps/api/src/scraper/WebScraper/crawler.ts b/apps/api/src/scraper/WebScraper/crawler.ts
index 9c49cf9b..dd949238 100644
--- a/apps/api/src/scraper/WebScraper/crawler.ts
+++ b/apps/api/src/scraper/WebScraper/crawler.ts
@@ -397,8 +397,7 @@ export class WebCrawler {
 
   private isInternalLink(link: string): boolean {
     const urlObj = new URL(link, this.baseUrl);
-    const baseDomain = this.baseUrl
-      .replace(/^https?:\/\//, "")
+    const baseDomain = new URL(this.baseUrl).hostname
       .replace(/^www\./, "")
       .trim();
     const linkDomain = urlObj.hostname.replace(/^www\./, "").trim();