From d28acfdb20003c7d17a9495f9e3f1b81973425e3 Mon Sep 17 00:00:00 2001
From: Pieter Vander Vennet <pietervdvn@posteo.net>
Date: Sat, 24 Aug 2024 01:50:34 +0200
Subject: [PATCH] Improve URL-validation for blocked and discouraged sites

---
 langs/en.json                                 |  5 ++--
 .../InputElement/Validators/UrlValidator.ts   | 28 +++++++++++++++----
 2 files changed, 26 insertions(+), 7 deletions(-)

diff --git a/langs/en.json b/langs/en.json
index f894c64115..23b50736ee 100644
--- a/langs/en.json
+++ b/langs/en.json
@@ -847,9 +847,10 @@
         },
         "tooLong": "The text is too long, at most 255 characters are allowed. You have {count} characters now.",
         "url": {
-            "aggregator": "{host} is a third-party aggregator website. If possible, search the official website.",
+            "aggregator": "{host} is a third-party website. If possible, search the official website.",
             "description": "link to a website",
-            "feedback": "This is not a valid web address"
+            "feedback": "This is not a valid web address",
+            "spamSite": "{host} is considered a low-quality website. Using this website is not allowed."
         },
         "wikidata": {
             "description": "A Wikidata identifier",
diff --git a/src/UI/InputElement/Validators/UrlValidator.ts b/src/UI/InputElement/Validators/UrlValidator.ts
index 148562f42e..6854c94153 100644
--- a/src/UI/InputElement/Validators/UrlValidator.ts
+++ b/src/UI/InputElement/Validators/UrlValidator.ts
@@ -5,14 +5,19 @@ import Translations from "../../i18n/Translations"
 export default class UrlValidator extends Validator {
     private readonly _forceHttps: boolean
 
-    private static readonly aggregatorWebsites = new Set<string>([
+    private static readonly spamWebsites = new Set<string>([
         "booking.com",
         "hotel-details-guide.com",
         "tripingguide.com",
         "tripadvisor.com",
         "tripadvisor.co.uk",
         "tripadvisor.com.au",
-        "katestravelexperience.eu"
+        "katestravelexperience.eu",
+        "hoteldetails.eu"
+    ])
+
+    private static readonly discouragedWebsites = new Set<string>([
+        "facebook.com"
     ])
 
     constructor(name?: string, explanation?: string, forceHttps?: boolean) {
@@ -89,15 +94,27 @@ export default class UrlValidator extends Validator {
      *
      */
     getFeedback(s: string, getCountry?: () => string): Translation | undefined {
+        if (
+            !s.startsWith("http://") &&
+            !s.startsWith("https://") &&
+            !s.startsWith("http:")
+        ) {
+            s = "https://" + s
+        }
         try{
             const url = new URL(s)
             let host = url.host.toLowerCase()
             if (host.startsWith("www.")) {
                 host = host.slice(4)
             }
-            if (UrlValidator.aggregatorWebsites.has(host)) {
+            if (UrlValidator.spamWebsites.has(host)) {
+                return Translations.t.validation.url.spamSite.Subs({ host })
+            }
+            if (UrlValidator.discouragedWebsites.has(host)) {
                 return Translations.t.validation.url.aggregator.Subs({ host })
             }
+
+
         }catch (e) {
             // pass
         }
@@ -111,6 +128,7 @@ export default class UrlValidator extends Validator {
     }
 
     isValid(str: string): boolean {
+
         try {
             if (
                 !str.startsWith("http://") &&
@@ -120,16 +138,16 @@ export default class UrlValidator extends Validator {
                 str = "https://" + str
             }
             const url = new URL(str)
-            const dotIndex = url.host.indexOf(".")
 
             let host = url.host.toLowerCase()
             if (host.startsWith("www.")) {
                 host = host.slice(4)
             }
-            if (UrlValidator.aggregatorWebsites.has(host)) {
+            if (UrlValidator.spamWebsites.has(host)) {
                 return false
             }
 
+            const dotIndex = url.host.indexOf(".")
             return dotIndex > 0 && url.host[url.host.length - 1] !== "."
         } catch (e) {
             return false