From 177b04dcfc064933a269bc05be64631ac90954d7 Mon Sep 17 00:00:00 2001
From: Shannon Booth <shannon@serenityos.org>
Date: Tue, 25 Jul 2023 19:43:00 +1200
Subject: [PATCH] AK: Fix url host parsing check for 'ends in a number'

I misunderstood the spec step for checking whether the host 'ends with a
number'. We can't simply check for it if ends with a number, this check
is actually an algorithm which is required to avoid detecting hosts that
end with a number from an IPv4 host.

Implement this missing step, and add a test to cover this.
---
 AK/URLParser.cpp     | 33 ++++++++++++++++++++++++++++++++-
 Tests/AK/TestURL.cpp | 10 ++++++++++
 2 files changed, 42 insertions(+), 1 deletion(-)

diff --git a/AK/URLParser.cpp b/AK/URLParser.cpp
index 4204cbcd1f..9e818aaaae 100644
--- a/AK/URLParser.cpp
+++ b/AK/URLParser.cpp
@@ -518,6 +518,37 @@ static Optional<Array<u16, 8>> parse_ipv6_address(StringView input)
     return address;
 }
 
+// https://url.spec.whatwg.org/#ends-in-a-number-checker
+static bool ends_in_a_number_checker(StringView input)
+{
+    // 1. Let parts be the result of strictly splitting input on U+002E (.).
+    auto parts = input.split_view("."sv, SplitBehavior::KeepEmpty);
+
+    // 2. If the last item in parts is the empty string, then:
+    if (parts.last().is_empty()) {
+        // 1. If parts’s size is 1, then return false.
+        if (parts.size() == 1)
+            return false;
+
+        // 2. Remove the last item from parts.
+        parts.take_last();
+    }
+
+    // 3. Let last be the last item in parts.
+    auto last = parts.last();
+
+    // 4. If last is non-empty and contains only ASCII digits, then return true.
+    if (!last.is_empty() && all_of(last, is_ascii_digit))
+        return true;
+
+    // 5. If parsing last as an IPv4 number does not return failure, then return true.
+    if (parse_ipv4_number(last).has_value())
+        return true;
+
+    // 6. Return false.
+    return false;
+}
+
 // https://url.spec.whatwg.org/#concept-host-parser
 // NOTE: This is a very bare-bones implementation.
 static Optional<DeprecatedString> parse_host(StringView input, bool is_not_special = false)
@@ -565,7 +596,7 @@ static Optional<DeprecatedString> parse_host(StringView input, bool is_not_speci
     }
 
     // 8. If asciiDomain ends in a number, then return the result of IPv4 parsing asciiDomain.
-    if (is_ascii_digit(ascii_domain[ascii_domain.length() - 1])) {
+    if (ends_in_a_number_checker(ascii_domain)) {
         auto ipv4_host = parse_ipv4_address(ascii_domain);
         if (!ipv4_host.has_value())
             return {};
diff --git a/Tests/AK/TestURL.cpp b/Tests/AK/TestURL.cpp
index cca71b31b8..a97abe28a6 100644
--- a/Tests/AK/TestURL.cpp
+++ b/Tests/AK/TestURL.cpp
@@ -38,6 +38,16 @@ TEST_CASE(basic)
         EXPECT(url.query().is_null());
         EXPECT(url.fragment().is_null());
     }
+    {
+        URL url("https://www.serenityos.org1/index.html"sv);
+        EXPECT_EQ(url.is_valid(), true);
+        EXPECT_EQ(url.scheme(), "https");
+        EXPECT_EQ(url.host(), "www.serenityos.org1");
+        EXPECT_EQ(url.port_or_default(), 443);
+        EXPECT_EQ(url.serialize_path(), "/index.html");
+        EXPECT(url.query().is_null());
+        EXPECT(url.fragment().is_null());
+    }
     {
         URL url("https://localhost:1234/~anon/test/page.html"sv);
         EXPECT_EQ(url.is_valid(), true);