LibURL: Correctly acquire the registrable domain for a URL

We were using the public suffix of the URL's host as its registrable
domain. But the registrable domain is actually the public suffix plus
one additional label.

Libraries/LibURL/Host.cpp

@@ -220,7 +220,7 @@ Optional<String> Host::registrable_domain() const
     // 3. Let registrableDomain be the registrable domain determined by running the Public Suffix List algorithm with host as domain. [PSL]
     // NOTE: The spec algorithm for the public suffix returns "*" by default, but get_public_suffix() returns an empty Optional.
     //       Remove the `value_or()` if and when we update it.
-    auto registrable_domain = get_public_suffix(host_string).value_or("*"_string);
+    auto registrable_domain = get_registrable_domain(host_string).value_or("*"_string);
 
     // 4. Assert: registrableDomain is an ASCII string that does not end with ".".
     VERIFY(all_of(registrable_domain.code_points(), is_ascii));

Libraries/LibURL/URL.cpp

@@ -527,4 +527,28 @@ Optional<String> get_public_suffix([[maybe_unused]] StringView host)
 #endif
 }
 
+// https://github.com/publicsuffix/list/wiki/Format#algorithm
+Optional<String> get_registrable_domain(StringView host)
+{
+    // The registered or registrable domain is the public suffix plus one additional label.
+    auto public_suffix = get_public_suffix(host);
+    if (!public_suffix.has_value() || !host.ends_with(*public_suffix))
+        return {};
+
+    if (host == *public_suffix)
+        return {};
+
+    auto subhost = host.substring_view(0, host.length() - public_suffix->bytes_as_string_view().length());
+    subhost = subhost.trim("."sv, TrimMode::Right);
+
+    if (subhost.is_empty())
+        return {};
+
+    size_t start_index = 0;
+    if (auto index = subhost.find_last('.'); index.has_value())
+        start_index = *index + 1;
+
+    return MUST(String::from_utf8(host.substring_view(start_index)));
+}
+
 }

Libraries/LibURL/URL.h

@@ -203,6 +203,7 @@ URL create_with_data(StringView mime_type, StringView payload, bool is_base64 =
 
 bool is_public_suffix(StringView host);
 Optional<String> get_public_suffix(StringView host);
+Optional<String> get_registrable_domain(StringView host);
 
 inline URL about_blank() { return URL::about("blank"_string); }
 inline URL about_srcdoc() { return URL::about("srcdoc"_string); }

Tests/LibURL/TestURL.cpp

@@ -558,3 +558,52 @@ TEST_CASE(invalid_domain_code_points)
         EXPECT(!url.has_value());
     }
 }
+
+TEST_CASE(get_registrable_domain)
+{
+    {
+        auto domain = URL::get_registrable_domain({});
+        EXPECT(!domain.has_value());
+    }
+    {
+        auto domain = URL::get_registrable_domain("foobar"sv);
+        EXPECT(!domain.has_value());
+    }
+    {
+        auto domain = URL::get_registrable_domain("com"sv);
+        EXPECT(!domain.has_value());
+    }
+    {
+        auto domain = URL::get_registrable_domain(".com"sv);
+        EXPECT(!domain.has_value());
+    }
+    {
+        auto domain = URL::get_registrable_domain("example.com"sv);
+        VERIFY(domain.has_value());
+        EXPECT_EQ(*domain, "example.com"sv);
+    }
+    {
+        auto domain = URL::get_registrable_domain(".example.com"sv);
+        VERIFY(domain.has_value());
+        EXPECT_EQ(*domain, "example.com"sv);
+    }
+    {
+        auto domain = URL::get_registrable_domain("www.example.com"sv);
+        VERIFY(domain.has_value());
+        EXPECT_EQ(*domain, "example.com"sv);
+    }
+    {
+        auto domain = URL::get_registrable_domain("sub.www.example.com"sv);
+        VERIFY(domain.has_value());
+        EXPECT_EQ(*domain, "example.com"sv);
+    }
+    {
+        auto domain = URL::get_registrable_domain("github.io"sv);
+        EXPECT(!domain.has_value());
+    }
+    {
+        auto domain = URL::get_registrable_domain("ladybird.github.io"sv);
+        VERIFY(domain.has_value());
+        EXPECT_EQ(*domain, "ladybird.github.io"sv);
+    }
+}