LibWeb: Only use code unit length for is_code_unit_prefix

Fixes a crash in the included test.

Libraries/LibWeb/Infra/Strings.cpp

@@ -69,10 +69,12 @@ ErrorOr<String> strip_and_collapse_whitespace(StringView string)
 }
 
 // https://infra.spec.whatwg.org/#code-unit-prefix
-bool is_code_unit_prefix(StringView potential_prefix, StringView input)
+bool is_code_unit_prefix(StringView potential_prefix_utf8, StringView input_utf8)
 {
-    auto potential_prefix_utf16 = MUST(utf8_to_utf16(potential_prefix));
+    auto potential_prefix_utf16_bytes = MUST(utf8_to_utf16(potential_prefix_utf8));
-    auto input_utf16 = MUST(utf8_to_utf16(input));
+    auto input_utf16_bytes = MUST(utf8_to_utf16(input_utf8));
+    Utf16View potential_prefix { potential_prefix_utf16_bytes };
+    Utf16View input { input_utf16_bytes };
 
     // 1. Let i be 0.
     size_t i = 0;
@@ -80,18 +82,18 @@ bool is_code_unit_prefix(StringView potential_prefix, StringView input)
     // 2. While true:
     while (true) {
         // 1. If i is greater than or equal to potentialPrefix’s length, then return true.
-        if (i >= potential_prefix.length())
+        if (i >= potential_prefix.length_in_code_units())
             return true;
 
         // 2. If i is greater than or equal to input’s length, then return false.
-        if (i >= input.length())
+        if (i >= input.length_in_code_units())
             return false;
 
         // 3. Let potentialPrefixCodeUnit be the ith code unit of potentialPrefix.
-        auto potential_prefix_code_unit = Utf16View(potential_prefix_utf16).code_unit_at(i);
+        auto potential_prefix_code_unit = potential_prefix.code_unit_at(i);
 
         // 4. Let inputCodeUnit be the ith code unit of input.
-        auto input_code_unit = Utf16View(input_utf16).code_unit_at(i);
+        auto input_code_unit = input.code_unit_at(i);
 
         // 5. Return false if potentialPrefixCodeUnit is not inputCodeUnit.
         if (potential_prefix_code_unit != input_code_unit)

Tests/LibWeb/CMakeLists.txt

@@ -9,6 +9,7 @@ set(TEST_SOURCES
     TestMicrosyntax.cpp
     TestMimeSniff.cpp
     TestNumbers.cpp
+    TestStrings.cpp
 )
 
 foreach(source IN LISTS TEST_SOURCES)

Tests/LibWeb/TestStrings.cpp

@@ -0,0 +1,45 @@
+/*
+ * Copyright (c) 2025, Shannon Booth <shannon@serenityos.org>
+ *
+ * SPDX-License-Identifier: BSD-2-Clause
+ */
+
+#include <LibTest/TestCase.h>
+#include <LibWeb/Infra/Strings.h>
+
+TEST_CASE(is_code_unit_prefix)
+{
+
+    // Basic prefix match
+    EXPECT(Web::Infra::is_code_unit_prefix("abc"sv, "abcde"sv));
+
+    // Exact match
+    EXPECT(Web::Infra::is_code_unit_prefix("abc"sv, "abc"sv));
+
+    // Empty prefix
+    EXPECT(Web::Infra::is_code_unit_prefix(""sv, "abc"sv));
+
+    // Empty input string
+    EXPECT(!Web::Infra::is_code_unit_prefix("abc"sv, ""sv));
+
+    // Both strings empty
+    EXPECT(Web::Infra::is_code_unit_prefix(""sv, ""sv));
+
+    // Prefix longer than input string
+    EXPECT(!Web::Infra::is_code_unit_prefix("abcdef"sv, "abc"sv));
+
+    // Non-ASCII characters
+    EXPECT(Web::Infra::is_code_unit_prefix("こんにちは"sv, "こんにちは世界"sv));
+    EXPECT(!Web::Infra::is_code_unit_prefix("世界"sv, "こんにちは世界"sv));
+
+    EXPECT(Web::Infra::is_code_unit_prefix("こ"sv, "こん"sv));
+    EXPECT(!Web::Infra::is_code_unit_prefix("こん"sv, "こ"sv));
+
+    // Special characters
+    EXPECT(Web::Infra::is_code_unit_prefix("!@#"sv, "!@#$%^"sv));
+    EXPECT(!Web::Infra::is_code_unit_prefix("!@#$"sv, "!@#"sv));
+
+    // Case sensitivity
+    EXPECT(!Web::Infra::is_code_unit_prefix("abc"sv, "ABC"sv));
+    EXPECT(!Web::Infra::is_code_unit_prefix("ABC"sv, "abc"sv));
+}