From 4db89dbbc6f3f920303ec7672d04aeb3a9cd93ac Mon Sep 17 00:00:00 2001
From: Frank Tang <ftang@chromium.org>
Date: Wed, 13 Dec 2023 18:02:21 -0800
Subject: [PATCH] ICU-22549 Add Normalizer2 Fuzzer

ICU-22549 Remove unnecessary include files
---
 icu4c/source/test/fuzzer/Makefile.in          |  1 +
 .../source/test/fuzzer/normalizer2_fuzzer.cpp | 79 +++++++++++++++++++
 2 files changed, 80 insertions(+)
 create mode 100644 icu4c/source/test/fuzzer/normalizer2_fuzzer.cpp

diff --git a/icu4c/source/test/fuzzer/Makefile.in b/icu4c/source/test/fuzzer/Makefile.in
index ab217ed10ba1..a8b28540e325 100644
--- a/icu4c/source/test/fuzzer/Makefile.in
+++ b/icu4c/source/test/fuzzer/Makefile.in
@@ -41,6 +41,7 @@ FUZZER_TARGETS = \
 		 date_time_pattern_generator_fuzzer \
 		 list_format_fuzzer locale_fuzzer \
 		 locale_morph_fuzzer \
+		 normalizer2_fuzzer \
 		 number_format_fuzzer \
 		 number_formatter_fuzzer \
 		 relative_date_time_formatter_fuzzer \
diff --git a/icu4c/source/test/fuzzer/normalizer2_fuzzer.cpp b/icu4c/source/test/fuzzer/normalizer2_fuzzer.cpp
new file mode 100644
index 000000000000..959da8c47b04
--- /dev/null
+++ b/icu4c/source/test/fuzzer/normalizer2_fuzzer.cpp
@@ -0,0 +1,79 @@
+// © 2023 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
+
+// Fuzzer for ICU Calendar.
+
+#include <cstring>
+
+#include "unicode/normalizer2.h"
+
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) {
+    uint16_t rnd;
+    UChar32 char1, char2;
+    if (size < sizeof(rnd) + sizeof(char1) + sizeof(char2)) return 0;
+    icu::StringPiece fuzzData(reinterpret_cast<const char *>(data), size);
+
+    std::memcpy(&rnd, fuzzData.data(), sizeof(rnd));
+    fuzzData.remove_prefix(sizeof(rnd));
+    std::memcpy(&char1, fuzzData.data(), sizeof(char1));
+    fuzzData.remove_prefix(sizeof(char1));
+    std::memcpy(&char2, fuzzData.data(), sizeof(char2));
+    fuzzData.remove_prefix(sizeof(char2));
+
+    size_t len = fuzzData.size() / sizeof(char16_t);
+    icu::UnicodeString text(false, reinterpret_cast<const char16_t*>(fuzzData.data()), len);
+
+    UErrorCode status = U_ZERO_ERROR;
+    const icu::Normalizer2* norm = nullptr;
+    switch (rnd % 6) {
+        case 0:
+            norm = icu::Normalizer2::getNFCInstance(status);
+            break;
+        case 1:
+            norm = icu::Normalizer2::getNFDInstance(status);
+            break;
+        case 2:
+            norm = icu::Normalizer2::getNFKCInstance(status);
+            break;
+        case 3:
+            norm = icu::Normalizer2::getNFKDInstance(status);
+            break;
+        case 4:
+            norm = icu::Normalizer2::getNFKCCasefoldInstance(status);
+            break;
+        case 5:
+            norm = icu::Normalizer2::getNFKCSimpleCasefoldInstance(status);
+            break;
+    }
+    if (U_SUCCESS(status)) {
+        norm->normalize(text, status);
+        status = U_ZERO_ERROR;
+
+        icu::UnicodeString out;
+
+        norm->normalize(text, out, status);
+        status = U_ZERO_ERROR;
+
+        norm->normalizeSecondAndAppend(out, text, status);
+        status = U_ZERO_ERROR;
+
+        norm->append(out, text, status);
+        status = U_ZERO_ERROR;
+
+        norm->getDecomposition(char1, out);
+        norm->getRawDecomposition(char1, out);
+        norm->composePair(char1, char2);
+        norm->getCombiningClass(char1);
+        norm->isNormalized(text, status);
+        status = U_ZERO_ERROR;
+
+        norm->quickCheck(text, status);
+        status = U_ZERO_ERROR;
+
+        norm->hasBoundaryBefore(char1);
+        norm->hasBoundaryAfter(char1);
+        norm->isInert(char1);
+    }
+
+    return EXIT_SUCCESS;
+}