ICU-22479 Add a new fuzzer to test more Locale methods

icu4c/source/test/fuzzer/Makefile.in

@@ -33,7 +33,7 @@ CPPFLAGS += -I$(srcdir) -I$(top_srcdir)/common -I$(top_srcdir)/i18n -I$(top_srcd
 DEFS += -D'U_TOPSRCDIR="$(top_srcdir)/"' -D'U_TOPBUILDDIR="$(BUILDDIR)"'
 LIBS = $(LIBCTESTFW) $(LIBICUTOOLUTIL) $(LIBICUIO) $(LIBICUI18N) $(LIBICUUC) $(DEFAULT_LIBS) $(LIB_M)
 
-FUZZER_TARGETS = break_iterator_fuzzer collator_compare_fuzzer collator_rulebased_fuzzer converter_fuzzer locale_fuzzer number_format_fuzzer ucasemap_fuzzer uloc_canonicalize_fuzzer uloc_for_language_tag_fuzzer uloc_get_name_fuzzer uloc_is_right_to_left_fuzzer uloc_open_keywords_fuzzer unicode_string_codepage_create_fuzzer uregex_open_fuzzer
+FUZZER_TARGETS = break_iterator_fuzzer collator_compare_fuzzer collator_rulebased_fuzzer converter_fuzzer locale_fuzzer locale_morph_fuzzer number_format_fuzzer ucasemap_fuzzer uloc_canonicalize_fuzzer uloc_for_language_tag_fuzzer uloc_get_name_fuzzer uloc_is_right_to_left_fuzzer uloc_open_keywords_fuzzer unicode_string_codepage_create_fuzzer uregex_open_fuzzer
 
 OBJECTS = $(FUZZER_TARGETS:%=%.o)
 OBJECTS += fuzzer_driver.o locale_util.o

icu4c/source/test/fuzzer/locale_morph_fuzzer.cpp

@@ -0,0 +1,109 @@
+// © 2019 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
+
+// Fuzzer for ICU Locales.
+
+#include <algorithm>
+#include <cstddef>
+#include <cstdint>
+#include <cstdlib>
+#include <set>
+#include <string>
+#include <vector>
+
+#include "unicode/locid.h"
+#include "unicode/localpointer.h"
+#include "unicode/stringpiece.h"
+
+#include "locale_util.h"
+
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) {
+    if (size < 1) return 0;
+    icu::StringPiece fuzzData(reinterpret_cast<const char *>(data), size);
+    uint8_t rnd = *fuzzData.data();
+    fuzzData.remove_prefix(1);
+    const std::string input = MakeZeroTerminatedInput(
+        (const uint8_t*)(fuzzData.data()), fuzzData.length());
+
+    icu::Locale locale(input.c_str());
+    UErrorCode status = U_ZERO_ERROR;
+    switch(rnd % 8) {
+        case 0:
+            locale.addLikelySubtags(status);
+            break;
+        case 1:
+            locale.minimizeSubtags(status);
+            break;
+        case 2:
+            locale.canonicalize(status);
+            break;
+        case 3:
+            {
+                icu::LocalPointer<icu::StringEnumeration> senum(
+                    locale.createKeywords(status), status);
+                while (U_SUCCESS(status) &&
+                       (senum->next(nullptr, status)) != nullptr) {
+                    // noop
+                }
+            }
+            break;
+        case 4:
+            {
+                icu::LocalPointer<icu::StringEnumeration> senum(
+                    locale.createUnicodeKeywords(status), status);
+                while (U_SUCCESS(status) &&
+                       (senum->next(nullptr, status)) != nullptr) {
+                    // noop
+                }
+            }
+            break;
+        case 5:
+            {
+                char buf[256];
+                icu::CheckedArrayByteSink sink(buf, rnd);
+                locale.toLanguageTag(sink, status);
+            }
+            break;
+        case 6:
+            {
+                std::set<std::string> keys;
+                locale.getKeywords<std::string>(
+                    std::insert_iterator<decltype(keys)>(keys, keys.begin()),
+                    status);
+                if (U_SUCCESS(status)) {
+                    char buf[256];
+                    icu::CheckedArrayByteSink sink(buf, rnd);
+                    for (std::set<std::string>::iterator it=keys.begin();
+                         it!=keys.end();
+                         ++it) {
+                        locale.getKeywordValue(
+                            icu::StringPiece(it->c_str(), it->length()), sink,
+                            status);
+                    }
+                }
+            }
+            break;
+        case 7:
+            {
+                std::set<std::string> keys;
+                locale.getUnicodeKeywords<std::string>(
+                    std::insert_iterator<decltype(keys)>(keys, keys.begin()),
+                    status);
+                if (U_SUCCESS(status)) {
+                    char buf[256];
+                    icu::CheckedArrayByteSink sink(buf, rnd);
+                    for (std::set<std::string>::iterator it=keys.begin();
+                         it!=keys.end();
+                         ++it) {
+                        locale.getUnicodeKeywordValue(
+                            icu::StringPiece(it->c_str(), it->length()), sink,
+                            status);
+                    }
+                }
+            }
+            break;
+        default:
+          break;
+    }
+    return EXIT_SUCCESS;
+}

icu4c/source/test/fuzzer/locale_morph_fuzzer_seed_corpus.txt

@@ -0,0 +1,30 @@
+1sr-Cyrl-RS
+2zz-UND
+3de_DE
+4De_dE
+5en-US-u-islamcal
+6zh-CN-a-myext-x-private
+7en-a-myext-b-another
+8de-419-DE
+9a-DE
+0ar-a-aaa-b-bbb-a-ccc
+asl-rozaj
+bsl-rozaj-biske
+cde-CH-1901
+dsl-IT-nedis
+ehy-Latn-IT-arevela
+faz-Arab-x-AZE-derbend
+gCompletely bad text for locale, even includes spaces and interpunction.!?
+hfr-u-nu-arabext
+ide-u-nu-beng-nu-bali-nu-khmr
+jfil-u-ca-buddhist
+kzh-Hant-CN-u-nu-arabext-ca-islamicc
+lja-u-ca-japanese
+mhr-HR-u-ca-roc
+n1234
+oAAAAaaaaaaaVVVV_-=+
+pja%-x-%e3
+qyq-U-98x-M85E-9pE-85Y-8xU985YE-89xxUy-xEyq-U85Y-8xE8-q5Y-8xU-9xE
+rUN-u-jTKMwi-a0q-aeae-KaG-aab-2aG2-36C-2uzeal-STqROK-U36-366-U86-83S-c3SZEC-SCG-1366-SG66-KMi-a0ae-qae-KaG-aab-1a3
+ssr-sr-u-z2r-4su-nms-5rsu-mns-6um-s5su-msu-ins1-7rzx-ianu-ssd-ss5r-d0r-U22su-n5sx-lvar-5su-ssu-nax-lvarUd-uimxE-112
+tHR-u-roc85Y-4xU-d0r-U22x-lvariant-ims-0d0U22-js17zr-rsm-u56su-csu-ins1-ins17Rz-zax-ianu-ssd-d0U22-js1-7rzrsm-u56su-csu-ins1-ins17Rz-z