+// © 2016 and later: Unicode, Inc. and others.
+// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
-* Copyright (C) 2002-2012, International Business Machines
+* Copyright (C) 2002-2016, International Business Machines
* Corporation and others. All Rights Reserved.
*
*******************************************************************************
* file name: strcase.cpp
-* encoding: US-ASCII
+* encoding: UTF-8
* tab size: 8 (not used)
* indentation:4
*
* Test file for string casing C++ API functions.
*/
+#include "unicode/std_string.h"
+#include "unicode/casemap.h"
+#include "unicode/edits.h"
#include "unicode/uchar.h"
#include "unicode/ures.h"
#include "unicode/uloc.h"
#include "ucase.h"
#include "ustrtest.h"
#include "unicode/tstdtmod.h"
+#include "cmemory.h"
-#define LENGTHOF(array) (int32_t)(sizeof(array)/sizeof((array)[0]))
+struct EditChange {
+ UBool change;
+ int32_t oldLength, newLength;
+};
+
+class StringCaseTest: public IntlTest {
+public:
+ StringCaseTest();
+ virtual ~StringCaseTest();
+
+ void runIndexedTest(int32_t index, UBool exec, const char *&name, char *par=0);
+
+ void TestCaseConversion();
+
+ void TestCasingImpl(const UnicodeString &input,
+ const UnicodeString &output,
+ int32_t whichCase,
+ void *iter, const char *localeID, uint32_t options);
+ void TestCasing();
+ void TestFullCaseFoldingIterator();
+ void TestGreekUpper();
+ void TestLongUpper();
+ void TestMalformedUTF8();
+ void TestBufferOverflow();
+ void TestEdits();
+ void TestCaseMapWithEdits();
+ void TestCaseMapUTF8WithEdits();
+ void TestLongUnicodeString();
+ void TestBug13127();
+
+private:
+ void assertGreekUpper(const char16_t *s, const char16_t *expected);
+ void checkEditsIter(
+ const UnicodeString &name, Edits::Iterator ei1, Edits::Iterator ei2, // two equal iterators
+ const EditChange expected[], int32_t expLength, UBool withUnchanged,
+ UErrorCode &errorCode);
+
+ Locale GREEK_LOCALE_;
+};
+
+StringCaseTest::StringCaseTest() : GREEK_LOCALE_("el") {}
StringCaseTest::~StringCaseTest() {}
+extern IntlTest *createStringCaseTest() {
+ return new StringCaseTest();
+}
+
void
StringCaseTest::runIndexedTest(int32_t index, UBool exec, const char *&name, char * /*par*/) {
if(exec) {
TESTCASE_AUTO(TestCaseConversion);
#if !UCONFIG_NO_BREAK_ITERATION && !UCONFIG_NO_FILE_IO && !UCONFIG_NO_LEGACY_CONVERSION
TESTCASE_AUTO(TestCasing);
+ TESTCASE_AUTO(TestBug13127);
#endif
TESTCASE_AUTO(TestFullCaseFoldingIterator);
+ TESTCASE_AUTO(TestGreekUpper);
+ TESTCASE_AUTO(TestLongUpper);
+ TESTCASE_AUTO(TestMalformedUTF8);
+ TESTCASE_AUTO(TestBufferOverflow);
+ TESTCASE_AUTO(TestEdits);
+ TESTCASE_AUTO(TestCaseMapWithEdits);
+ TESTCASE_AUTO(TestCaseMapUTF8WithEdits);
+ TESTCASE_AUTO(TestLongUnicodeString);
TESTCASE_AUTO_END;
}
UnicodeString s;
/* lowercase with root locale */
- s=UnicodeString(FALSE, beforeLower, (int32_t)(sizeof(beforeLower)/U_SIZEOF_UCHAR));
+ s=UnicodeString(FALSE, beforeLower, UPRV_LENGTHOF(beforeLower));
s.toLower("");
- if( s.length()!=(sizeof(lowerRoot)/U_SIZEOF_UCHAR) ||
+ if( s.length()!=UPRV_LENGTHOF(lowerRoot) ||
s!=UnicodeString(FALSE, lowerRoot, s.length())
) {
- errln("error in toLower(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, lowerRoot, (int32_t)(sizeof(lowerRoot)/U_SIZEOF_UCHAR)) + "\"");
+ errln("error in toLower(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, lowerRoot, UPRV_LENGTHOF(lowerRoot)) + "\"");
}
/* lowercase with turkish locale */
- s=UnicodeString(FALSE, beforeLower, (int32_t)(sizeof(beforeLower)/U_SIZEOF_UCHAR));
+ s=UnicodeString(FALSE, beforeLower, UPRV_LENGTHOF(beforeLower));
s.setCharAt(0, beforeLower[0]).toLower(Locale("tr"));
- if( s.length()!=(sizeof(lowerTurkish)/U_SIZEOF_UCHAR) ||
+ if( s.length()!=UPRV_LENGTHOF(lowerTurkish) ||
s!=UnicodeString(FALSE, lowerTurkish, s.length())
) {
- errln("error in toLower(turkish locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, lowerTurkish, (int32_t)(sizeof(lowerTurkish)/U_SIZEOF_UCHAR)) + "\"");
+ errln("error in toLower(turkish locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, lowerTurkish, UPRV_LENGTHOF(lowerTurkish)) + "\"");
}
/* uppercase with root locale */
- s=UnicodeString(FALSE, beforeUpper, (int32_t)(sizeof(beforeUpper)/U_SIZEOF_UCHAR));
+ s=UnicodeString(FALSE, beforeUpper, UPRV_LENGTHOF(beforeUpper));
s.setCharAt(0, beforeUpper[0]).toUpper(Locale(""));
- if( s.length()!=(sizeof(upperRoot)/U_SIZEOF_UCHAR) ||
+ if( s.length()!=UPRV_LENGTHOF(upperRoot) ||
s!=UnicodeString(FALSE, upperRoot, s.length())
) {
- errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, upperRoot, (int32_t)(sizeof(upperRoot)/U_SIZEOF_UCHAR)) + "\"");
+ errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, upperRoot, UPRV_LENGTHOF(upperRoot)) + "\"");
}
/* uppercase with turkish locale */
- s=UnicodeString(FALSE, beforeUpper, (int32_t)(sizeof(beforeUpper)/U_SIZEOF_UCHAR));
+ s=UnicodeString(FALSE, beforeUpper, UPRV_LENGTHOF(beforeUpper));
s.toUpper(Locale("tr"));
- if( s.length()!=(sizeof(upperTurkish)/U_SIZEOF_UCHAR) ||
+ if( s.length()!=UPRV_LENGTHOF(upperTurkish) ||
s!=UnicodeString(FALSE, upperTurkish, s.length())
) {
- errln("error in toUpper(turkish locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, upperTurkish, (int32_t)(sizeof(upperTurkish)/U_SIZEOF_UCHAR)) + "\"");
+ errln("error in toUpper(turkish locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, upperTurkish, UPRV_LENGTHOF(upperTurkish)) + "\"");
}
/* uppercase a short string with root locale */
- s=UnicodeString(FALSE, beforeMiniUpper, (int32_t)(sizeof(beforeMiniUpper)/U_SIZEOF_UCHAR));
+ s=UnicodeString(FALSE, beforeMiniUpper, UPRV_LENGTHOF(beforeMiniUpper));
s.setCharAt(0, beforeMiniUpper[0]).toUpper("");
- if( s.length()!=(sizeof(miniUpper)/U_SIZEOF_UCHAR) ||
+ if( s.length()!=UPRV_LENGTHOF(miniUpper) ||
s!=UnicodeString(FALSE, miniUpper, s.length())
) {
- errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, miniUpper, (int32_t)(sizeof(miniUpper)/U_SIZEOF_UCHAR)) + "\"");
+ errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, miniUpper, UPRV_LENGTHOF(miniUpper)) + "\"");
}
}
#if !UCONFIG_NO_BREAK_ITERATION
if(iter!=NULL) {
// Clone the break iterator so that the UCaseMap can safely adopt it.
- int32_t size=1; // Not 0 because that only gives preflighting.
- UBreakIterator *clone=ubrk_safeClone((UBreakIterator *)iter, NULL, &size, errorCode);
+ UBreakIterator *clone=ubrk_safeClone((UBreakIterator *)iter, NULL, NULL, errorCode);
ucasemap_setBreakIterator(csm.getAlias(), clone, errorCode);
}
#endif
// or even just { 0 } as boundaries.
static const UChar rules[] = { 0x2e, 0x2a, 0x3b }; // ".*;"
UParseError parseError;
- iter.adoptInstead(ubrk_openRules(rules, LENGTHOF(rules), NULL, 0, &parseError, &status));
+ iter.adoptInstead(ubrk_openRules(rules, UPRV_LENGTHOF(rules), NULL, 0, &parseError, &status));
}
}
#endif
errln("error: FullCaseFoldingIterator yielded only %d (cp, full) pairs", (int)count);
}
}
+
+void
+StringCaseTest::assertGreekUpper(const char16_t *s, const char16_t *expected) {
+ UnicodeString s16(s);
+ UnicodeString expected16(expected);
+ UnicodeString msg = UnicodeString("UnicodeString::toUpper/Greek(\"") + s16 + "\")";
+ UnicodeString result16(s16);
+ result16.toUpper(GREEK_LOCALE_);
+ assertEquals(msg, expected16, result16);
+
+ msg = UnicodeString("u_strToUpper/Greek(\"") + s16 + "\") cap=";
+ int32_t length = expected16.length();
+ int32_t capacities[] = {
+ // Keep in sync with the UTF-8 capacities near the bottom of this function.
+ 0, length / 2, length - 1, length, length + 1
+ };
+ for (int32_t i = 0; i < UPRV_LENGTHOF(capacities); ++i) {
+ int32_t cap = capacities[i];
+ UChar *dest16 = result16.getBuffer(expected16.length() + 1);
+ u_memset(dest16, 0x55AA, result16.getCapacity());
+ UErrorCode errorCode = U_ZERO_ERROR;
+ length = u_strToUpper(dest16, cap, s16.getBuffer(), s16.length(), "el", &errorCode);
+ assertEquals(msg + cap, expected16.length(), length);
+ UErrorCode expectedErrorCode;
+ if (cap < expected16.length()) {
+ expectedErrorCode = U_BUFFER_OVERFLOW_ERROR;
+ } else if (cap == expected16.length()) {
+ expectedErrorCode = U_STRING_NOT_TERMINATED_WARNING;
+ } else {
+ expectedErrorCode = U_ZERO_ERROR;
+ assertEquals(msg + cap + " NUL", 0, dest16[length]);
+ }
+ assertEquals(msg + cap + " errorCode", expectedErrorCode, errorCode);
+ result16.releaseBuffer(length);
+ if (cap >= expected16.length()) {
+ assertEquals(msg + cap, expected16, result16);
+ }
+ }
+
+ UErrorCode errorCode = U_ZERO_ERROR;
+ LocalUCaseMapPointer csm(ucasemap_open("el", 0, &errorCode));
+ assertSuccess("ucasemap_open", errorCode);
+ std::string s8;
+ s16.toUTF8String(s8);
+ msg = UnicodeString("ucasemap_utf8ToUpper/Greek(\"") + s16 + "\")";
+ char dest8[1000];
+ length = ucasemap_utf8ToUpper(csm.getAlias(), dest8, UPRV_LENGTHOF(dest8),
+ s8.data(), s8.length(), &errorCode);
+ assertSuccess("ucasemap_utf8ToUpper", errorCode);
+ StringPiece result8(dest8, length);
+ UnicodeString result16From8 = UnicodeString::fromUTF8(result8);
+ assertEquals(msg, expected16, result16From8);
+
+ msg += " cap=";
+ capacities[1] = length / 2;
+ capacities[2] = length - 1;
+ capacities[3] = length;
+ capacities[4] = length + 1;
+ char dest8b[1000];
+ int32_t expected8Length = length; // Assuming the previous call worked.
+ for (int32_t i = 0; i < UPRV_LENGTHOF(capacities); ++i) {
+ int32_t cap = capacities[i];
+ memset(dest8b, 0x5A, UPRV_LENGTHOF(dest8b));
+ UErrorCode errorCode = U_ZERO_ERROR;
+ length = ucasemap_utf8ToUpper(csm.getAlias(), dest8b, cap,
+ s8.data(), s8.length(), &errorCode);
+ assertEquals(msg + cap, expected8Length, length);
+ UErrorCode expectedErrorCode;
+ if (cap < expected8Length) {
+ expectedErrorCode = U_BUFFER_OVERFLOW_ERROR;
+ } else if (cap == expected8Length) {
+ expectedErrorCode = U_STRING_NOT_TERMINATED_WARNING;
+ } else {
+ expectedErrorCode = U_ZERO_ERROR;
+ // Casts to int32_t to avoid matching UBool.
+ assertEquals(msg + cap + " NUL", (int32_t)0, (int32_t)dest8b[length]);
+ }
+ assertEquals(msg + cap + " errorCode", expectedErrorCode, errorCode);
+ if (cap >= expected8Length) {
+ assertEquals(msg + cap + " (memcmp)", 0, memcmp(dest8, dest8b, expected8Length));
+ }
+ }
+}
+
+void
+StringCaseTest::TestGreekUpper() {
+ // http://bugs.icu-project.org/trac/ticket/5456
+ assertGreekUpper(u"άδικος, κείμενο, ίριδα", u"ΑΔΙΚΟΣ, ΚΕΙΜΕΝΟ, ΙΡΙΔΑ");
+ // https://bugzilla.mozilla.org/show_bug.cgi?id=307039
+ // https://bug307039.bmoattachments.org/attachment.cgi?id=194893
+ assertGreekUpper(u"Πατάτα", u"ΠΑΤΑΤΑ");
+ assertGreekUpper(u"Αέρας, Μυστήριο, Ωραίο", u"ΑΕΡΑΣ, ΜΥΣΤΗΡΙΟ, ΩΡΑΙΟ");
+ assertGreekUpper(u"Μαΐου, Πόρος, Ρύθμιση", u"ΜΑΪΟΥ, ΠΟΡΟΣ, ΡΥΘΜΙΣΗ");
+ assertGreekUpper(u"ΰ, Τηρώ, Μάιος", u"Ϋ, ΤΗΡΩ, ΜΑΪΟΣ");
+ assertGreekUpper(u"άυλος", u"ΑΫΛΟΣ");
+ assertGreekUpper(u"ΑΫΛΟΣ", u"ΑΫΛΟΣ");
+ assertGreekUpper(u"Άκλιτα ρήματα ή άκλιτες μετοχές", u"ΑΚΛΙΤΑ ΡΗΜΑΤΑ Ή ΑΚΛΙΤΕΣ ΜΕΤΟΧΕΣ");
+ // http://www.unicode.org/udhr/d/udhr_ell_monotonic.html
+ assertGreekUpper(u"Επειδή η αναγνώριση της αξιοπρέπειας", u"ΕΠΕΙΔΗ Η ΑΝΑΓΝΩΡΙΣΗ ΤΗΣ ΑΞΙΟΠΡΕΠΕΙΑΣ");
+ assertGreekUpper(u"νομικού ή διεθνούς", u"ΝΟΜΙΚΟΥ Ή ΔΙΕΘΝΟΥΣ");
+ // http://unicode.org/udhr/d/udhr_ell_polytonic.html
+ assertGreekUpper(u"Ἐπειδὴ ἡ ἀναγνώριση", u"ΕΠΕΙΔΗ Η ΑΝΑΓΝΩΡΙΣΗ");
+ assertGreekUpper(u"νομικοῦ ἢ διεθνοῦς", u"ΝΟΜΙΚΟΥ Ή ΔΙΕΘΝΟΥΣ");
+ // From Google bug report
+ assertGreekUpper(u"Νέο, Δημιουργία", u"ΝΕΟ, ΔΗΜΙΟΥΡΓΙΑ");
+ // http://crbug.com/234797
+ assertGreekUpper(u"Ελάτε να φάτε τα καλύτερα παϊδάκια!", u"ΕΛΑΤΕ ΝΑ ΦΑΤΕ ΤΑ ΚΑΛΥΤΕΡΑ ΠΑΪΔΑΚΙΑ!");
+ assertGreekUpper(u"Μαΐου, τρόλεϊ", u"ΜΑΪΟΥ, ΤΡΟΛΕΪ");
+ assertGreekUpper(u"Το ένα ή το άλλο.", u"ΤΟ ΕΝΑ Ή ΤΟ ΑΛΛΟ.");
+ // http://multilingualtypesetting.co.uk/blog/greek-typesetting-tips/
+ assertGreekUpper(u"ρωμέικα", u"ΡΩΜΕΪΚΑ");
+}
+
+void
+StringCaseTest::TestLongUpper() {
+ if (quick) {
+ logln("not exhaustive mode: skipping this test");
+ return;
+ }
+ // Ticket #12663, crash with an extremely long string where
+ // U+0390 maps to 0399 0308 0301 so that the result is three times as long
+ // and overflows an int32_t.
+ int32_t length = 0x40000004; // more than 1G UChars
+ UnicodeString s(length, (UChar32)0x390, length);
+ UnicodeString result;
+ UChar *dest = result.getBuffer(length + 1);
+ if (s.isBogus() || dest == NULL) {
+ logln("Out of memory, unable to run this test on this machine.");
+ return;
+ }
+ IcuTestErrorCode errorCode(*this, "TestLongUpper");
+ int32_t destLength = u_strToUpper(dest, result.getCapacity(),
+ s.getBuffer(), s.length(), "", errorCode);
+ result.releaseBuffer(destLength);
+ if (errorCode.reset() != U_INDEX_OUTOFBOUNDS_ERROR) {
+ errln("expected U_INDEX_OUTOFBOUNDS_ERROR, got %s (destLength is undefined, got %ld)",
+ errorCode.errorName(), (long)destLength);
+ }
+}
+
+void StringCaseTest::TestMalformedUTF8() {
+ // ticket #12639
+ IcuTestErrorCode errorCode(*this, "TestMalformedUTF8");
+ LocalUCaseMapPointer csm(ucasemap_open("en", U_TITLECASE_NO_BREAK_ADJUSTMENT, errorCode));
+ if (errorCode.isFailure()) {
+ errln("ucasemap_open(English) failed - %s", errorCode.errorName());
+ return;
+ }
+ char src[1] = { (char)0x85 }; // malformed UTF-8
+ char dest[3] = { 0, 0, 0 };
+ int32_t destLength;
+#if !UCONFIG_NO_BREAK_ITERATION
+ destLength = ucasemap_utf8ToTitle(csm.getAlias(), dest, 3, src, 1, errorCode);
+ if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) {
+ errln("ucasemap_utf8ToTitle(\\x85) failed: %s destLength=%d dest[0]=0x%02x",
+ errorCode.errorName(), (int)destLength, dest[0]);
+ }
+#endif
+
+ errorCode.reset();
+ dest[0] = 0;
+ destLength = ucasemap_utf8ToLower(csm.getAlias(), dest, 3, src, 1, errorCode);
+ if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) {
+ errln("ucasemap_utf8ToLower(\\x85) failed: %s destLength=%d dest[0]=0x%02x",
+ errorCode.errorName(), (int)destLength, dest[0]);
+ }
+
+ errorCode.reset();
+ dest[0] = 0;
+ destLength = ucasemap_utf8ToUpper(csm.getAlias(), dest, 3, src, 1, errorCode);
+ if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) {
+ errln("ucasemap_utf8ToUpper(\\x85) failed: %s destLength=%d dest[0]=0x%02x",
+ errorCode.errorName(), (int)destLength, dest[0]);
+ }
+
+ errorCode.reset();
+ dest[0] = 0;
+ destLength = ucasemap_utf8FoldCase(csm.getAlias(), dest, 3, src, 1, errorCode);
+ if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) {
+ errln("ucasemap_utf8FoldCase(\\x85) failed: %s destLength=%d dest[0]=0x%02x",
+ errorCode.errorName(), (int)destLength, dest[0]);
+ }
+}
+
+void StringCaseTest::TestBufferOverflow() {
+ // Ticket #12849, incorrect result from Title Case preflight operation,
+ // when buffer overflow error is expected.
+ IcuTestErrorCode errorCode(*this, "TestBufferOverflow");
+ LocalUCaseMapPointer csm(ucasemap_open("en", 0, errorCode));
+ if (errorCode.isFailure()) {
+ errln("ucasemap_open(English) failed - %s", errorCode.errorName());
+ return;
+ }
+
+ UnicodeString data("hello world");
+ int32_t result;
+#if !UCONFIG_NO_BREAK_ITERATION
+ result = ucasemap_toTitle(csm.getAlias(), NULL, 0, data.getBuffer(), data.length(), errorCode);
+ if (errorCode.get() != U_BUFFER_OVERFLOW_ERROR || result != data.length()) {
+ errln("%s:%d ucasemap_toTitle(\"hello world\") failed: "
+ "expected (U_BUFFER_OVERFLOW_ERROR, %d), got (%s, %d)",
+ __FILE__, __LINE__, data.length(), errorCode.errorName(), result);
+ }
+#endif
+ errorCode.reset();
+
+ std::string data_utf8;
+ data.toUTF8String(data_utf8);
+#if !UCONFIG_NO_BREAK_ITERATION
+ result = ucasemap_utf8ToTitle(csm.getAlias(), NULL, 0, data_utf8.c_str(), data_utf8.length(), errorCode);
+ if (errorCode.get() != U_BUFFER_OVERFLOW_ERROR || result != (int32_t)data_utf8.length()) {
+ errln("%s:%d ucasemap_toTitle(\"hello world\") failed: "
+ "expected (U_BUFFER_OVERFLOW_ERROR, %d), got (%s, %d)",
+ __FILE__, __LINE__, data_utf8.length(), errorCode.errorName(), result);
+ }
+#endif
+ errorCode.reset();
+}
+
+void StringCaseTest::checkEditsIter(
+ const UnicodeString &name,
+ Edits::Iterator ei1, Edits::Iterator ei2, // two equal iterators
+ const EditChange expected[], int32_t expLength, UBool withUnchanged,
+ UErrorCode &errorCode) {
+ assertFalse(name, ei2.findSourceIndex(-1, errorCode));
+
+ int32_t expSrcIndex = 0;
+ int32_t expDestIndex = 0;
+ int32_t expReplIndex = 0;
+ for (int32_t expIndex = 0; expIndex < expLength; ++expIndex) {
+ const EditChange &expect = expected[expIndex];
+ UnicodeString msg = UnicodeString(name).append(u' ') + expIndex;
+ if (withUnchanged || expect.change) {
+ assertTrue(msg, ei1.next(errorCode));
+ assertEquals(msg, expect.change, ei1.hasChange());
+ assertEquals(msg, expect.oldLength, ei1.oldLength());
+ assertEquals(msg, expect.newLength, ei1.newLength());
+ assertEquals(msg, expSrcIndex, ei1.sourceIndex());
+ assertEquals(msg, expDestIndex, ei1.destinationIndex());
+ assertEquals(msg, expReplIndex, ei1.replacementIndex());
+ }
+
+ if (expect.oldLength > 0) {
+ assertTrue(msg, ei2.findSourceIndex(expSrcIndex, errorCode));
+ assertEquals(msg, expect.change, ei2.hasChange());
+ assertEquals(msg, expect.oldLength, ei2.oldLength());
+ assertEquals(msg, expect.newLength, ei2.newLength());
+ assertEquals(msg, expSrcIndex, ei2.sourceIndex());
+ assertEquals(msg, expDestIndex, ei2.destinationIndex());
+ assertEquals(msg, expReplIndex, ei2.replacementIndex());
+ if (!withUnchanged) {
+ // For some iterators, move past the current range
+ // so that findSourceIndex() has to look before the current index.
+ ei2.next(errorCode);
+ ei2.next(errorCode);
+ }
+ }
+
+ expSrcIndex += expect.oldLength;
+ expDestIndex += expect.newLength;
+ if (expect.change) {
+ expReplIndex += expect.newLength;
+ }
+ }
+ // TODO: remove casts from u"" when merging into trunk
+ UnicodeString msg = UnicodeString(name).append(u" end");
+ assertFalse(msg, ei1.next(errorCode));
+ assertFalse(msg, ei1.hasChange());
+ assertEquals(msg, 0, ei1.oldLength());
+ assertEquals(msg, 0, ei1.newLength());
+ assertEquals(msg, expSrcIndex, ei1.sourceIndex());
+ assertEquals(msg, expDestIndex, ei1.destinationIndex());
+ assertEquals(msg, expReplIndex, ei1.replacementIndex());
+
+ assertFalse(name, ei2.findSourceIndex(expSrcIndex, errorCode));
+}
+
+void StringCaseTest::TestEdits() {
+ IcuTestErrorCode errorCode(*this, "TestEdits");
+ Edits edits;
+ assertFalse("new Edits", edits.hasChanges());
+ assertEquals("new Edits", 0, edits.lengthDelta());
+ edits.addUnchanged(1); // multiple unchanged ranges are combined
+ edits.addUnchanged(10000); // too long, and they are split
+ edits.addReplace(0, 0);
+ edits.addUnchanged(2);
+ assertFalse("unchanged 10003", edits.hasChanges());
+ assertEquals("unchanged 10003", 0, edits.lengthDelta());
+ edits.addReplace(1, 1); // multiple short equal-length edits are compressed
+ edits.addUnchanged(0);
+ edits.addReplace(1, 1);
+ edits.addReplace(1, 1);
+ edits.addReplace(0, 10);
+ edits.addReplace(100, 0);
+ edits.addReplace(3000, 4000); // variable-length encoding
+ edits.addReplace(100000, 100000);
+ assertTrue("some edits", edits.hasChanges());
+ assertEquals("some edits", 10 - 100 + 1000, edits.lengthDelta());
+ UErrorCode outErrorCode = U_ZERO_ERROR;
+ assertFalse("edits done: copyErrorTo", edits.copyErrorTo(outErrorCode));
+
+ static const EditChange coarseExpectedChanges[] = {
+ { FALSE, 10003, 10003 },
+ { TRUE, 103103, 104013 }
+ };
+ checkEditsIter(u"coarse",
+ edits.getCoarseIterator(), edits.getCoarseIterator(),
+ coarseExpectedChanges, UPRV_LENGTHOF(coarseExpectedChanges), TRUE, errorCode);
+ checkEditsIter(u"coarse changes",
+ edits.getCoarseChangesIterator(), edits.getCoarseChangesIterator(),
+ coarseExpectedChanges, UPRV_LENGTHOF(coarseExpectedChanges), FALSE, errorCode);
+
+ static const EditChange fineExpectedChanges[] = {
+ { FALSE, 10003, 10003 },
+ { TRUE, 1, 1 },
+ { TRUE, 1, 1 },
+ { TRUE, 1, 1 },
+ { TRUE, 0, 10 },
+ { TRUE, 100, 0 },
+ { TRUE, 3000, 4000 },
+ { TRUE, 100000, 100000 }
+ };
+ checkEditsIter(u"fine",
+ edits.getFineIterator(), edits.getFineIterator(),
+ fineExpectedChanges, UPRV_LENGTHOF(fineExpectedChanges), TRUE, errorCode);
+ checkEditsIter(u"fine changes",
+ edits.getFineChangesIterator(), edits.getFineChangesIterator(),
+ fineExpectedChanges, UPRV_LENGTHOF(fineExpectedChanges), FALSE, errorCode);
+
+ edits.reset();
+ assertFalse("reset", edits.hasChanges());
+ assertEquals("reset", 0, edits.lengthDelta());
+ Edits::Iterator ei = edits.getCoarseChangesIterator();
+ assertFalse("reset then iterator", ei.next(errorCode));
+}
+
+void StringCaseTest::TestCaseMapWithEdits() {
+ IcuTestErrorCode errorCode(*this, "TestEdits");
+ UChar dest[20];
+ Edits edits;
+
+ int32_t length = CaseMap::toLower("tr", UCASEMAP_OMIT_UNCHANGED_TEXT,
+ u"IstanBul", 8, dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"toLower(IstanBul)", UnicodeString(u"ıb"), UnicodeString(TRUE, dest, length));
+ static const EditChange lowerExpectedChanges[] = {
+ { TRUE, 1, 1 },
+ { FALSE, 4, 4 },
+ { TRUE, 1, 1 },
+ { FALSE, 2, 2 }
+ };
+ checkEditsIter(u"toLower(IstanBul)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ lowerExpectedChanges, UPRV_LENGTHOF(lowerExpectedChanges),
+ TRUE, errorCode);
+
+ edits.reset();
+ length = CaseMap::toUpper("el", UCASEMAP_OMIT_UNCHANGED_TEXT,
+ u"Πατάτα", 6, dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"toUpper(Πατάτα)", UnicodeString(u"ΑΤΑΤΑ"), UnicodeString(TRUE, dest, length));
+ static const EditChange upperExpectedChanges[] = {
+ { FALSE, 1, 1 },
+ { TRUE, 1, 1 },
+ { TRUE, 1, 1 },
+ { TRUE, 1, 1 },
+ { TRUE, 1, 1 },
+ { TRUE, 1, 1 }
+ };
+ checkEditsIter(u"toUpper(Πατάτα)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ upperExpectedChanges, UPRV_LENGTHOF(upperExpectedChanges),
+ TRUE, errorCode);
+
+ edits.reset();
+
+#if !UCONFIG_NO_BREAK_ITERATION
+ length = CaseMap::toTitle("nl",
+ UCASEMAP_OMIT_UNCHANGED_TEXT |
+ U_TITLECASE_NO_BREAK_ADJUSTMENT |
+ U_TITLECASE_NO_LOWERCASE,
+ NULL, u"IjssEL IglOo", 12,
+ dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"toTitle(IjssEL IglOo)", UnicodeString(u"J"), UnicodeString(TRUE, dest, length));
+ static const EditChange titleExpectedChanges[] = {
+ { FALSE, 1, 1 },
+ { TRUE, 1, 1 },
+ { FALSE, 10, 10 }
+ };
+ checkEditsIter(u"toTitle(IjssEL IglOo)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ titleExpectedChanges, UPRV_LENGTHOF(titleExpectedChanges),
+ TRUE, errorCode);
+#endif
+
+ edits.reset();
+ length = CaseMap::fold(UCASEMAP_OMIT_UNCHANGED_TEXT | U_FOLD_CASE_EXCLUDE_SPECIAL_I,
+ u"IßtanBul", 8, dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"foldCase(IßtanBul)", UnicodeString(u"ıssb"), UnicodeString(TRUE, dest, length));
+ static const EditChange foldExpectedChanges[] = {
+ { TRUE, 1, 1 },
+ { TRUE, 1, 2 },
+ { FALSE, 3, 3 },
+ { TRUE, 1, 1 },
+ { FALSE, 2, 2 }
+ };
+ checkEditsIter(u"foldCase(IßtanBul)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ foldExpectedChanges, UPRV_LENGTHOF(foldExpectedChanges),
+ TRUE, errorCode);
+}
+
+void StringCaseTest::TestCaseMapUTF8WithEdits() {
+ IcuTestErrorCode errorCode(*this, "TestEdits");
+ char dest[50];
+ Edits edits;
+
+ int32_t length = CaseMap::utf8ToLower("tr", UCASEMAP_OMIT_UNCHANGED_TEXT,
+ u8"IstanBul", 8, dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"toLower(IstanBul)", UnicodeString(u"ıb"),
+ UnicodeString::fromUTF8(StringPiece(dest, length)));
+ static const EditChange lowerExpectedChanges[] = {
+ { TRUE, 1, 2 },
+ { FALSE, 4, 4 },
+ { TRUE, 1, 1 },
+ { FALSE, 2, 2 }
+ };
+ checkEditsIter(u"toLower(IstanBul)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ lowerExpectedChanges, UPRV_LENGTHOF(lowerExpectedChanges),
+ TRUE, errorCode);
+
+ edits.reset();
+ length = CaseMap::utf8ToUpper("el", UCASEMAP_OMIT_UNCHANGED_TEXT,
+ u8"Πατάτα", 6 * 2, dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"toUpper(Πατάτα)", UnicodeString(u"ΑΤΑΤΑ"),
+ UnicodeString::fromUTF8(StringPiece(dest, length)));
+ static const EditChange upperExpectedChanges[] = {
+ { FALSE, 2, 2 },
+ { TRUE, 2, 2 },
+ { TRUE, 2, 2 },
+ { TRUE, 2, 2 },
+ { TRUE, 2, 2 },
+ { TRUE, 2, 2 }
+ };
+ checkEditsIter(u"toUpper(Πατάτα)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ upperExpectedChanges, UPRV_LENGTHOF(upperExpectedChanges),
+ TRUE, errorCode);
+
+ edits.reset();
+#if !UCONFIG_NO_BREAK_ITERATION
+ length = CaseMap::utf8ToTitle("nl",
+ UCASEMAP_OMIT_UNCHANGED_TEXT |
+ U_TITLECASE_NO_BREAK_ADJUSTMENT |
+ U_TITLECASE_NO_LOWERCASE,
+ NULL, u8"IjssEL IglOo", 12,
+ dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"toTitle(IjssEL IglOo)", UnicodeString(u"J"),
+ UnicodeString::fromUTF8(StringPiece(dest, length)));
+ static const EditChange titleExpectedChanges[] = {
+ { FALSE, 1, 1 },
+ { TRUE, 1, 1 },
+ { FALSE, 10, 10 }
+ };
+ checkEditsIter(u"toTitle(IjssEL IglOo)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ titleExpectedChanges, UPRV_LENGTHOF(titleExpectedChanges),
+ TRUE, errorCode);
+#endif
+
+ edits.reset();
+ length = CaseMap::utf8Fold(UCASEMAP_OMIT_UNCHANGED_TEXT | U_FOLD_CASE_EXCLUDE_SPECIAL_I,
+ u8"IßtanBul", 1 + 2 + 6, dest, UPRV_LENGTHOF(dest), &edits, errorCode);
+ assertEquals(u"foldCase(IßtanBul)", UnicodeString(u"ıssb"),
+ UnicodeString::fromUTF8(StringPiece(dest, length)));
+ static const EditChange foldExpectedChanges[] = {
+ { TRUE, 1, 2 },
+ { TRUE, 2, 2 },
+ { FALSE, 3, 3 },
+ { TRUE, 1, 1 },
+ { FALSE, 2, 2 }
+ };
+ checkEditsIter(u"foldCase(IßtanBul)",
+ edits.getFineIterator(), edits.getFineIterator(),
+ foldExpectedChanges, UPRV_LENGTHOF(foldExpectedChanges),
+ TRUE, errorCode);
+}
+
+void StringCaseTest::TestLongUnicodeString() {
+ // Code coverage for UnicodeString case mapping code handling
+ // long strings or many changes in a string.
+ UnicodeString s(TRUE,
+ (const UChar *)
+ u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF"
+ u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF"
+ u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF"
+ u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF"
+ u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF"
+ u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF", 6 * 51);
+ UnicodeString expected(TRUE,
+ (const UChar *)
+ u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF"
+ u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF"
+ u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF"
+ u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF"
+ u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF"
+ u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF", 6 * 51);
+ s.toUpper(Locale::getRoot());
+ assertEquals("string length 306", expected, s);
+}
+
+void StringCaseTest::TestBug13127() {
+ // Test case crashed when the bug was present.
+ const char16_t *s16 = u"日本語";
+ UnicodeString s(TRUE, s16, -1);
+ s.toTitle(0, Locale::getEnglish());
+}
projects / apple / icu.git / blobdiff