xref: /external/icu/icu4c/source/test/intltest/apicoll.cpp

//  2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/********************************************************************
 * COPYRIGHT:
 * Copyright (c) 1997-2016, International Business Machines Corporation and
 * others. All Rights Reserved.
 ********************************************************************/
//===============================================================================
//
// File apicoll.cpp
//
//
//
// Created by: Helena Shih
//
// Modification History:
//
//  Date         Name          Description
//  2/5/97      aliu        Added streamIn and streamOut methods.  Added
//                          constructor which reads RuleBasedCollator object from
//                          a binary file.  Added writeToFile method which streams
//                          RuleBasedCollator out to a binary file.  The streamIn
//                          and streamOut methods use istream and ostream objects
//                          in binary mode.
//  6/30/97     helena      Added tests for CollationElementIterator::setText, getOffset
//                          setOffset and DecompositionIterator::getOffset, setOffset.
//                          DecompositionIterator is made public so add class scope
//                          testing.
//  02/10/98    damiba      Added test for compare(UnicodeString&, UnicodeString&, int32_t)
//===============================================================================

#include "unicode/utypes.h"

#if !UCONFIG_NO_COLLATION

#include "unicode/localpointer.h"
#include "unicode/coll.h"
#include "unicode/tblcoll.h"
#include "unicode/coleitr.h"
#include "unicode/sortkey.h"
#include "apicoll.h"
#include "unicode/chariter.h"
#include "unicode/schriter.h"
#include "unicode/strenum.h"
#include "unicode/ustring.h"
#include "unicode/ucol.h"

#include "sfwdchit.h"
#include "cmemory.h"
#include <stdlib.h>

void
CollationAPITest::doAssert(UBool condition, const char *message)
{
    if (!condition) {
        errln(UnicodeString("ERROR : ") + message);
    }
}

// Collator Class Properties
// ctor, dtor, createInstance, compare, getStrength/setStrength
// getDecomposition/setDecomposition, getDisplayName
void
CollationAPITest::TestProperty(/* char* par */)
{
    UErrorCode success = U_ZERO_ERROR;
    Collator *col = 0;
    /*
     * Expected version of the English collator.
     * Currently, the major/minor version numbers change when the builder code
     * changes,
     * number 2 is from the tailoring data version and
     * number 3 is the UCA version.
     * This changes with every UCA version change, and the expected value
     * needs to be adjusted.
     * Same in cintltst/capitst.c.
     */
    UVersionInfo currVersionArray = {0x31, 0xC0, 0x05, 0x2A};  // from ICU 4.4/UCA 5.2
    UVersionInfo versionArray;

    logln("The property tests begin : ");
    logln("Test ctors : ");
    col = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success)){
        errcheckln(success, "English Collator creation failed. - %s", u_errorName(success));
        return;
    }

    col->getVersion(versionArray);
    // Check for a version greater than some value rather than equality
    // so that we need not update the expected version each time.
    if (uprv_memcmp(versionArray, currVersionArray, 4)<0) {
      errln("Testing Collator::getVersion() - unexpected result: %02x.%02x.%02x.%02x",
            versionArray[0], versionArray[1], versionArray[2], versionArray[3]);
    } else {
      logln("Collator::getVersion() result: %02x.%02x.%02x.%02x",
            versionArray[0], versionArray[1], versionArray[2], versionArray[3]);
    }

    doAssert((col->compare("ab", "abc") == Collator::LESS), "ab < abc comparison failed");
    doAssert((col->compare("ab", "AB") == Collator::LESS), "ab < AB comparison failed");
    doAssert((col->compare("blackbird", "black-bird") == Collator::GREATER), "black-bird > blackbird comparison failed");
    doAssert((col->compare("black bird", "black-bird") == Collator::LESS), "black bird > black-bird comparison failed");
    doAssert((col->compare("Hello", "hello") == Collator::GREATER), "Hello > hello comparison failed");
    doAssert((col->compare("","",success) == UCOL_EQUAL), "Comparison between empty strings failed");

    doAssert((col->compareUTF8("\x61\x62\xc3\xa4", "\x61\x62\xc3\x9f", success) == UCOL_LESS), "ab a-umlaut < ab sharp-s UTF-8 comparison failed");
    success = U_ZERO_ERROR;
    {
        UnicodeString abau=UNICODE_STRING_SIMPLE("\\x61\\x62\\xe4").unescape();
        UnicodeString abss=UNICODE_STRING_SIMPLE("\\x61\\x62\\xdf").unescape();
        UCharIterator abauIter, abssIter;
        uiter_setReplaceable(&abauIter, &abau);
        uiter_setReplaceable(&abssIter, &abss);
        doAssert((col->compare(abauIter, abssIter, success) == UCOL_LESS), "ab a-umlaut < ab sharp-s UCharIterator comparison failed");
        success = U_ZERO_ERROR;
    }

    /*start of update [Bertrand A. D. 02/10/98]*/
    doAssert((col->compare("ab", "abc", 2) == Collator::EQUAL), "ab = abc with length 2 comparison failed");
    doAssert((col->compare("ab", "AB", 2) == Collator::LESS), "ab < AB  with length 2 comparison failed");
    doAssert((col->compare("ab", "Aa", 1) == Collator::LESS), "ab < Aa  with length 1 comparison failed");
    doAssert((col->compare("ab", "Aa", 2) == Collator::GREATER), "ab > Aa  with length 2 comparison failed");
    doAssert((col->compare("black-bird", "blackbird", 5) == Collator::EQUAL), "black-bird = blackbird with length of 5 comparison failed");
    doAssert((col->compare("black bird", "black-bird", 10) == Collator::LESS), "black bird < black-bird with length 10 comparison failed");
    doAssert((col->compare("Hello", "hello", 5) == Collator::GREATER), "Hello > hello with length 5 comparison failed");
    /*end of update [Bertrand A. D. 02/10/98]*/


    logln("Test ctors ends.");
    logln("testing Collator::getStrength() method ...");
    doAssert((col->getStrength() == Collator::TERTIARY), "collation object has the wrong strength");
    doAssert((col->getStrength() != Collator::PRIMARY), "collation object's strength is primary difference");


    logln("testing Collator::setStrength() method ...");
    col->setStrength(Collator::SECONDARY);
    doAssert((col->getStrength() != Collator::TERTIARY), "collation object's strength is secondary difference");
    doAssert((col->getStrength() != Collator::PRIMARY), "collation object's strength is primary difference");
    doAssert((col->getStrength() == Collator::SECONDARY), "collation object has the wrong strength");

    UnicodeString name;

    logln("Get display name for the US English collation in German : ");
    logln(Collator::getDisplayName(Locale::getUS(), Locale::getGerman(), name));
    doAssert((name == UnicodeString("Englisch (Vereinigte Staaten)")), "getDisplayName failed");

    logln("Get display name for the US English collation in English : ");
    logln(Collator::getDisplayName(Locale::getUS(), Locale::getEnglish(), name));
    doAssert((name == UnicodeString("English (United States)")), "getDisplayName failed");
#if 0
    // weiv : this test is bogus if we're running on any machine that has different default locale than English.
    // Therefore, it is banned!
    logln("Get display name for the US English in default locale language : ");
    logln(Collator::getDisplayName(Locale::US, name));
    doAssert((name == UnicodeString("English (United States)")), "getDisplayName failed if this is an English machine");
#endif
    delete col; col = 0;
    RuleBasedCollator *rcol = (RuleBasedCollator *)Collator::createInstance("da_DK",
                                                                            success);
    if (U_FAILURE(success)) {
        errcheckln(success, "Collator::createInstance(\"da_DK\") failed - %s", u_errorName(success));
        return;
    }
    const UnicodeString &daRules = rcol->getRules();
    if(daRules.isEmpty()) {
        dataerrln("missing da_DK tailoring rule string");
    } else {
        doAssert(daRules.indexOf("aa") >= 0, "da_DK rules do not contain 'aa'");
    }
    delete rcol;

    col = Collator::createInstance(Locale::getFrench(), success);
    if (U_FAILURE(success))
    {
        errln("Creating French collation failed.");
        return;
    }

    col->setStrength(Collator::PRIMARY);
    logln("testing Collator::getStrength() method again ...");
    doAssert((col->getStrength() != Collator::TERTIARY), "collation object has the wrong strength");
    doAssert((col->getStrength() == Collator::PRIMARY), "collation object's strength is not primary difference");

    logln("testing French Collator::setStrength() method ...");
    col->setStrength(Collator::TERTIARY);
    doAssert((col->getStrength() == Collator::TERTIARY), "collation object's strength is not tertiary difference");
    doAssert((col->getStrength() != Collator::PRIMARY), "collation object's strength is primary difference");
    doAssert((col->getStrength() != Collator::SECONDARY), "collation object's strength is secondary difference");
    delete col;

    logln("Create junk collation: ");
    Locale abcd("ab", "CD", "");
    success = U_ZERO_ERROR;
    Collator *junk = 0;
    junk = Collator::createInstance(abcd, success);

    if (U_FAILURE(success))
    {
        errln("Junk collation creation failed, should at least return default.");
        return;
    }

    doAssert(((RuleBasedCollator *)junk)->getRules().isEmpty(),
               "The root collation should be returned for an unsupported language.");
    Collator *frCol = Collator::createInstance(Locale::getCanadaFrench(), success);
    if (U_FAILURE(success))
    {
        errln("Creating fr_CA collator failed.");
        delete junk;
        return;
    }

    // If the default locale isn't French, the French and non-French collators
    // should be different
    if (frCol->getLocale(ULOC_ACTUAL_LOCALE, success) != Locale::getCanadaFrench()) {
        doAssert((*frCol != *junk), "The junk is the same as the fr_CA collator.");
    }
    Collator *aFrCol = frCol->clone();
    doAssert((*frCol == *aFrCol), "The cloning of a fr_CA collator failed.");
    logln("Collator property test ended.");

    delete frCol;
    delete aFrCol;
    delete junk;
}

void CollationAPITest::TestKeywordValues() {
    IcuTestErrorCode errorCode(*this, "TestKeywordValues");
    LocalPointer<Collator> col(Collator::createInstance(Locale::getEnglish(), errorCode));
    if (errorCode.errIfFailureAndReset("English Collator creation failed")) {
        return;
    }

    LocalPointer<StringEnumeration> kwEnum(
        col->getKeywordValuesForLocale("collation", Locale::getEnglish(), TRUE, errorCode));
    if (errorCode.errIfFailureAndReset("Get Keyword Values for English Collator failed")) {
        return;
    }
    assertTrue("expect at least one collation tailoring for English", kwEnum->count(errorCode) > 0);
    const char *kw;
    UBool hasStandard = FALSE;
    while ((kw = kwEnum->next(NULL, errorCode)) != NULL) {
        if (strcmp(kw, "standard") == 0) {
            hasStandard = TRUE;
        }
    }
    assertTrue("expect at least the 'standard' collation tailoring for English", hasStandard);
}

void
CollationAPITest::TestRuleBasedColl()
{
    RuleBasedCollator *col1, *col2, *col3, *col4;
    UErrorCode status = U_ZERO_ERROR;

    UnicodeString ruleset1("&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E");
    UnicodeString ruleset2("&9 < a, A < b, B < c, C < d, D, e, E");

    col1 = new RuleBasedCollator(ruleset1, status);
    if (U_FAILURE(status)) {
        errcheckln(status, "RuleBased Collator creation failed. - %s", u_errorName(status));
        return;
    }
    else {
        logln("PASS: RuleBased Collator creation passed\n");
    }

    status = U_ZERO_ERROR;
    col2 = new RuleBasedCollator(ruleset2, status);
    if (U_FAILURE(status)) {
        errln("RuleBased Collator creation failed.\n");
        return;
    }
    else {
        logln("PASS: RuleBased Collator creation passed\n");
    }

    status = U_ZERO_ERROR;
    Locale locale("aa", "AA");
    col3 = (RuleBasedCollator *)Collator::createInstance(locale, status);
    if (U_FAILURE(status)) {
        errln("Fallback Collator creation failed.: %s\n");
        return;
    }
    else {
        logln("PASS: Fallback Collator creation passed\n");
    }
    delete col3;

    status = U_ZERO_ERROR;
    col3 = (RuleBasedCollator *)Collator::createInstance(status);
    if (U_FAILURE(status)) {
        errln("Default Collator creation failed.: %s\n");
        return;
    }
    else {
        logln("PASS: Default Collator creation passed\n");
    }

    UnicodeString rule1 = col1->getRules();
    UnicodeString rule2 = col2->getRules();
    UnicodeString rule3 = col3->getRules();

    doAssert(rule1 != rule2, "Default collator getRules failed");
    doAssert(rule2 != rule3, "Default collator getRules failed");
    doAssert(rule1 != rule3, "Default collator getRules failed");

    col4 = new RuleBasedCollator(rule2, status);
    if (U_FAILURE(status)) {
        errln("RuleBased Collator creation failed.\n");
        return;
    }

    UnicodeString rule4 = col4->getRules();
    doAssert(rule2 == rule4, "Default collator getRules failed");
    int32_t length4 = 0;
    uint8_t *clonedrule4 = col4->cloneRuleData(length4, status);
    if (U_FAILURE(status)) {
        errln("Cloned rule data failed.\n");
        return;
    }

 //   free(clonedrule4);     BAD API!!!!
    uprv_free(clonedrule4);


    delete col1;
    delete col2;
    delete col3;
    delete col4;
}

void
CollationAPITest::TestRules()
{
    RuleBasedCollator *coll;
    UErrorCode status = U_ZERO_ERROR;
    UnicodeString rules;

    coll = (RuleBasedCollator *)Collator::createInstance(Locale::getEnglish(), status);
    if (U_FAILURE(status)) {
        errcheckln(status, "English Collator creation failed. - %s", u_errorName(status));
        return;
    }
    else {
        logln("PASS: RuleBased Collator creation passed\n");
    }

    coll->getRules(UCOL_TAILORING_ONLY, rules);
    if (rules.length() != 0x00) {
      errln("English tailored rules failed - length is 0x%x expected 0x%x", rules.length(), 0x00);
    }

    coll->getRules(UCOL_FULL_RULES, rules);
    if (rules.length() < 0) {
        errln("English full rules failed");
    }
    delete coll;
}

void
CollationAPITest::TestDecomposition() {
  UErrorCode status = U_ZERO_ERROR;
  Collator *en_US = Collator::createInstance("en_US", status),
    *el_GR = Collator::createInstance("el_GR", status),
    *vi_VN = Collator::createInstance("vi_VN", status);

  if (U_FAILURE(status)) {
    errcheckln(status, "ERROR: collation creation failed. - %s", u_errorName(status));
    return;
  }

  /* there is no reason to have canonical decomposition in en_US OR default locale */
  if (vi_VN->getAttribute(UCOL_NORMALIZATION_MODE, status) != UCOL_ON)
  {
    errln("ERROR: vi_VN collation did not have canonical decomposition for normalization!\n");
  }

  if (el_GR->getAttribute(UCOL_NORMALIZATION_MODE, status) != UCOL_ON)
  {
    errln("ERROR: el_GR collation did not have canonical decomposition for normalization!\n");
  }

  if (en_US->getAttribute(UCOL_NORMALIZATION_MODE, status) != UCOL_OFF)
  {
    errln("ERROR: en_US collation had canonical decomposition for normalization!\n");
  }

  delete en_US;
  delete el_GR;
  delete vi_VN;
}

void
CollationAPITest::TestSafeClone() {
    static const int CLONETEST_COLLATOR_COUNT = 3;
    Collator *someCollators [CLONETEST_COLLATOR_COUNT];
    Collator *col;
    UErrorCode err = U_ZERO_ERROR;
    int index;

    UnicodeString test1("abCda");
    UnicodeString test2("abcda");

    /* one default collator & two complex ones */
    someCollators[0] = Collator::createInstance("en_US", err);
    someCollators[1] = Collator::createInstance("ko", err);
    someCollators[2] = Collator::createInstance("ja_JP", err);
    if(U_FAILURE(err)) {
      errcheckln(err, "Couldn't instantiate collators. Error: %s", u_errorName(err));
      delete someCollators[0];
      delete someCollators[1];
      delete someCollators[2];
      return;
    }

    /* change orig & clone & make sure they are independent */

    for (index = 0; index < CLONETEST_COLLATOR_COUNT; index++)
    {
        col = someCollators[index]->safeClone();
        if (col == 0) {
            errln("SafeClone of collator should not return null\n");
            break;
        }
        col->setStrength(Collator::TERTIARY);
        someCollators[index]->setStrength(Collator::PRIMARY);
        col->setAttribute(UCOL_CASE_LEVEL, UCOL_OFF, err);
        someCollators[index]->setAttribute(UCOL_CASE_LEVEL, UCOL_OFF, err);

        doAssert(col->greater(test1, test2), "Result should be \"abCda\" >>> \"abcda\" ");
        doAssert(someCollators[index]->equals(test1, test2), "Result should be \"abcda\" == \"abCda\"");
        delete col;
        delete someCollators[index];
    }
}

void
CollationAPITest::TestHashCode(/* char* par */)
{
    logln("hashCode tests begin.");
    UErrorCode success = U_ZERO_ERROR;
    Collator *col1 = 0;
    col1 = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success))
    {
        errcheckln(success, "Default collation creation failed. - %s", u_errorName(success));
        return;
    }

    Collator *col2 = 0;
    Locale dk("da", "DK", "");
    col2 = Collator::createInstance(dk, success);
    if (U_FAILURE(success))
    {
        errln("Danish collation creation failed.");
        return;
    }

    Collator *col3 = 0;
    col3 = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success))
    {
        errln("2nd default collation creation failed.");
        return;
    }

    logln("Collator::hashCode() testing ...");

    doAssert(col1->hashCode() != col2->hashCode(), "Hash test1 result incorrect" );
    doAssert(!(col1->hashCode() == col2->hashCode()), "Hash test2 result incorrect" );
    doAssert(col1->hashCode() == col3->hashCode(), "Hash result not equal" );

    logln("hashCode tests end.");
    delete col1;
    delete col2;

    UnicodeString test1("Abcda");
    UnicodeString test2("abcda");

    CollationKey sortk1, sortk2, sortk3;
    UErrorCode status = U_ZERO_ERROR;

    col3->getCollationKey(test1, sortk1, status);
    col3->getCollationKey(test2, sortk2, status);
    col3->getCollationKey(test2, sortk3, status);

    doAssert(sortk1.hashCode() != sortk2.hashCode(), "Hash test1 result incorrect");
    doAssert(sortk2.hashCode() == sortk3.hashCode(), "Hash result not equal" );

    delete col3;
}

//----------------------------------------------------------------------------
// CollationKey -- Tests the CollationKey methods
//
void
CollationAPITest::TestCollationKey(/* char* par */)
{
    logln("testing CollationKey begins...");
    Collator *col = 0;
    UErrorCode success=U_ZERO_ERROR;
    col = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success))
    {
        errcheckln(success, "Default collation creation failed. - %s", u_errorName(success));
        return;
    }
    col->setStrength(Collator::TERTIARY);

    CollationKey sortk1, sortk2;
    UnicodeString test1("Abcda"), test2("abcda");
    UErrorCode key1Status = U_ZERO_ERROR, key2Status = U_ZERO_ERROR;

    logln("Testing weird arguments");
    // No string vs. empty string vs. completely-ignorable string:
    // See ICU ticket #10495.
    CollationKey sortkNone;
    int32_t length;
    sortkNone.getByteArray(length);
    doAssert(!sortkNone.isBogus() && length == 0,
             "Default-constructed collation key should be empty");
    CollationKey sortkEmpty;
    col->getCollationKey(NULL, 0, sortkEmpty, key1Status);
    // key gets reset here
    const uint8_t* byteArrayEmpty = sortkEmpty.getByteArray(length);
    doAssert(sortkEmpty.isBogus() == FALSE && length == 3 &&
             byteArrayEmpty[0] == 1 && byteArrayEmpty[1] == 1 && byteArrayEmpty[2] == 0,
             "Empty string should return a collation key with empty levels");
    doAssert(sortkNone.compareTo(sortkEmpty) == Collator::LESS,
             "Expected no collation key < collation key for empty string");
    doAssert(sortkEmpty.compareTo(sortkNone) == Collator::GREATER,
             "Expected collation key for empty string > no collation key");

    CollationKey sortkIgnorable;
    // Most control codes and CGJ are completely ignorable.
    // A string with only completely ignorables must compare equal to an empty string.
    col->getCollationKey(UnicodeString((UChar)1).append((UChar)0x34f), sortkIgnorable, key1Status);
    sortkIgnorable.getByteArray(length);
    doAssert(!sortkIgnorable.isBogus() && length == 3,
             "Completely ignorable string should return a collation key with empty levels");
    doAssert(sortkIgnorable.compareTo(sortkEmpty) == Collator::EQUAL,
             "Completely ignorable string should compare equal to empty string");

    // bogus key returned here
    key1Status = U_ILLEGAL_ARGUMENT_ERROR;
    col->getCollationKey(NULL, 0, sortk1, key1Status);
    doAssert(sortk1.isBogus() && (sortk1.getByteArray(length), length) == 0,
        "Error code should return bogus collation key");

    key1Status = U_ZERO_ERROR;
    logln("Use tertiary comparison level testing ....");

    col->getCollationKey(test1, sortk1, key1Status);
    if (U_FAILURE(key1Status)) {
        errln("getCollationKey(Abcda) failed - %s", u_errorName(key1Status));
        return;
    }
    doAssert((sortk1.compareTo(col->getCollationKey(test2, sortk2, key2Status)))
                 == Collator::GREATER,
                "Result should be \"Abcda\" >>> \"abcda\"");

    CollationKey sortk3(sortk2), sortkNew;

    sortkNew = sortk1;
    doAssert((sortk1 != sortk2), "The sort keys should be different");
    doAssert((sortk1.hashCode() != sortk2.hashCode()), "sort key hashCode() failed");
    doAssert((sortk2 == sortk3), "The sort keys should be the same");
    doAssert((sortk1 == sortkNew), "The sort keys assignment failed");
    doAssert((sortk1.hashCode() == sortkNew.hashCode()), "sort key hashCode() failed");
    doAssert((sortkNew != sortk3), "The sort keys should be different");
    doAssert(sortk1.compareTo(sortk3) == Collator::GREATER, "Result should be \"Abcda\" >>> \"abcda\"");
    doAssert(sortk2.compareTo(sortk3) == Collator::EQUAL, "Result should be \"abcda\" == \"abcda\"");
    doAssert(sortkEmpty.compareTo(sortk1) == Collator::LESS, "Result should be (empty key) <<< \"Abcda\"");
    doAssert(sortk1.compareTo(sortkEmpty) == Collator::GREATER, "Result should be \"Abcda\" >>> (empty key)");
    doAssert(sortkEmpty.compareTo(sortkEmpty) == Collator::EQUAL, "Result should be (empty key) == (empty key)");
    doAssert(sortk1.compareTo(sortk3, success) == UCOL_GREATER, "Result should be \"Abcda\" >>> \"abcda\"");
    doAssert(sortk2.compareTo(sortk3, success) == UCOL_EQUAL, "Result should be \"abcda\" == \"abcda\"");
    doAssert(sortkEmpty.compareTo(sortk1, success) == UCOL_LESS, "Result should be (empty key) <<< \"Abcda\"");
    doAssert(sortk1.compareTo(sortkEmpty, success) == UCOL_GREATER, "Result should be \"Abcda\" >>> (empty key)");
    doAssert(sortkEmpty.compareTo(sortkEmpty, success) == UCOL_EQUAL, "Result should be (empty key) == (empty key)");

    int32_t    cnt1, cnt2, cnt3, cnt4;

    const uint8_t* byteArray1 = sortk1.getByteArray(cnt1);
    const uint8_t* byteArray2 = sortk2.getByteArray(cnt2);

    const uint8_t* byteArray3 = 0;
    byteArray3 = sortk1.getByteArray(cnt3);

    const uint8_t* byteArray4 = 0;
    byteArray4 = sortk2.getByteArray(cnt4);

    CollationKey sortk4(byteArray1, cnt1), sortk5(byteArray2, cnt2);
    CollationKey sortk6(byteArray3, cnt3), sortk7(byteArray4, cnt4);

    doAssert(sortk1.compareTo(sortk4) == Collator::EQUAL, "CollationKey::toByteArray(sortk1) Failed.");
    doAssert(sortk2.compareTo(sortk5) == Collator::EQUAL, "CollationKey::toByteArray(sortk2) Failed.");
    doAssert(sortk4.compareTo(sortk5) == Collator::GREATER, "sortk4 >>> sortk5 Failed");
    doAssert(sortk1.compareTo(sortk6) == Collator::EQUAL, "CollationKey::getByteArray(sortk1) Failed.");
    doAssert(sortk2.compareTo(sortk7) == Collator::EQUAL, "CollationKey::getByteArray(sortk2) Failed.");
    doAssert(sortk6.compareTo(sortk7) == Collator::GREATER, "sortk6 >>> sortk7 Failed");

    logln("Equality tests : ");
    doAssert(sortk1 == sortk4, "sortk1 == sortk4 Failed.");
    doAssert(sortk2 == sortk5, "sortk2 == sortk5 Failed.");
    doAssert(sortk1 != sortk5, "sortk1 != sortk5 Failed.");
    doAssert(sortk1 == sortk6, "sortk1 == sortk6 Failed.");
    doAssert(sortk2 == sortk7, "sortk2 == sortk7 Failed.");
    doAssert(sortk1 != sortk7, "sortk1 != sortk7 Failed.");

    byteArray1 = 0;
    byteArray2 = 0;

    sortk3 = sortk1;
    doAssert(sortk1 == sortk3, "sortk1 = sortk3 assignment Failed.");
    doAssert(sortk2 != sortk3, "sortk2 != sortk3 Failed.");
    logln("testing sortkey ends...");

    col->setStrength(Collator::SECONDARY);
    doAssert(col->getCollationKey(test1, sortk1, key1Status).compareTo(
                                  col->getCollationKey(test2, sortk2, key2Status))
                                  == Collator::EQUAL,
                                  "Result should be \"Abcda\" == \"abcda\"");
    delete col;
}

//----------------------------------------------------------------------------
// Tests the CollatorElementIterator class.
// ctor, RuleBasedCollator::createCollationElementIterator(), operator==, operator!=
//
void
CollationAPITest::TestElemIter(/* char* par */)
{
    logln("testing sortkey begins...");
    Collator *col = 0;
    UErrorCode success = U_ZERO_ERROR;
    col = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success))
    {
        errcheckln(success, "Default collation creation failed. - %s", u_errorName(success));
        return;
    }

    UnicodeString testString1("XFILE What subset of all possible test cases has the highest probability of detecting the most errors?");
    UnicodeString testString2("Xf_ile What subset of all possible test cases has the lowest probability of detecting the least errors?");
    logln("Constructors and comparison testing....");
    CollationElementIterator *iterator1 = ((RuleBasedCollator*)col)->createCollationElementIterator(testString1);

    CharacterIterator *chariter=new StringCharacterIterator(testString1);
    CollationElementIterator *coliter=((RuleBasedCollator*)col)->createCollationElementIterator(*chariter);

    // copy ctor
    CollationElementIterator *iterator2 = ((RuleBasedCollator*)col)->createCollationElementIterator(testString1);
    CollationElementIterator *iterator3 = ((RuleBasedCollator*)col)->createCollationElementIterator(testString2);

    int32_t offset = iterator1->getOffset();
    if (offset != 0) {
        errln("Error in getOffset for collation element iterator\n");
        return;
    }
    iterator1->setOffset(6, success);
    if (U_FAILURE(success)) {
        errln("Error in setOffset for collation element iterator\n");
        return;
    }
    iterator1->setOffset(0, success);
    int32_t order1, order2, order3;
    doAssert((*iterator1 == *iterator2), "The two iterators should be the same");
    doAssert((*iterator1 != *iterator3), "The two iterators should be different");

    doAssert((*coliter == *iterator1), "The two iterators should be the same");
    doAssert((*coliter == *iterator2), "The two iterators should be the same");
    doAssert((*coliter != *iterator3), "The two iterators should be different");

    order1 = iterator1->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((*iterator1 != *iterator2), "The first iterator advance failed");
    order2 = iterator2->getOffset();
    doAssert((order1 != order2), "The order result should not be the same");
    order2 = iterator2->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((*iterator1 == *iterator2), "The second iterator advance failed");
    doAssert((order1 == order2), "The order result should be the same");
    order3 = iterator3->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((CollationElementIterator::primaryOrder(order1) ==
        CollationElementIterator::primaryOrder(order3)), "The primary orders should be the same");
    doAssert((CollationElementIterator::secondaryOrder(order1) ==
        CollationElementIterator::secondaryOrder(order3)), "The secondary orders should be the same");
    doAssert((CollationElementIterator::tertiaryOrder(order1) ==
        CollationElementIterator::tertiaryOrder(order3)), "The tertiary orders should be the same");

    order1 = iterator1->next(success); order3 = iterator3->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((CollationElementIterator::primaryOrder(order1) ==
        CollationElementIterator::primaryOrder(order3)), "The primary orders should be identical");
    doAssert((CollationElementIterator::tertiaryOrder(order1) !=
        CollationElementIterator::tertiaryOrder(order3)), "The tertiary orders should be different");

    order1 = iterator1->next(success);
    order3 = iterator3->next(success);
    /* NO! Secondary orders of two CEs are not related, especially in the case of '_' vs 'I' */
    /*
    doAssert((CollationElementIterator::secondaryOrder(order1) !=
        CollationElementIterator::secondaryOrder(order3)), "The secondary orders should not be the same");
    */
    doAssert((order1 != CollationElementIterator::NULLORDER), "Unexpected end of iterator reached");

    iterator1->reset(); iterator2->reset(); iterator3->reset();
    order1 = iterator1->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((*iterator1 != *iterator2), "The first iterator advance failed");

    order2 = iterator2->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((*iterator1 == *iterator2), "The second iterator advance failed");
    doAssert((order1 == order2), "The order result should be the same");

    order3 = iterator3->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((CollationElementIterator::primaryOrder(order1) ==
        CollationElementIterator::primaryOrder(order3)), "The primary orders should be the same");
    doAssert((CollationElementIterator::secondaryOrder(order1) ==
        CollationElementIterator::secondaryOrder(order3)), "The secondary orders should be the same");
    doAssert((CollationElementIterator::tertiaryOrder(order1) ==
        CollationElementIterator::tertiaryOrder(order3)), "The tertiary orders should be the same");

    order1 = iterator1->next(success); order2 = iterator2->next(success); order3 = iterator3->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    doAssert((CollationElementIterator::primaryOrder(order1) ==
        CollationElementIterator::primaryOrder(order3)), "The primary orders should be identical");
    doAssert((CollationElementIterator::tertiaryOrder(order1) !=
        CollationElementIterator::tertiaryOrder(order3)), "The tertiary orders should be different");

    order1 = iterator1->next(success); order3 = iterator3->next(success);
    if (U_FAILURE(success))
    {
        errln("Somehow ran out of memory stepping through the iterator.");
        return;
    }

    /* NO! Secondary orders of two CEs are not related, especially in the case of '_' vs 'I' */
    /*
    doAssert((CollationElementIterator::secondaryOrder(order1) !=
        CollationElementIterator::secondaryOrder(order3)), "The secondary orders should not be the same");
    */
    doAssert((order1 != CollationElementIterator::NULLORDER), "Unexpected end of iterator reached");
    doAssert((*iterator2 != *iterator3), "The iterators should be different");


    //test error values
    success=U_UNSUPPORTED_ERROR;
    Collator *colerror=NULL;
    colerror=Collator::createInstance(Locale::getEnglish(), success);
    if (colerror != 0 || success == U_ZERO_ERROR){
        errln("Error: createInstance(UErrorCode != U_ZERO_ERROR) should just return and not create an instance\n");
    }
    int32_t position=coliter->previous(success);
    if(position != CollationElementIterator::NULLORDER){
        errln((UnicodeString)"Expected NULLORDER got" + position);
    }
    coliter->reset();
    coliter->setText(*chariter, success);
    if(!U_FAILURE(success)){
        errln("Expeceted error");
    }
    iterator1->setText((UnicodeString)"hello there", success);
    if(!U_FAILURE(success)){
        errln("Expeceted error");
    }

    delete chariter;
    delete coliter;
    delete iterator1;
    delete iterator2;
    delete iterator3;
    delete col;



    logln("testing CollationElementIterator ends...");
}

// Test RuleBasedCollator ctor, dtor, operator==, operator!=, clone, copy, and getRules
void
CollationAPITest::TestOperators(/* char* par */)
{
    UErrorCode success = U_ZERO_ERROR;
    UnicodeString ruleset1("&9 < a, A < b, B < c, C; ch, cH, Ch, CH < d, D, e, E");
    UnicodeString ruleset2("&9 < a, A < b, B < c, C < d, D, e, E");
    RuleBasedCollator *col1 = new RuleBasedCollator(ruleset1, success);
    if (U_FAILURE(success)) {
        errcheckln(success, "RuleBasedCollator creation failed. - %s", u_errorName(success));
        return;
    }
    success = U_ZERO_ERROR;
    RuleBasedCollator *col2 = new RuleBasedCollator(ruleset2, success);
    if (U_FAILURE(success)) {
        errln("The RuleBasedCollator constructor failed when building with the 2nd rule set.");
        return;
    }
    logln("The operator tests begin : ");
    logln("testing operator==, operator!=, clone  methods ...");
    doAssert((*col1 != *col2), "The two different table collations compared equal");
    *col1 = *col2;
    doAssert((*col1 == *col2), "Collator objects not equal after assignment (operator=)");

    success = U_ZERO_ERROR;
    Collator *col3 = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success)) {
        errln("Default collation creation failed.");
        return;
    }
    doAssert((*col1 != *col3), "The two different table collations compared equal");
    Collator* col4 = col1->clone();
    Collator* col5 = col3->clone();
    doAssert((*col1 == *col4), "Cloned collation objects not equal");
    doAssert((*col3 != *col4), "Two different table collations compared equal");
    doAssert((*col3 == *col5), "Cloned collation objects not equal");
    doAssert((*col4 != *col5), "Two cloned collations compared equal");

    const UnicodeString& defRules = ((RuleBasedCollator*)col3)->getRules();
    RuleBasedCollator* col6 = new RuleBasedCollator(defRules, success);
    if (U_FAILURE(success)) {
        errln("Creating default collation with rules failed.");
        return;
    }
    doAssert((((RuleBasedCollator*)col3)->getRules() == col6->getRules()), "Default collator getRules failed");

    success = U_ZERO_ERROR;
    RuleBasedCollator *col7 = new RuleBasedCollator(ruleset2, Collator::TERTIARY, success);
    if (U_FAILURE(success)) {
        errln("The RuleBasedCollator constructor failed when building with the 2nd rule set with tertiary strength.");
        return;
    }
    success = U_ZERO_ERROR;
    RuleBasedCollator *col8 = new RuleBasedCollator(ruleset2, UCOL_OFF, success);
    if (U_FAILURE(success)) {
        errln("The RuleBasedCollator constructor failed when building with the 2nd rule set with Normalizer::NO_OP.");
        return;
    }
    success = U_ZERO_ERROR;
    RuleBasedCollator *col9 = new RuleBasedCollator(ruleset2, Collator::PRIMARY, UCOL_ON, success);
    if (U_FAILURE(success)) {
        errln("The RuleBasedCollator constructor failed when building with the 2nd rule set with tertiary strength and Normalizer::NO_OP.");
        return;
    }
  //  doAssert((*col7 == *col8), "The two equal table collations compared different");
    doAssert((*col7 != *col9), "The two different table collations compared equal");
    doAssert((*col8 != *col9), "The two different table collations compared equal");

    logln("operator tests ended.");
    delete col1;
    delete col2;
    delete col3;
    delete col4;
    delete col5;
    delete col6;
    delete col7;
    delete col8;
    delete col9;
}

// test clone and copy
void
CollationAPITest::TestDuplicate(/* char* par */)
{
    UErrorCode status = U_ZERO_ERROR;
    Collator *col1 = Collator::createInstance(Locale::getEnglish(), status);
    if (U_FAILURE(status)) {
        logln("Default collator creation failed.");
        return;
    }
    Collator *col2 = col1->clone();
    doAssert((*col1 == *col2), "Cloned object is not equal to the orginal");
    UnicodeString ruleset("&9 < a, A < b, B < c, C < d, D, e, E");
    RuleBasedCollator *col3 = new RuleBasedCollator(ruleset, status);
    if (U_FAILURE(status)) {
        logln("Collation tailoring failed.");
        return;
    }
    doAssert((*col1 != *col3), "Cloned object is equal to some dummy");
    *col3 = *((RuleBasedCollator*)col1);
    doAssert((*col1 == *col3), "Copied object is not equal to the orginal");

    UCollationResult res;
    UnicodeString first((UChar)0x0061);
    UnicodeString second((UChar)0x0062);
    UnicodeString copiedEnglishRules(((RuleBasedCollator*)col1)->getRules());

    delete col1;

    // Try using the cloned collators after deleting the original data
    res = col2->compare(first, second, status);
    if(res != UCOL_LESS) {
        errln("a should be less then b after tailoring");
    }
    if (((RuleBasedCollator*)col2)->getRules() != copiedEnglishRules) {
        errln(UnicodeString("English rule difference. ")
            + copiedEnglishRules + UnicodeString("\ngetRules=") + ((RuleBasedCollator*)col2)->getRules());
    }
    res = col3->compare(first, second, status);
    if(res != UCOL_LESS) {
        errln("a should be less then b after tailoring");
    }
    if (col3->getRules() != copiedEnglishRules) {
        errln(UnicodeString("English rule difference. ")
            + copiedEnglishRules + UnicodeString("\ngetRules=") + col3->getRules());
    }

    delete col2;
    delete col3;
}

void
CollationAPITest::TestCompare(/* char* par */)
{
    logln("The compare tests begin : ");
    Collator *col = 0;
    UErrorCode success = U_ZERO_ERROR;
    col = Collator::createInstance(Locale::getEnglish(), success);
    if (U_FAILURE(success)) {
        errcheckln(success, "Default collation creation failed. - %s", u_errorName(success));
        return;
    }
    UnicodeString test1("Abcda"), test2("abcda");
    logln("Use tertiary comparison level testing ....");

    doAssert((!col->equals(test1, test2) ), "Result should be \"Abcda\" != \"abcda\"");
    doAssert((col->greater(test1, test2) ), "Result should be \"Abcda\" >>> \"abcda\"");
    doAssert((col->greaterOrEqual(test1, test2) ), "Result should be \"Abcda\" >>> \"abcda\"");

    col->setStrength(Collator::SECONDARY);
    logln("Use secondary comparison level testing ....");

    doAssert((col->equals(test1, test2) ), "Result should be \"Abcda\" == \"abcda\"");
    doAssert((!col->greater(test1, test2) ), "Result should be \"Abcda\" == \"abcda\"");
    doAssert((col->greaterOrEqual(test1, test2) ), "Result should be \"Abcda\" == \"abcda\"");

    col->setStrength(Collator::PRIMARY);
    logln("Use primary comparison level testing ....");

    doAssert((col->equals(test1, test2) ), "Result should be \"Abcda\" == \"abcda\"");
    doAssert((!col->greater(test1, test2) ), "Result should be \"Abcda\" == \"abcda\"");
    doAssert((col->greaterOrEqual(test1, test2) ), "Result should be \"Abcda\" == \"abcda\"");

    // Test different APIs
    const UChar* t1 = test1.getBuffer();
    int32_t t1Len = test1.length();
    const UChar* t2 = test2.getBuffer();
    int32_t t2Len = test2.length();

    doAssert((col->compare(test1, test2) == Collator::EQUAL), "Problem");
    doAssert((col->compare(test1, test2, success) == UCOL_EQUAL), "Problem");
    doAssert((col->compare(t1, t1Len, t2, t2Len) == Collator::EQUAL), "Problem");
    doAssert((col->compare(t1, t1Len, t2, t2Len, success) == UCOL_EQUAL), "Problem");
    doAssert((col->compare(test1, test2, t1Len) == Collator::EQUAL), "Problem");
    doAssert((col->compare(test1, test2, t1Len, success) == UCOL_EQUAL), "Problem");

    col->setAttribute(UCOL_STRENGTH, UCOL_TERTIARY, success);
    doAssert((col->compare(test1, test2) == Collator::GREATER), "Problem");
    doAssert((col->compare(test1, test2, success) == UCOL_GREATER), "Problem");
    doAssert((col->compare(t1, t1Len, t2, t2Len) == Collator::GREATER), "Problem");
    doAssert((col->compare(t1, t1Len, t2, t2Len, success) == UCOL_GREATER), "Problem");
    doAssert((col->compare(test1, test2, t1Len) == Collator::GREATER), "Problem");
    doAssert((col->compare(test1, test2, t1Len, success) == UCOL_GREATER), "Problem");



    logln("The compare tests end.");
    delete col;
}

void
CollationAPITest::TestGetAll(/* char* par */)
{
    int32_t count1, count2;
    UErrorCode status = U_ZERO_ERROR;

    logln("Trying Collator::getAvailableLocales(int&)");

    const Locale* list = Collator::getAvailableLocales(count1);
    for (int32_t i = 0; i < count1; ++i) {
        UnicodeString dispName;
        logln(UnicodeString("Locale name: ")
            + UnicodeString(list[i].getName())
            + UnicodeString(" , the display name is : ")
            + UnicodeString(list[i].getDisplayName(dispName)));
    }

    if (count1 == 0 || list == NULL) {
        dataerrln("getAvailableLocales(int&) returned an empty list");
    }

    logln("Trying Collator::getAvailableLocales()");
    StringEnumeration* localeEnum = Collator::getAvailableLocales();
    const UnicodeString* locStr;
    const char *locCStr;
    count2 = 0;

    if (localeEnum == NULL) {
        dataerrln("getAvailableLocales() returned NULL");
        return;
    }

    while ((locStr = localeEnum->snext(status)) != NULL)
    {
        logln(UnicodeString("Locale name is: ") + *locStr);
        count2++;
    }
    if (count1 != count2) {
        errln("getAvailableLocales(int&) returned %d and getAvailableLocales() returned %d", count1, count2);
    }

    logln("Trying Collator::getAvailableLocales() clone");
    count1 = 0;
    StringEnumeration* localeEnum2 = localeEnum->clone();
    localeEnum2->reset(status);
    while ((locCStr = localeEnum2->next(NULL, status)) != NULL)
    {
        logln(UnicodeString("Locale name is: ") + UnicodeString(locCStr));
        count1++;
    }
    if (count1 != count2) {
        errln("getAvailableLocales(3rd time) returned %d and getAvailableLocales(2nd time) returned %d", count1, count2);
    }
    if (localeEnum->count(status) != count1) {
        errln("localeEnum->count() returned %d and getAvailableLocales() returned %d", localeEnum->count(status), count1);
    }
    delete localeEnum;
    delete localeEnum2;
}

void CollationAPITest::TestSortKey()
{
    UErrorCode status = U_ZERO_ERROR;
    /*
    this is supposed to open default date format, but later on it treats
    it like it is "en_US"
    - very bad if you try to run the tests on machine where default
      locale is NOT "en_US"
    */
    Collator *col = Collator::createInstance(Locale::getEnglish(), status);
    if (U_FAILURE(status)) {
        errcheckln(status, "ERROR: Default collation creation failed.: %s\n", u_errorName(status));
        return;
    }

    if (col->getStrength() != Collator::TERTIARY)
    {
        errln("ERROR: default collation did not have UCOL_DEFAULT_STRENGTH !\n");
    }

    /* Need to use identical strength */
    col->setAttribute(UCOL_STRENGTH, UCOL_IDENTICAL, status);

    UChar test1[6] = {0x41, 0x62, 0x63, 0x64, 0x61, 0},
          test2[6] = {0x61, 0x62, 0x63, 0x64, 0x61, 0},
          test3[6] = {0x61, 0x62, 0x63, 0x64, 0x61, 0};

    uint8_t sortkey1[64];
    uint8_t sortkey2[64];
    uint8_t sortkey3[64];

    logln("Use tertiary comparison level testing ....\n");

    CollationKey key1;
    col->getCollationKey(test1, u_strlen(test1), key1, status);

    CollationKey key2;
    col->getCollationKey(test2, u_strlen(test2), key2, status);

    CollationKey key3;
    col->getCollationKey(test3, u_strlen(test3), key3, status);

    doAssert(key1.compareTo(key2) == Collator::GREATER,
        "Result should be \"Abcda\" > \"abcda\"");
    doAssert(key2.compareTo(key1) == Collator::LESS,
        "Result should be \"abcda\" < \"Abcda\"");
    doAssert(key2.compareTo(key3) == Collator::EQUAL,
        "Result should be \"abcda\" ==  \"abcda\"");

    // Clone the key2 sortkey for later.
    int32_t keylength = 0;
    const uint8_t *key2primary_alias = key2.getByteArray(keylength);
    LocalArray<uint8_t> key2primary(new uint8_t[keylength]);
    memcpy(key2primary.getAlias(), key2primary_alias, keylength);

    col->getSortKey(test1, sortkey1, 64);
    col->getSortKey(test2, sortkey2, 64);
    col->getSortKey(test3, sortkey3, 64);

    const uint8_t *tempkey = key1.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey1, keylength) == 0,
        "Test1 string should have the same collation key and sort key");
    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey2, keylength) == 0,
        "Test2 string should have the same collation key and sort key");
    tempkey = key3.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey3, keylength) == 0,
        "Test3 string should have the same collation key and sort key");

    col->getSortKey(test1, 5, sortkey1, 64);
    col->getSortKey(test2, 5, sortkey2, 64);
    col->getSortKey(test3, 5, sortkey3, 64);

    tempkey = key1.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey1, keylength) == 0,
        "Test1 string should have the same collation key and sort key");
    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey2, keylength) == 0,
        "Test2 string should have the same collation key and sort key");
    tempkey = key3.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey3, keylength) == 0,
        "Test3 string should have the same collation key and sort key");

    UnicodeString strtest1(test1);
    col->getSortKey(strtest1, sortkey1, 64);
    UnicodeString strtest2(test2);
    col->getSortKey(strtest2, sortkey2, 64);
    UnicodeString strtest3(test3);
    col->getSortKey(strtest3, sortkey3, 64);

    tempkey = key1.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey1, keylength) == 0,
        "Test1 string should have the same collation key and sort key");
    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey2, keylength) == 0,
        "Test2 string should have the same collation key and sort key");
    tempkey = key3.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey3, keylength) == 0,
        "Test3 string should have the same collation key and sort key");

    logln("Use secondary comparision level testing ...\n");
    col->setStrength(Collator::SECONDARY);

    col->getCollationKey(test1, u_strlen(test1), key1, status);
    col->getCollationKey(test2, u_strlen(test2), key2, status);
    col->getCollationKey(test3, u_strlen(test3), key3, status);

    doAssert(key1.compareTo(key2) == Collator::EQUAL,
        "Result should be \"Abcda\" == \"abcda\"");
    doAssert(key2.compareTo(key3) == Collator::EQUAL,
        "Result should be \"abcda\" ==  \"abcda\"");

    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, key2primary.getAlias(), keylength - 1) == 0,
             "Binary format for 'abcda' sortkey different for secondary strength!");

    col->getSortKey(test1, sortkey1, 64);
    col->getSortKey(test2, sortkey2, 64);
    col->getSortKey(test3, sortkey3, 64);

    tempkey = key1.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey1, keylength) == 0,
        "Test1 string should have the same collation key and sort key");
    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey2, keylength) == 0,
        "Test2 string should have the same collation key and sort key");
    tempkey = key3.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey3, keylength) == 0,
        "Test3 string should have the same collation key and sort key");

    col->getSortKey(test1, 5, sortkey1, 64);
    col->getSortKey(test2, 5, sortkey2, 64);
    col->getSortKey(test3, 5, sortkey3, 64);

    tempkey = key1.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey1, keylength) == 0,
        "Test1 string should have the same collation key and sort key");
    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey2, keylength) == 0,
        "Test2 string should have the same collation key and sort key");
    tempkey = key3.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey3, keylength) == 0,
        "Test3 string should have the same collation key and sort key");

    col->getSortKey(strtest1, sortkey1, 64);
    col->getSortKey(strtest2, sortkey2, 64);
    col->getSortKey(strtest3, sortkey3, 64);

    tempkey = key1.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey1, keylength) == 0,
        "Test1 string should have the same collation key and sort key");
    tempkey = key2.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey2, keylength) == 0,
        "Test2 string should have the same collation key and sort key");
    tempkey = key3.getByteArray(keylength);
    doAssert(memcmp(tempkey, sortkey3, keylength) == 0,
        "Test3 string should have the same collation key and sort key");

    logln("testing sortkey ends...");
    delete col;
}

void CollationAPITest::TestSortKeyOverflow() {
    IcuTestErrorCode errorCode(*this, "TestSortKeyOverflow()");
    LocalPointer<Collator> col(Collator::createInstance(Locale::getEnglish(), errorCode));
    if (errorCode.errDataIfFailureAndReset("Collator::createInstance(English) failed")) {
        return;
    }
    col->setAttribute(UCOL_STRENGTH, UCOL_PRIMARY, errorCode);
    UChar i_and_phi[] = { 0x438, 0x3c6 };  // Cyrillic small i & Greek small phi.
    // The sort key should be 6 bytes:
    // 2 bytes for the Cyrillic i, 1 byte for the primary-compression terminator,
    // 2 bytes for the Greek phi, and 1 byte for the NUL terminator.
    uint8_t sortKey[12];
    int32_t length = col->getSortKey(i_and_phi, 2, sortKey, UPRV_LENGTHOF(sortKey));
    uint8_t sortKey2[12];
    for (int32_t capacity = 0; capacity < length; ++capacity) {
        uprv_memset(sortKey2, 2, UPRV_LENGTHOF(sortKey2));
        int32_t length2 = col->getSortKey(i_and_phi, 2, sortKey2, capacity);
        if (length2 != length || 0 != uprv_memcmp(sortKey, sortKey2, capacity)) {
            errln("getSortKey(i_and_phi, capacity=%d) failed to write proper prefix", capacity);
        } else if (sortKey2[capacity] != 2 || sortKey2[capacity + 1] != 2) {
            errln("getSortKey(i_and_phi, capacity=%d) wrote beyond capacity", capacity);
        }
    }

    // Now try to break getCollationKey().
    // Internally, it always starts with a large stack buffer.
    // Since we cannot control the initial capacity, we throw an increasing number
    // of characters at it, with the problematic part at the end.
    const int32_t longCapacity = 2000;
    // Each 'a' in the prefix should result in one primary sort key byte.
    // For i_and_phi we expect 6 bytes, then the NUL terminator.
    const int32_t maxPrefixLength = longCapacity - 6 - 1;
    LocalArray<uint8_t> longSortKey(new uint8_t[longCapacity]);
    UnicodeString s(FALSE, i_and_phi, 2);
    for (int32_t prefixLength = 0; prefixLength < maxPrefixLength; ++prefixLength) {
        length = col->getSortKey(s, longSortKey.getAlias(), longCapacity);
        CollationKey collKey;
        col->getCollationKey(s, collKey, errorCode);
        int32_t collKeyLength;
        const uint8_t *collSortKey = collKey.getByteArray(collKeyLength);
        if (collKeyLength != length || 0 != uprv_memcmp(longSortKey.getAlias(), collSortKey, length)) {
            errln("getCollationKey(prefix[%d]+i_and_phi) failed to write proper sort key", prefixLength);
        }

        // Insert an 'a' to match ++prefixLength.
        s.insert(prefixLength, (UChar)0x61);
    }
}

void CollationAPITest::TestMaxExpansion()
{
    UErrorCode          status = U_ZERO_ERROR;
    UChar               ch     = 0;
    UChar32             unassigned = 0xEFFFD;
    uint32_t            sorder = 0;
    uint32_t            temporder = 0;

    UnicodeString rule("&a < ab < c/aba < d < z < ch");
    RuleBasedCollator coll(rule, status);
    if(U_FAILURE(status)) {
      errcheckln(status, "Collator creation failed with error %s", u_errorName(status));
      return;
    }
    UnicodeString str(ch);
    CollationElementIterator *iter =
                                  coll.createCollationElementIterator(str);

    while (ch < 0xFFFF && U_SUCCESS(status)) {
        int      count = 1;
        uint32_t order;
        int32_t  size = 0;

        ch ++;

        str.setCharAt(0, ch);
        iter->setText(str, status);
        order = iter->previous(status);

        /* thai management */
        if (order == 0)
            order = iter->previous(status);

        while (U_SUCCESS(status) && iter->previous(status) != CollationElementIterator::NULLORDER) {
            count ++;
        }

        size = coll.getMaxExpansion(order);
        if (U_FAILURE(status) || size < count) {
            errln("Failure at codepoint U+%04X, maximum expansion count %d < %d",
                  ch, size, count);
        }
    }

    /* testing for exact max expansion */
    int32_t size;
    ch = 0;
    while (ch < 0x61) {
        uint32_t order;
        str.setCharAt(0, ch);
        iter->setText(str, status);
        order = iter->previous(status);
        size  = coll.getMaxExpansion(order);
        if (U_FAILURE(status) || size != 1) {
            errln("Failure at codepoint U+%04X, maximum expansion count %d < %d",
                  ch, size, 1);
        }
        ch ++;
    }

    ch = 0x63;
    str.setTo(ch);
    iter->setText(str, status);
    temporder = iter->previous(status);
    size = coll.getMaxExpansion(temporder);
    if (U_FAILURE(status) || size != 3) {
        errln("Failure at codepoint U+%04X, CE %08x, maximum expansion count %d != %d",
              ch, temporder, size, 3);
    }

    ch = 0x64;
    str.setTo(ch);
    iter->setText(str, status);
    temporder = iter->previous(status);
    size = coll.getMaxExpansion(temporder);
    if (U_FAILURE(status) || size != 1) {
        errln("Failure at codepoint U+%04X, CE %08x, maximum expansion count %d != %d",
              ch, temporder, size, 1);
    }

    str.setTo(unassigned);
    iter->setText(str, status);
    sorder = iter->previous(status);
    size = coll.getMaxExpansion(sorder);
    if (U_FAILURE(status) || size != 2) {
        errln("Failure at supplementary codepoints, maximum expansion count %d < %d",
              size, 2);
    }

    /* testing jamo */
    ch = 0x1165;
    str.setTo(ch);
    iter->setText(str, status);
    temporder = iter->previous(status);
    size = coll.getMaxExpansion(temporder);
    if (U_FAILURE(status) || size > 3) {
        errln("Failure at codepoint U+%04X, maximum expansion count %d > %d",
              ch, size, 3);
    }

    delete iter;

    /* testing special jamo &a<\u1160 */
    rule = CharsToUnicodeString("\\u0026\\u0071\\u003c\\u1165\\u002f\\u0071\\u0071\\u0071\\u0071");

    RuleBasedCollator jamocoll(rule, status);
    iter = jamocoll.createCollationElementIterator(str);
    temporder = iter->previous(status);
    size = iter->getMaxExpansion(temporder);
    if (U_FAILURE(status) || size != 6) {
        errln("Failure at codepoint U+%04X, maximum expansion count %d > %d",
              ch, size, 5);
    }

    delete iter;
}

void CollationAPITest::TestDisplayName()
{
    UErrorCode error = U_ZERO_ERROR;
    Collator *coll = Collator::createInstance("en_US", error);
    if (U_FAILURE(error)) {
        errcheckln(error, "Failure creating english collator - %s", u_errorName(error));
        return;
    }
    UnicodeString name;
    UnicodeString result;
    coll->getDisplayName(Locale::getCanadaFrench(), result);
    Locale::getCanadaFrench().getDisplayName(name);
    if (result.compare(name)) {
        errln("Failure getting the correct name for locale en_US");
    }

    coll->getDisplayName(Locale::getSimplifiedChinese(), result);
    Locale::getSimplifiedChinese().getDisplayName(name);
    if (result.compare(name)) {
        errln("Failure getting the correct name for locale zh_SG");
    }
    delete coll;
}

void CollationAPITest::TestAttribute()
{
    UErrorCode error = U_ZERO_ERROR;
    Collator *coll = Collator::createInstance(error);

    if (U_FAILURE(error)) {
        errcheckln(error, "Creation of default collator failed - %s", u_errorName(error));
        return;
    }

    coll->setAttribute(UCOL_FRENCH_COLLATION, UCOL_OFF, error);
    if (coll->getAttribute(UCOL_FRENCH_COLLATION, error) != UCOL_OFF ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the french collation failed");
    }

    coll->setAttribute(UCOL_FRENCH_COLLATION, UCOL_ON, error);
    if (coll->getAttribute(UCOL_FRENCH_COLLATION, error) != UCOL_ON ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the french collation failed");
    }

    coll->setAttribute(UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, error);
    if (coll->getAttribute(UCOL_ALTERNATE_HANDLING, error) != UCOL_SHIFTED ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the alternate handling failed");
    }

    coll->setAttribute(UCOL_ALTERNATE_HANDLING, UCOL_NON_IGNORABLE, error);
    if (coll->getAttribute(UCOL_ALTERNATE_HANDLING, error) != UCOL_NON_IGNORABLE ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the alternate handling failed");
    }

    coll->setAttribute(UCOL_CASE_FIRST, UCOL_LOWER_FIRST, error);
    if (coll->getAttribute(UCOL_CASE_FIRST, error) != UCOL_LOWER_FIRST ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the case first attribute failed");
    }

    coll->setAttribute(UCOL_CASE_FIRST, UCOL_UPPER_FIRST, error);
    if (coll->getAttribute(UCOL_CASE_FIRST, error) != UCOL_UPPER_FIRST ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the case first attribute failed");
    }

    coll->setAttribute(UCOL_CASE_LEVEL, UCOL_ON, error);
    if (coll->getAttribute(UCOL_CASE_LEVEL, error) != UCOL_ON ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the case level attribute failed");
    }

    coll->setAttribute(UCOL_CASE_LEVEL, UCOL_OFF, error);
    if (coll->getAttribute(UCOL_CASE_LEVEL, error) != UCOL_OFF ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the case level attribute failed");
    }

    coll->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_ON, error);
    if (coll->getAttribute(UCOL_NORMALIZATION_MODE, error) != UCOL_ON ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the normalization on/off attribute failed");
    }

    coll->setAttribute(UCOL_NORMALIZATION_MODE, UCOL_OFF, error);
    if (coll->getAttribute(UCOL_NORMALIZATION_MODE, error) != UCOL_OFF ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the normalization on/off attribute failed");
    }

    coll->setAttribute(UCOL_STRENGTH, UCOL_PRIMARY, error);
    if (coll->getAttribute(UCOL_STRENGTH, error) != UCOL_PRIMARY ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the collation strength failed");
    }

    coll->setAttribute(UCOL_STRENGTH, UCOL_SECONDARY, error);
    if (coll->getAttribute(UCOL_STRENGTH, error) != UCOL_SECONDARY ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the collation strength failed");
    }

    coll->setAttribute(UCOL_STRENGTH, UCOL_TERTIARY, error);
    if (coll->getAttribute(UCOL_STRENGTH, error) != UCOL_TERTIARY ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the collation strength failed");
    }

    coll->setAttribute(UCOL_STRENGTH, UCOL_QUATERNARY, error);
    if (coll->getAttribute(UCOL_STRENGTH, error) != UCOL_QUATERNARY ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the collation strength failed");
    }

    coll->setAttribute(UCOL_STRENGTH, UCOL_IDENTICAL, error);
    if (coll->getAttribute(UCOL_STRENGTH, error) != UCOL_IDENTICAL ||
        U_FAILURE(error)) {
        errln("Setting and retrieving of the collation strength failed");
    }

    delete coll;
}

void CollationAPITest::TestVariableTopSetting() {
  UErrorCode status = U_ZERO_ERROR;

  UChar vt[256] = { 0 };

  // Use the root collator, not the default collator.
  // This test fails with en_US_POSIX which tailors the dollar sign after 'A'.
  Collator *coll = Collator::createInstance(Locale::getRoot(), status);
  if(U_FAILURE(status)) {
    delete coll;
    errcheckln(status, "Collator creation failed with error %s", u_errorName(status));
    return;
  }

  uint32_t oldVarTop = coll->getVariableTop(status);

  // ICU 53+: The character must be in a supported reordering group,
  // and the variable top is pinned to the end of that group.
  vt[0] = 0x0041;

  (void)coll->setVariableTop(vt, 1, status);
  if(status != U_ILLEGAL_ARGUMENT_ERROR) {
    errln("setVariableTop(letter) did not detect illegal argument - %s", u_errorName(status));
  }

  status = U_ZERO_ERROR;
  vt[0] = 0x24;  // dollar sign (currency symbol)
  uint32_t newVarTop = coll->setVariableTop(vt, 1, status);
  if(U_FAILURE(status)) {
    errln("setVariableTop(dollar sign) failed: %s", u_errorName(status));
    return;
  }
  if(newVarTop != coll->getVariableTop(status)) {
    errln("setVariableTop(dollar sign) != following getVariableTop()");
  }

  UnicodeString dollar((UChar)0x24);
  UnicodeString euro((UChar)0x20AC);
  uint32_t newVarTop2 = coll->setVariableTop(euro, status);
  assertEquals("setVariableTop(Euro sign) == following getVariableTop()",
               (int64_t)newVarTop2, (int64_t)coll->getVariableTop(status));
  assertEquals("setVariableTop(Euro sign) == setVariableTop(dollar sign) (should pin to top of currency group)",
               (int64_t)newVarTop2, (int64_t)newVarTop);

  coll->setAttribute(UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, status);
  assertEquals("empty==dollar", (int32_t)UCOL_EQUAL, (int32_t)coll->compare(UnicodeString(), dollar));
  assertEquals("empty==euro", (int32_t)UCOL_EQUAL, (int32_t)coll->compare(UnicodeString(), euro));
  assertEquals("dollar<zero", (int32_t)UCOL_LESS, (int32_t)coll->compare(dollar, UnicodeString((UChar)0x30)));

  coll->setVariableTop(oldVarTop, status);

  uint32_t newerVarTop = coll->setVariableTop(UnicodeString(vt, 1), status);

  if(newVarTop != newerVarTop) {
    errln("Didn't set vartop properly from UnicodeString!\n");
  }

  delete coll;

}

void CollationAPITest::TestMaxVariable() {
  UErrorCode errorCode = U_ZERO_ERROR;
  LocalPointer<Collator> coll(Collator::createInstance(Locale::getRoot(), errorCode));
  if(U_FAILURE(errorCode)) {
    errcheckln(errorCode, "Collator creation failed with error %s", u_errorName(errorCode));
    return;
  }

  (void)coll->setMaxVariable(UCOL_REORDER_CODE_OTHERS, errorCode);
  if(errorCode != U_ILLEGAL_ARGUMENT_ERROR) {
    errln("setMaxVariable(others) did not detect illegal argument - %s", u_errorName(errorCode));
  }

  errorCode = U_ZERO_ERROR;
  (void)coll->setMaxVariable(UCOL_REORDER_CODE_CURRENCY, errorCode);

  if(UCOL_REORDER_CODE_CURRENCY != coll->getMaxVariable()) {
    errln("setMaxVariable(currency) != following getMaxVariable()");
  }

  coll->setAttribute(UCOL_ALTERNATE_HANDLING, UCOL_SHIFTED, errorCode);
  assertEquals("empty==dollar", (int32_t)UCOL_EQUAL, (int32_t)coll->compare(UnicodeString(), UnicodeString((UChar)0x24)));
  assertEquals("empty==euro", (int32_t)UCOL_EQUAL, (int32_t)coll->compare(UnicodeString(), UnicodeString((UChar)0x20AC)));
  assertEquals("dollar<zero", (int32_t)UCOL_LESS, (int32_t)coll->compare(UnicodeString((UChar)0x24), UnicodeString((UChar)0x30)));
}

void CollationAPITest::TestGetLocale() {
  UErrorCode status = U_ZERO_ERROR;
  const char *rules = "&a<x<y<z";
  UChar rlz[256] = {0};

  Collator *coll = Collator::createInstance("root", status);
  if(U_FAILURE(status)) {
    dataerrln("Failed to open collator for \"root\" with %s", u_errorName(status));
    return;
  }
  Locale locale = coll->getLocale(ULOC_ACTUAL_LOCALE, status);
  if(locale != Locale::getRoot()) {
    errln("Collator::createInstance(\"root\").getLocale(actual) != Locale::getRoot(); "
          "getLocale().getName() = \"%s\"",
          locale.getName());
  }
  delete coll;

  coll = Collator::createInstance("", status);
  if(U_FAILURE(status)) {
    dataerrln("Failed to open collator for \"\" with %s", u_errorName(status));
    return;
  }
  locale = coll->getLocale(ULOC_ACTUAL_LOCALE, status);
  if(locale != Locale::getRoot()) {
    errln("Collator::createInstance(\"\").getLocale(actual) != Locale::getRoot(); "
          "getLocale().getName() = \"%s\"",
          locale.getName());
  }
  delete coll;

  int32_t i = 0;

  static const struct {
    const char* requestedLocale;
    const char* validLocale;
    const char* actualLocale;
  } testStruct[] = {
    // Note: Locale::getRoot().getName() == "" not "root".
    { "de_DE", "de", "" },
    { "sr_RS", "sr_Cyrl_RS", "sr" },
    { "en_US_CALIFORNIA", "en_US", "" },
    { "fr_FR_NONEXISTANT", "fr", "" },
    // pinyin is the default, therefore suppressed.
    { "zh_CN", "zh_Hans_CN", "zh" },
    // zh_Hant has default=stroke but the data is in zh.
    { "zh_TW", "zh_Hant_TW", "zh@collation=stroke" },
    { "zh_TW@collation=pinyin", "zh_Hant_TW@collation=pinyin", "zh" },
    { "zh_CN@collation=stroke", "zh_Hans_CN@collation=stroke", "zh@collation=stroke" },
    // yue/yue_Hant aliased to zh_Hant, yue_Hans aliased to zh_Hans.
    { "yue", "zh_Hant", "zh@collation=stroke" },
    { "yue_HK", "zh_Hant", "zh@collation=stroke" },
    { "yue_Hant", "zh_Hant", "zh@collation=stroke" },
    { "yue_Hant_HK", "zh_Hant", "zh@collation=stroke" },
    { "yue@collation=pinyin", "zh_Hant@collation=pinyin", "zh" },
    { "yue_HK@collation=pinyin", "zh_Hant@collation=pinyin", "zh" },
    { "yue_CN", "zh_Hans", "zh" },
    { "yue_Hans", "zh_Hans", "zh" },
    { "yue_Hans_CN", "zh_Hans", "zh" },
    { "yue_Hans@collation=stroke", "zh_Hans@collation=stroke", "zh@collation=stroke" },
    { "yue_CN@collation=stroke", "zh_Hans@collation=stroke", "zh@collation=stroke" }
  };

  u_unescape(rules, rlz, 256);

  /* test opening collators for different locales */
  for(i = 0; i<UPRV_LENGTHOF(testStruct); i++) {
    status = U_ZERO_ERROR;
    coll = Collator::createInstance(testStruct[i].requestedLocale, status);
    if(U_FAILURE(status)) {
      errln("Failed to open collator for %s with %s", testStruct[i].requestedLocale, u_errorName(status));
      delete coll;
      continue;
    }
    // The requested locale may be the same as the valid locale,
    // or may not be supported at all. See ticket #10477.
    locale = coll->getLocale(ULOC_REQUESTED_LOCALE, status);
    if(U_SUCCESS(status) &&
        locale != testStruct[i].requestedLocale && locale != testStruct[i].validLocale) {
      errln("[Coll %s]: Error in requested locale, expected %s or %s, got %s",
            testStruct[i].requestedLocale,
            testStruct[i].requestedLocale, testStruct[i].validLocale, locale.getName());
    }
    status = U_ZERO_ERROR;
    locale = coll->getLocale(ULOC_VALID_LOCALE, status);
    if(locale != testStruct[i].validLocale) {
      errln("[Coll %s]: Error in valid locale, expected %s, got %s",
            testStruct[i].requestedLocale, testStruct[i].validLocale, locale.getName());
    }
    locale = coll->getLocale(ULOC_ACTUAL_LOCALE, status);
    if(locale != testStruct[i].actualLocale) {
      errln("[Coll %s]: Error in actual locale, expected %s, got %s",
            testStruct[i].requestedLocale, testStruct[i].actualLocale, locale.getName());
    }
    // If we open a collator for the actual locale, we should get an equivalent one again.
    LocalPointer<Collator> coll2(Collator::createInstance(locale, status));
    if(U_FAILURE(status)) {
      errln("Failed to open collator for actual locale \"%s\" with %s",
            locale.getName(), u_errorName(status));
    } else {
      Locale actual2 = coll2->getLocale(ULOC_ACTUAL_LOCALE, status);
      if(actual2 != locale) {
        errln("[Coll actual \"%s\"]: Error in actual locale, got different one: \"%s\"",
              locale.getName(), actual2.getName());
      }
      if(*coll2 != *coll) {
        errln("[Coll actual \"%s\"]: Got different collator than before", locale.getName());
      }
    }
    delete coll;
  }

  /* completely non-existent locale for collator should get a root collator */
  {
    LocalPointer<Collator> coll(Collator::createInstance("blahaha", status));
    if(U_FAILURE(status)) {
      errln("Failed to open collator with %s", u_errorName(status));
      return;
    }
    Locale valid = coll->getLocale(ULOC_VALID_LOCALE, status);
    const char *name = valid.getName();
    if(*name != 0 && strcmp(name, "root") != 0) {
      errln("Valid locale for nonexisting-locale collator is \"%s\" not root", name);
    }
    Locale actual = coll->getLocale(ULOC_ACTUAL_LOCALE, status);
    name = actual.getName();
    if(*name != 0 && strcmp(name, "root") != 0) {
      errln("Actual locale for nonexisting-locale collator is \"%s\" not root", name);
    }
  }



  /* collator instantiated from rules should have all three locales NULL */
  coll = new RuleBasedCollator(rlz, status);
  locale = coll->getLocale(ULOC_REQUESTED_LOCALE, status);
  if(U_SUCCESS(status) && !locale.isBogus()) {
    errln("For collator instantiated from rules, requested locale %s is not bogus", locale.getName());
  }
  status = U_ZERO_ERROR;
  locale = coll->getLocale(ULOC_VALID_LOCALE, status);
  if(!locale.isBogus()) {
    errln("For collator instantiated from rules, valid locale %s is not bogus", locale.getName());
  }
  locale = coll->getLocale(ULOC_ACTUAL_LOCALE, status);
  if(!locale.isBogus()) {
    errln("For collator instantiated from rules, actual locale %s is not bogus", locale.getName());
  }
  delete coll;
}

struct teststruct {
    const char *original;
    uint8_t key[256];
};



U_CDECL_BEGIN
static int U_CALLCONV
compare_teststruct(const void *string1, const void *string2) {
  return(strcmp((const char *)((struct teststruct *)string1)->key, (const char *)((struct teststruct *)string2)->key));
}
U_CDECL_END


void CollationAPITest::TestBounds(void) {
    UErrorCode status = U_ZERO_ERROR;

    Collator *coll = Collator::createInstance(Locale("sh"), status);
    if(U_FAILURE(status)) {
      delete coll;
      errcheckln(status, "Collator creation failed with %s", u_errorName(status));
      return;
    }

    uint8_t sortkey[512], lower[512], upper[512];
    UChar buffer[512];

    static const char * const test[] = {
        "John Smith",
        "JOHN SMITH",
        "john SMITH",
        "j\\u00F6hn sm\\u00EFth",
        "J\\u00F6hn Sm\\u00EFth",
        "J\\u00D6HN SM\\u00CFTH",
        "john smithsonian",
        "John Smithsonian"
    };

    struct teststruct tests[] = {
        {"\\u010CAKI MIHALJ", {0}},
        {"\\u010CAKI MIHALJ", {0}},
        {"\\u010CAKI PIRO\\u0160KA", {0}},
        {"\\u010CABAI ANDRIJA", {0}},
        {"\\u010CABAI LAJO\\u0160", {0}},
        {"\\u010CABAI MARIJA", {0}},
        {"\\u010CABAI STEVAN", {0}},
        {"\\u010CABAI STEVAN", {0}},
        {"\\u010CABARKAPA BRANKO", {0}},
        {"\\u010CABARKAPA MILENKO", {0}},
        {"\\u010CABARKAPA MIROSLAV", {0}},
        {"\\u010CABARKAPA SIMO", {0}},
        {"\\u010CABARKAPA STANKO", {0}},
        {"\\u010CABARKAPA TAMARA", {0}},
        {"\\u010CABARKAPA TOMA\\u0160", {0}},
        {"\\u010CABDARI\\u0106 NIKOLA", {0}},
        {"\\u010CABDARI\\u0106 ZORICA", {0}},
        {"\\u010CABI NANDOR", {0}},
        {"\\u010CABOVI\\u0106 MILAN", {0}},
        {"\\u010CABRADI AGNEZIJA", {0}},
        {"\\u010CABRADI IVAN", {0}},
        {"\\u010CABRADI JELENA", {0}},
        {"\\u010CABRADI LJUBICA", {0}},
        {"\\u010CABRADI STEVAN", {0}},
        {"\\u010CABRDA MARTIN", {0}},
        {"\\u010CABRILO BOGDAN", {0}},
        {"\\u010CABRILO BRANISLAV", {0}},
        {"\\u010CABRILO LAZAR", {0}},
        {"\\u010CABRILO LJUBICA", {0}},
        {"\\u010CABRILO SPASOJA", {0}},
        {"\\u010CADE\\u0160 ZDENKA", {0}},
        {"\\u010CADESKI BLAGOJE", {0}},
        {"\\u010CADOVSKI VLADIMIR", {0}},
        {"\\u010CAGLJEVI\\u0106 TOMA", {0}},
        {"\\u010CAGOROVI\\u0106 VLADIMIR", {0}},
        {"\\u010CAJA VANKA", {0}},
        {"\\u010CAJI\\u0106 BOGOLJUB", {0}},
        {"\\u010CAJI\\u0106 BORISLAV", {0}},
        {"\\u010CAJI\\u0106 RADOSLAV", {0}},
        {"\\u010CAK\\u0160IRAN MILADIN", {0}},
        {"\\u010CAKAN EUGEN", {0}},
        {"\\u010CAKAN EVGENIJE", {0}},
        {"\\u010CAKAN IVAN", {0}},
        {"\\u010CAKAN JULIJAN", {0}},
        {"\\u010CAKAN MIHAJLO", {0}},
        {"\\u010CAKAN STEVAN", {0}},
        {"\\u010CAKAN VLADIMIR", {0}},
        {"\\u010CAKAN VLADIMIR", {0}},
        {"\\u010CAKAN VLADIMIR", {0}},
        {"\\u010CAKARA ANA", {0}},
        {"\\u010CAKAREVI\\u0106 MOMIR", {0}},
        {"\\u010CAKAREVI\\u0106 NEDELJKO", {0}},
        {"\\u010CAKI \\u0160ANDOR", {0}},
        {"\\u010CAKI AMALIJA", {0}},
        {"\\u010CAKI ANDRA\\u0160", {0}},
        {"\\u010CAKI LADISLAV", {0}},
        {"\\u010CAKI LAJO\\u0160", {0}},
        {"\\u010CAKI LASLO", {0}}
    };



    int32_t i = 0, j = 0, k = 0, buffSize = 0, skSize = 0, lowerSize = 0, upperSize = 0;
    int32_t arraySize = UPRV_LENGTHOF(tests);

    (void)lowerSize;  // Suppress unused variable warnings.
    (void)upperSize;

    for(i = 0; i<arraySize; i++) {
        buffSize = u_unescape(tests[i].original, buffer, 512);
        skSize = coll->getSortKey(buffer, buffSize, tests[i].key, 512);
    }

    qsort(tests, arraySize, sizeof(struct teststruct), compare_teststruct);

    for(i = 0; i < arraySize-1; i++) {
        for(j = i+1; j < arraySize; j++) {
            lowerSize = coll->getBound(tests[i].key, -1, UCOL_BOUND_LOWER, 1, lower, 512, status);
            upperSize = coll->getBound(tests[j].key, -1, UCOL_BOUND_UPPER, 1, upper, 512, status);
            for(k = i; k <= j; k++) {
                if(strcmp((const char *)lower, (const char *)tests[k].key) > 0) {
                    errln("Problem with lower! j = %i (%s vs %s)", k, tests[k].original, tests[i].original);
                }
                if(strcmp((const char *)upper, (const char *)tests[k].key) <= 0) {
                    errln("Problem with upper! j = %i (%s vs %s)", k, tests[k].original, tests[j].original);
                }
            }
        }
    }


    for(i = 0; i<UPRV_LENGTHOF(test); i++) {
        buffSize = u_unescape(test[i], buffer, 512);
        skSize = coll->getSortKey(buffer, buffSize, sortkey, 512);
        lowerSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_LOWER, 1, lower, 512, &status);
        upperSize = ucol_getBound(sortkey, skSize, UCOL_BOUND_UPPER_LONG, 1, upper, 512, &status);
        for(j = i+1; j<UPRV_LENGTHOF(test); j++) {
            buffSize = u_unescape(test[j], buffer, 512);
            skSize = coll->getSortKey(buffer, buffSize, sortkey, 512);
            if(strcmp((const char *)lower, (const char *)sortkey) > 0) {
                errln("Problem with lower! i = %i, j = %i (%s vs %s)", i, j, test[i], test[j]);
            }
            if(strcmp((const char *)upper, (const char *)sortkey) <= 0) {
                errln("Problem with upper! i = %i, j = %i (%s vs %s)", i, j, test[i], test[j]);
            }
        }
    }
    delete coll;
}


void CollationAPITest::TestGetTailoredSet()
{
  struct {
    const char *rules;
    const char *tests[20];
    int32_t testsize;
  } setTest[] = {
    { "&a < \\u212b", { "\\u212b", "A\\u030a", "\\u00c5" }, 3},
    { "& S < \\u0161 <<< \\u0160", { "\\u0161", "s\\u030C", "\\u0160", "S\\u030C" }, 4}
  };

  int32_t i = 0, j = 0;
  UErrorCode status = U_ZERO_ERROR;

  UnicodeString buff;
  UnicodeSet *set = NULL;

  for(i = 0; i < UPRV_LENGTHOF(setTest); i++) {
    buff = UnicodeString(setTest[i].rules, -1, US_INV).unescape();
    RuleBasedCollator coll(buff, status);
    if(U_SUCCESS(status)) {
      set = coll.getTailoredSet(status);
      if(set->size() < setTest[i].testsize) {
        errln("Tailored set size smaller (%d) than expected (%d)", set->size(), setTest[i].testsize);
      }
      for(j = 0; j < setTest[i].testsize; j++) {
        buff = UnicodeString(setTest[i].tests[j], -1, US_INV).unescape();
        if(!set->contains(buff)) {
          errln("Tailored set doesn't contain %s... It should", setTest[i].tests[j]);
        }
      }
      delete set;
    } else {
      errcheckln(status, "Couldn't open collator with rules %s - %s", setTest[i].rules, u_errorName(status));
    }
  }
}

void CollationAPITest::TestUClassID()
{
    char id = *((char *)RuleBasedCollator::getStaticClassID());
    if (id != 0) {
        errln("Static class id for RuleBasedCollator should be 0");
    }
    UErrorCode status = U_ZERO_ERROR;
    RuleBasedCollator *coll
        = (RuleBasedCollator *)Collator::createInstance(status);
    if(U_FAILURE(status)) {
      delete coll;
      errcheckln(status, "Collator creation failed with %s", u_errorName(status));
      return;
    }
    id = *((char *)coll->getDynamicClassID());
    if (id != 0) {
        errln("Dynamic class id for RuleBasedCollator should be 0");
    }
    id = *((char *)CollationKey::getStaticClassID());
    if (id != 0) {
        errln("Static class id for CollationKey should be 0");
    }
    CollationKey *key = new CollationKey();
    id = *((char *)key->getDynamicClassID());
    if (id != 0) {
        errln("Dynamic class id for CollationKey should be 0");
    }
    id = *((char *)CollationElementIterator::getStaticClassID());
    if (id != 0) {
        errln("Static class id for CollationElementIterator should be 0");
    }
    UnicodeString str("testing");
    CollationElementIterator *iter = coll->createCollationElementIterator(str);
    id = *((char *)iter->getDynamicClassID());
    if (id != 0) {
        errln("Dynamic class id for CollationElementIterator should be 0");
    }
    delete key;
    delete iter;
    delete coll;
}

class TestCollator  : public Collator
{
public:
    virtual Collator* clone(void) const;

    using Collator::compare;

    virtual UCollationResult compare(const UnicodeString& source,
                                      const UnicodeString& target,
                                      UErrorCode& status) const;
    virtual UCollationResult compare(const UnicodeString& source,
                                      const UnicodeString& target,
                                      int32_t length,
                                      UErrorCode& status) const;
    virtual UCollationResult compare(const UChar* source,
                                      int32_t sourceLength,
                                      const UChar* target,
                                      int32_t targetLength,
                                      UErrorCode& status) const;
    virtual CollationKey& getCollationKey(const UnicodeString&  source,
                                          CollationKey& key,
                                          UErrorCode& status) const;
    virtual CollationKey& getCollationKey(const UChar*source,
                                          int32_t sourceLength,
                                          CollationKey& key,
                                          UErrorCode& status) const;
    virtual int32_t hashCode(void) const;
    virtual Locale getLocale(ULocDataLocaleType type, UErrorCode& status) const;
    virtual ECollationStrength getStrength(void) const;
    virtual void setStrength(ECollationStrength newStrength);
    virtual UClassID getDynamicClassID(void) const;
    virtual void getVersion(UVersionInfo info) const;
    virtual void setAttribute(UColAttribute attr, UColAttributeValue value,
                              UErrorCode &status);
    virtual UColAttributeValue getAttribute(UColAttribute attr,
                                            UErrorCode &status) const;
    virtual uint32_t setVariableTop(const UChar *varTop, int32_t len,
                                    UErrorCode &status);
    virtual uint32_t setVariableTop(const UnicodeString &varTop,
                                    UErrorCode &status);
    virtual void setVariableTop(uint32_t varTop, UErrorCode &status);
    virtual uint32_t getVariableTop(UErrorCode &status) const;
    virtual int32_t getSortKey(const UnicodeString& source,
                            uint8_t* result,
                            int32_t resultLength) const;
    virtual int32_t getSortKey(const UChar*source, int32_t sourceLength,
                             uint8_t*result, int32_t resultLength) const;
    virtual UnicodeSet *getTailoredSet(UErrorCode &status) const;
    virtual UBool operator==(const Collator& other) const;
    // Collator::operator!= calls !Collator::operator== which works for all subclasses.
    virtual void setLocales(const Locale& requestedLocale, const Locale& validLocale, const Locale& actualLocale);
    TestCollator() : Collator() {};
    TestCollator(UCollationStrength collationStrength,
           UNormalizationMode decompositionMode) : Collator(collationStrength, decompositionMode) {};
};

inline UBool TestCollator::operator==(const Collator& other) const {
    // TestCollator has no fields, so we test for identity.
    return this == &other;

    // Normally, subclasses should do something like the following:
    //    if (this == &other) { return TRUE; }
    //    if (!Collator::operator==(other)) { return FALSE; }  // not the same class
    //
    //    const TestCollator &o = (const TestCollator&)other;
    //    (compare this vs. o's subclass fields)
}

Collator* TestCollator::clone() const
{
    return new TestCollator();
}

UCollationResult TestCollator::compare(const UnicodeString& source,
                                        const UnicodeString& target,
                                        UErrorCode& status) const
{
  if(U_SUCCESS(status)) {
    return UCollationResult(source.compare(target));
  } else {
    return UCOL_EQUAL;
  }
}

UCollationResult TestCollator::compare(const UnicodeString& source,
                                        const UnicodeString& target,
                                        int32_t length,
                                        UErrorCode& status) const
{
  if(U_SUCCESS(status)) {
    return UCollationResult(source.compare(0, length, target));
  } else {
    return UCOL_EQUAL;
  }
}

UCollationResult TestCollator::compare(const UChar* source,
                                        int32_t sourceLength,
                                        const UChar* target,
                                        int32_t targetLength,
                                        UErrorCode& status) const
{
    UnicodeString s(source, sourceLength);
    UnicodeString t(target, targetLength);
    return compare(s, t, status);
}

CollationKey& TestCollator::getCollationKey(const UnicodeString& source,
                                            CollationKey& key,
                                            UErrorCode& status) const
{
    char temp[100];
    int length = 100;
    length = source.extract(temp, length, NULL, status);
    temp[length] = 0;
    CollationKey tempkey((uint8_t*)temp, length);
    key = tempkey;
    return key;
}

CollationKey& TestCollator::getCollationKey(const UChar*source,
                                          int32_t sourceLength,
                                          CollationKey& key,
                                          UErrorCode& status) const
{
    //s tack allocation used since collationkey does not keep the unicodestring
    UnicodeString str(source, sourceLength);
    return getCollationKey(str, key, status);
}

int32_t TestCollator::getSortKey(const UnicodeString& source, uint8_t* result,
                                 int32_t resultLength) const
{
    UErrorCode status = U_ZERO_ERROR;
    int32_t length = source.extract((char *)result, resultLength, NULL,
                                    status);
    result[length] = 0;
    return length;
}

int32_t TestCollator::getSortKey(const UChar*source, int32_t sourceLength,
                                 uint8_t*result, int32_t resultLength) const
{
    UnicodeString str(source, sourceLength);
    return getSortKey(str, result, resultLength);
}

int32_t TestCollator::hashCode() const
{
    return 0;
}

Locale TestCollator::getLocale(ULocDataLocaleType type, UErrorCode& status) const
{
    // api not used, this is to make the compiler happy
    if (U_FAILURE(status)) {
        (void)type;
    }
    return NULL;
}

Collator::ECollationStrength TestCollator::getStrength() const
{
    return TERTIARY;
}

void TestCollator::setStrength(Collator::ECollationStrength newStrength)
{
    // api not used, this is to make the compiler happy
    (void)newStrength;
}

UClassID TestCollator::getDynamicClassID(void) const
{
    return 0;
}

void TestCollator::getVersion(UVersionInfo info) const
{
    // api not used, this is to make the compiler happy
    memset(info, 0, U_MAX_VERSION_LENGTH);
}

void TestCollator::setAttribute(UColAttribute /*attr*/, UColAttributeValue /*value*/,
                                UErrorCode & /*status*/)
{
}

UColAttributeValue TestCollator::getAttribute(UColAttribute attr,
                                              UErrorCode &status) const
{
    // api not used, this is to make the compiler happy
    if (U_FAILURE(status) || attr == UCOL_ATTRIBUTE_COUNT) {
        return UCOL_OFF;
    }
    return UCOL_DEFAULT;
}

uint32_t TestCollator::setVariableTop(const UChar *varTop, int32_t len,
                                  UErrorCode &status)
{
    // api not used, this is to make the compiler happy
    if (U_SUCCESS(status) && (varTop == 0 || len < -1)) {
        status = U_ILLEGAL_ARGUMENT_ERROR;
    }
    return 0;
}

uint32_t TestCollator::setVariableTop(const UnicodeString &varTop,
                                  UErrorCode &status)
{
    // api not used, this is to make the compiler happy
    if (U_SUCCESS(status) && varTop.length() == 0) {
        status = U_ILLEGAL_ARGUMENT_ERROR;
    }
    return 0;
}

void TestCollator::setVariableTop(uint32_t varTop, UErrorCode &status)
{
    // api not used, this is to make the compiler happy
    if (U_SUCCESS(status) && varTop == 0) {
        status = U_ILLEGAL_ARGUMENT_ERROR;
    }
}

uint32_t TestCollator::getVariableTop(UErrorCode &status) const
{

    // api not used, this is to make the compiler happy
    if (U_SUCCESS(status)) {
        return 0;
    }
    return (uint32_t)(0xFFFFFFFFu);
}

UnicodeSet * TestCollator::getTailoredSet(UErrorCode &status) const
{
    return Collator::getTailoredSet(status);
}

void TestCollator::setLocales(const Locale& requestedLocale, const Locale& validLocale, const Locale& actualLocale)
{
    Collator::setLocales(requestedLocale, validLocale, actualLocale);
}


void CollationAPITest::TestSubclass()
{
    TestCollator col1;
    TestCollator col2;
    doAssert(col1 != col2, "2 instances of TestCollator should be different");
    if (col1.hashCode() != col2.hashCode()) {
        errln("Every TestCollator has the same hashcode");
    }
    UnicodeString abc("abc", 3);
    UnicodeString bcd("bcd", 3);
    if (col1.compare(abc, bcd) != abc.compare(bcd)) {
        errln("TestCollator compare should be the same as the default "
              "string comparison");
    }
    CollationKey key;
    UErrorCode status = U_ZERO_ERROR;
    col1.getCollationKey(abc, key, status);
    int32_t length = 0;
    const char* bytes = (const char *)key.getByteArray(length);
    UnicodeString keyarray(bytes, length, NULL, status);
    if (abc != keyarray) {
        errln("TestCollator collationkey API is returning wrong values");
    }

    UnicodeSet expectedset(0, 0x10FFFF);
    UnicodeSet *defaultset = col1.getTailoredSet(status);
    if (!defaultset->containsAll(expectedset)
        || !expectedset.containsAll(*defaultset)) {
        errln("Error: expected default tailoring to be 0 to 0x10ffff");
    }
    delete defaultset;

    // use base class implementation
    Locale loc1 = Locale::getGermany();
    Locale loc2 = Locale::getFrance();
    col1.setLocales(loc1, loc2, loc2); // default implementation has no effect

    UnicodeString displayName;
    col1.getDisplayName(loc1, loc2, displayName); // de_DE collator in fr_FR locale

    TestCollator col3(UCOL_TERTIARY, UNORM_NONE);
    UnicodeString a("a");
    UnicodeString b("b");
    Collator::EComparisonResult result = Collator::EComparisonResult(a.compare(b));
    if(col1.compare(a, b) != result) {
      errln("Collator doesn't give default result");
    }
    if(col1.compare(a, b, 1) != result) {
      errln("Collator doesn't give default result");
    }
    if(col1.compare(a.getBuffer(), a.length(), b.getBuffer(), b.length()) != result) {
      errln("Collator doesn't give default result");
    }
}

void CollationAPITest::TestNULLCharTailoring()
{
    UErrorCode status = U_ZERO_ERROR;
    UChar buf[256] = {0};
    int32_t len = u_unescape("&a < '\\u0000'", buf, 256);
    UnicodeString first((UChar)0x0061);
    UnicodeString second((UChar)0);
    RuleBasedCollator *coll = new RuleBasedCollator(UnicodeString(buf, len), status);
    if(U_FAILURE(status)) {
        delete coll;
        errcheckln(status, "Failed to open collator - %s", u_errorName(status));
        return;
    }
    UCollationResult res = coll->compare(first, second, status);
    if(res != UCOL_LESS) {
        errln("a should be less then NULL after tailoring");
    }
    delete coll;
}

void CollationAPITest::TestClone() {
    logln("\ninit c0");
    UErrorCode status = U_ZERO_ERROR;
    RuleBasedCollator* c0 = (RuleBasedCollator*)Collator::createInstance(status);

    if (U_FAILURE(status)) {
        errcheckln(status, "Collator::CreateInstance(status) failed with %s", u_errorName(status));
        return;
    }

    c0->setStrength(Collator::TERTIARY);
    dump("c0", c0, status);

    logln("\ninit c1");
    RuleBasedCollator* c1 = (RuleBasedCollator*)Collator::createInstance(status);
    c1->setStrength(Collator::TERTIARY);
    UColAttributeValue val = c1->getAttribute(UCOL_CASE_FIRST, status);
    if(val == UCOL_LOWER_FIRST){
        c1->setAttribute(UCOL_CASE_FIRST, UCOL_UPPER_FIRST, status);
    }else{
        c1->setAttribute(UCOL_CASE_FIRST, UCOL_LOWER_FIRST, status);
    }
    dump("c0", c0, status);
    dump("c1", c1, status);

    logln("\ninit c2");
    RuleBasedCollator* c2 = (RuleBasedCollator*)c1->clone();
    val = c2->getAttribute(UCOL_CASE_FIRST, status);
    if(val == UCOL_LOWER_FIRST){
        c2->setAttribute(UCOL_CASE_FIRST, UCOL_UPPER_FIRST, status);
    }else{
        c2->setAttribute(UCOL_CASE_FIRST, UCOL_LOWER_FIRST, status);
    }
    if(U_FAILURE(status)){
        errln("set and get attributes of collator failed. %s\n", u_errorName(status));
        return;
    }
    dump("c0", c0, status);
    dump("c1", c1, status);
    dump("c2", c2, status);
    if(*c1 == *c2){
        errln("The cloned objects refer to same data");
    }
    delete c0;
    delete c1;
    delete c2;
}

void CollationAPITest::TestCloneBinary() {
    IcuTestErrorCode errorCode(*this, "TestCloneBinary");
    LocalPointer<Collator> root(Collator::createInstance(Locale::getRoot(), errorCode));
    LocalPointer<Collator> coll(Collator::createInstance("de@collation=phonebook", errorCode));
    if(errorCode.errDataIfFailureAndReset("Collator::createInstance(de@collation=phonebook)")) {
        return;
    }
    RuleBasedCollator *rbRoot = dynamic_cast<RuleBasedCollator *>(root.getAlias());
    RuleBasedCollator *rbc = dynamic_cast<RuleBasedCollator *>(coll.getAlias());
    if(rbRoot == NULL || rbc == NULL) {
        infoln("root or de@collation=phonebook is not a RuleBasedCollator");
        return;
    }
    rbc->setAttribute(UCOL_STRENGTH, UCOL_PRIMARY, errorCode);
    UnicodeString uUmlaut((UChar)0xfc);
    UnicodeString ue = UNICODE_STRING_SIMPLE("ue");
    assertEquals("rbc/primary: u-umlaut==ue", (int32_t)UCOL_EQUAL, rbc->compare(uUmlaut, ue, errorCode));
    uint8_t bin[25000];
    int32_t binLength = rbc->cloneBinary(bin, UPRV_LENGTHOF(bin), errorCode);
    if(errorCode.errDataIfFailureAndReset("rbc->cloneBinary()")) {
        return;
    }
    logln("rbc->cloneBinary() -> %d bytes", (int)binLength);

    RuleBasedCollator rbc2(bin, binLength, rbRoot, errorCode);
    if(errorCode.errDataIfFailureAndReset("RuleBasedCollator(rbc binary)")) {
        return;
    }
    assertEquals("rbc2.strength==primary", (int32_t)UCOL_PRIMARY, rbc2.getAttribute(UCOL_STRENGTH, errorCode));
    assertEquals("rbc2: u-umlaut==ue", (int32_t)UCOL_EQUAL, rbc2.compare(uUmlaut, ue, errorCode));
    assertTrue("rbc==rbc2", *rbc == rbc2);
    uint8_t bin2[25000];
    int32_t bin2Length = rbc2.cloneBinary(bin2, UPRV_LENGTHOF(bin2), errorCode);
    assertEquals("len(rbc binary)==len(rbc2 binary)", binLength, bin2Length);
    assertTrue("rbc binary==rbc2 binary", binLength == bin2Length && memcmp(bin, bin2, binLength) == 0);

    RuleBasedCollator rbc3(bin, -1, rbRoot, errorCode);
    if(errorCode.errDataIfFailureAndReset("RuleBasedCollator(rbc binary, length<0)")) {
        return;
    }
    assertEquals("rbc3.strength==primary", (int32_t)UCOL_PRIMARY, rbc3.getAttribute(UCOL_STRENGTH, errorCode));
    assertEquals("rbc3: u-umlaut==ue", (int32_t)UCOL_EQUAL, rbc3.compare(uUmlaut, ue, errorCode));
    assertTrue("rbc==rbc3", *rbc == rbc3);
}

void CollationAPITest::TestIterNumeric() {
    // Regression test for ticket #9915.
    // The collation code sometimes masked the continuation marker away
    // but later tested the result for isContinuation().
    // This test case failed because the third bytes of the computed numeric-collation primaries
    // were permutated with the script reordering table.
    // It should have been possible to reproduce this with the root collator
    // and characters with appropriate 3-byte primary weights.
    // The effectiveness of this test depends completely on the collation elements
    // and on the implementation code.
    IcuTestErrorCode errorCode(*this, "TestIterNumeric");
    RuleBasedCollator coll(UnicodeString("[reorder Hang Hani]"), errorCode);
    if(errorCode.errDataIfFailureAndReset("RuleBasedCollator constructor")) {
        return;
    }
    coll.setAttribute(UCOL_NUMERIC_COLLATION, UCOL_ON, errorCode);
    UCharIterator iter40, iter72;
    uiter_setUTF8(&iter40, "\x34\x30", 2);
    uiter_setUTF8(&iter72, "\x37\x32", 2);
    UCollationResult result = coll.compare(iter40, iter72, errorCode);
    assertEquals("40<72", (int32_t)UCOL_LESS, (int32_t)result);
}

void CollationAPITest::TestBadKeywords() {
    // Test locale IDs with errors.
    // Valid locale IDs are tested via data-driven tests.
    UErrorCode errorCode = U_ZERO_ERROR;
    Locale bogusLocale(Locale::getRoot());
    bogusLocale.setToBogus();
    LocalPointer<Collator> coll(Collator::createInstance(bogusLocale, errorCode));
    if(errorCode != U_ILLEGAL_ARGUMENT_ERROR) {
        errln("Collator::createInstance(bogus locale) did not fail as expected - %s",
              u_errorName(errorCode));
    }

    // Unknown value.
    const char *localeID = "it-u-ks-xyz";
    errorCode = U_ZERO_ERROR;
    coll.adoptInstead(Collator::createInstance(localeID, errorCode));
    if(errorCode != U_ILLEGAL_ARGUMENT_ERROR) {
        dataerrln("Collator::createInstance(%s) did not fail as expected - %s",
              localeID, u_errorName(errorCode));
    }

    // Unsupported attributes.
    localeID = "it@colHiraganaQuaternary=true";
    errorCode = U_ZERO_ERROR;
    coll.adoptInstead(Collator::createInstance(localeID, errorCode));
    if(errorCode != U_UNSUPPORTED_ERROR) {
        if (errorCode == U_FILE_ACCESS_ERROR) {
            dataerrln("Collator::createInstance(it@colHiraganaQuaternary=true) : %s", u_errorName(errorCode));
        } else {
            errln("Collator::createInstance(%s) did not fail as expected - %s",
                  localeID, u_errorName(errorCode));
        }
    }

    localeID = "it-u-vt-u24";
    errorCode = U_ZERO_ERROR;
    coll.adoptInstead(Collator::createInstance(localeID, errorCode));
    if(errorCode != U_UNSUPPORTED_ERROR) {
        if (errorCode == U_ILLEGAL_ARGUMENT_ERROR || errorCode == U_FILE_ACCESS_ERROR) {
            dataerrln("Collator::createInstance(it-u-vt-u24) : %s", u_errorName(errorCode));
        } else {
           errln("Collator::createInstance(%s) did not fail as expected - %s",
                  localeID, u_errorName(errorCode));
        }
    }
}

void CollationAPITest::TestGapTooSmall() {
    IcuTestErrorCode errorCode(*this, "TestGapTooSmall");
    // Try to tailor >20k characters into a too-small primary gap between symbols
    // that have 3-byte primary weights.
    // In FractionalUCA.txt:
    // 263A; [0C BA D0, 05, 05]  # Zyyy So  [084A.0020.0002]  * WHITE SMILING FACE
    // 263B; [0C BA D7, 05, 05]  # Zyyy So  [084B.0020.0002]  * BLACK SMILING FACE
    {
        RuleBasedCollator(u"&<*\u4E00-\u9FFF", errorCode);
        if(errorCode.isSuccess()) {
            errln("no exception for primary-gap overflow");
        } else if(errorCode.get() == U_BUFFER_OVERFLOW_ERROR) {
            // This is the expected error.
            // assertTrue("exception message mentions 'gap'", e.getMessage().contains("gap"));
        } else {
            errln("unexpected error for primary-gap overflow: %s", errorCode.errorName());
        }
        errorCode.reset();
    }

    // CLDR 32/ICU 60 FractionalUCA.txt makes room at the end of the symbols range
    // for several 2-byte primaries, or a large number of 3-byters.
    // The reset point is primary-before what should be
    // the special currency-first-primary contraction,
    // which is hopefully fairly stable, but not guaranteed stable.
    // In FractionalUCA.txt:
    // FDD1 20AC; [0D 70 02, 05, 05]  # CURRENCY first primary
    {
        RuleBasedCollator coll(u"&[before 1]\uFDD1<*\u4E00-\u9FFF", errorCode);
        assertTrue("tailored Han before currency", coll.compare(u"\u4E00", u"$", errorCode) < 0);
        errorCode.errIfFailureAndReset(
            "unexpected exception for tailoring many characters at the end of symbols");
    }
}

 void CollationAPITest::dump(UnicodeString msg, RuleBasedCollator* c, UErrorCode& status) {
    const char* bigone = "One";
    const char* littleone = "one";

    logln(msg + " " + c->compare(bigone, littleone) +
                        " s: " + c->getStrength() +
                        " u: " + c->getAttribute(UCOL_CASE_FIRST, status));
}
void CollationAPITest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par */)
{
    if (exec) logln("TestSuite CollationAPITest: ");
    TESTCASE_AUTO_BEGIN;
    TESTCASE_AUTO(TestProperty);
    TESTCASE_AUTO(TestKeywordValues);
    TESTCASE_AUTO(TestOperators);
    TESTCASE_AUTO(TestDuplicate);
    TESTCASE_AUTO(TestCompare);
    TESTCASE_AUTO(TestHashCode);
    TESTCASE_AUTO(TestCollationKey);
    TESTCASE_AUTO(TestElemIter);
    TESTCASE_AUTO(TestGetAll);
    TESTCASE_AUTO(TestRuleBasedColl);
    TESTCASE_AUTO(TestDecomposition);
    TESTCASE_AUTO(TestSafeClone);
    TESTCASE_AUTO(TestSortKey);
    TESTCASE_AUTO(TestSortKeyOverflow);
    TESTCASE_AUTO(TestMaxExpansion);
    TESTCASE_AUTO(TestDisplayName);
    TESTCASE_AUTO(TestAttribute);
    TESTCASE_AUTO(TestVariableTopSetting);
    TESTCASE_AUTO(TestMaxVariable);
    TESTCASE_AUTO(TestRules);
    TESTCASE_AUTO(TestGetLocale);
    TESTCASE_AUTO(TestBounds);
    TESTCASE_AUTO(TestGetTailoredSet);
    TESTCASE_AUTO(TestUClassID);
    TESTCASE_AUTO(TestSubclass);
    TESTCASE_AUTO(TestNULLCharTailoring);
    TESTCASE_AUTO(TestClone);
    TESTCASE_AUTO(TestCloneBinary);
    TESTCASE_AUTO(TestIterNumeric);
    TESTCASE_AUTO(TestBadKeywords);
    TESTCASE_AUTO(TestGapTooSmall);
    TESTCASE_AUTO_END;
}

#endif /* #if !UCONFIG_NO_COLLATION */