/*
******************************************************************************
*
-* Copyright (C) 2002-2004, International Business Machines
+* Copyright (C) 2002-2014, International Business Machines
* Corporation and others. All Rights Reserved.
*
******************************************************************************
* Tests of ustring.h Unicode string API functions.
*/
-#include "unicode/utypes.h"
#include "unicode/ustring.h"
-#include "unicode/uloc.h"
#include "unicode/ucnv.h"
#include "unicode/uiter.h"
#include "cintltst.h"
-#include "cucdtst.h"
#include <string.h>
-#include <stdlib.h>
#define LENGTHOF(array) (sizeof(array)/sizeof((array)[0]))
static void TestUnescape(void);
static void TestCountChar32(void);
static void TestUCharIterator(void);
-static void TestUNormIterator(void);
void addUStringTest(TestNode** root);
addTest(root, &TestUnescape, "tsutil/custrtst/TestUnescape");
addTest(root, &TestCountChar32, "tsutil/custrtst/TestCountChar32");
addTest(root, &TestUCharIterator, "tsutil/custrtst/TestUCharIterator");
- addTest(root, &TestUNormIterator, "tsutil/custrtst/TestUNormIterator");
-
- /* cstrcase.c functions, declared in cucdtst.h */
- addTest(root, &TestCaseLower, "tsutil/custrtst/TestCaseLower");
- addTest(root, &TestCaseUpper, "tsutil/custrtst/TestCaseUpper");
-#if !UCONFIG_NO_BREAK_ITERATION
- addTest(root, &TestCaseTitle, "tsutil/custrtst/TestCaseTitle");
-#endif
- addTest(root, &TestCaseFolding, "tsutil/custrtst/TestCaseFolding");
- addTest(root, &TestCaseCompare, "tsutil/custrtst/TestCaseCompare");
}
/* test data for TestStringFunctions ---------------------------------------- */
/* ### TODO test other iterators: CharacterIterator, Replaceable */
}
-
-#if UCONFIG_NO_COLLATION
-
-static void
-TestUNormIterator() {
- /* test nothing */
-}
-
-#else
-
-#include "unicode/unorm.h"
-#include "unorm_it.h"
-
-/*
- * Compare results from two iterators, should be same.
- * Assume that the text is not empty and that
- * iteration start==0 and iteration limit==length.
- *
- * Modified version of compareIterators() but does not assume that indexes
- * are available.
- */
-static void
-compareIterNoIndexes(UCharIterator *iter1, const char *n1,
- UCharIterator *iter2, const char *n2,
- int32_t middle) {
- uint32_t state;
- int32_t i;
- UChar32 c1, c2;
- UErrorCode errorCode;
-
- /* set into the middle */
- iter1->move(iter1, middle, UITER_ZERO);
- iter2->move(iter2, middle, UITER_ZERO);
-
- /* test current() */
- c1=iter1->current(iter1);
- c2=iter2->current(iter2);
- if(c1!=c2) {
- log_err("%s->current()=U+%04x != U+%04x=%s->current() at middle=%d\n", n1, c1, c2, n2, middle);
- return;
- }
-
- /* move forward 3 UChars */
- for(i=0; i<3; ++i) {
- c1=iter1->next(iter1);
- c2=iter2->next(iter2);
- if(c1!=c2) {
- log_err("%s->next()=U+%04x != U+%04x=%s->next() at %d (started in middle)\n", n1, c1, c2, n2, iter1->getIndex(iter1, UITER_CURRENT));
- return;
- }
- }
-
- /* move backward 5 UChars */
- for(i=0; i<5; ++i) {
- c1=iter1->previous(iter1);
- c2=iter2->previous(iter2);
- if(c1!=c2) {
- log_err("%s->previous()=U+%04x != U+%04x=%s->previous() at %d (started in middle)\n", n1, c1, c2, n2, iter1->getIndex(iter1, UITER_CURRENT));
- return;
- }
- }
-
- /* iterate forward from the beginning */
- iter1->move(iter1, 0, UITER_START);
- if(!iter1->hasNext(iter1)) {
- log_err("%s->hasNext() at the start returns FALSE\n", n1);
- return;
- }
-
- iter2->move(iter2, 0, UITER_START);
- if(!iter2->hasNext(iter2)) {
- log_err("%s->hasNext() at the start returns FALSE\n", n2);
- return;
- }
-
- do {
- c1=iter1->next(iter1);
- c2=iter2->next(iter2);
- if(c1!=c2) {
- log_err("%s->next()=U+%04x != U+%04x=%s->next() at %d\n", n1, c1, c2, n2, iter1->getIndex(iter1, UITER_CURRENT));
- return;
- }
- } while(c1>=0);
-
- if(iter1->hasNext(iter1)) {
- log_err("%s->hasNext() at the end returns TRUE\n", n1);
- return;
- }
- if(iter2->hasNext(iter2)) {
- log_err("%s->hasNext() at the end returns TRUE\n", n2);
- return;
- }
-
- /* iterate backward */
- do {
- c1=iter1->previous(iter1);
- c2=iter2->previous(iter2);
- if(c1!=c2) {
- log_err("%s->previous()=U+%04x != U+%04x=%s->previous() at %d\n", n1, c1, c2, n2, iter1->getIndex(iter1, UITER_CURRENT));
- return;
- }
- } while(c1>=0);
-
- /* back to the middle */
- iter1->move(iter1, middle, UITER_ZERO);
- iter2->move(iter2, middle, UITER_ZERO);
-
- /* try get/set state */
- while((state=uiter_getState(iter2))==UITER_NO_STATE) {
- if(!iter2->hasNext(iter2)) {
- log_err("%s has no known state from middle=%d to the end\n", n2, middle);
- return;
- }
- iter2->next(iter2);
- }
-
- errorCode=U_ZERO_ERROR;
-
- c2=iter2->current(iter2);
- iter2->move(iter2, 0, UITER_ZERO);
- uiter_setState(iter2, state, &errorCode);
- c1=iter2->current(iter2);
- if(U_FAILURE(errorCode) || c1!=c2) {
- log_err("%s->current() differs across get/set state, U+%04x vs. U+%04x\n", n2, c2, c1);
- return;
- }
-
- c2=iter2->previous(iter2);
- iter2->move(iter2, 0, UITER_ZERO);
- uiter_setState(iter2, state, &errorCode);
- c1=iter2->previous(iter2);
- if(U_FAILURE(errorCode) || c1!=c2) {
- log_err("%s->previous() differs across get/set state, U+%04x vs. U+%04x\n", n2, c2, c1);
- return;
- }
-
- /* iterate backward from the end */
- iter1->move(iter1, 0, UITER_LIMIT);
- if(!iter1->hasPrevious(iter1)) {
- log_err("%s->hasPrevious() at the end returns FALSE\n", n1);
- return;
- }
-
- iter2->move(iter2, 0, UITER_LIMIT);
- if(!iter2->hasPrevious(iter2)) {
- log_err("%s->hasPrevious() at the end returns FALSE\n", n2);
- return;
- }
-
- do {
- c1=iter1->previous(iter1);
- c2=iter2->previous(iter2);
- if(c1!=c2) {
- log_err("%s->previous()=U+%04x != U+%04x=%s->previous() at %d\n", n1, c1, c2, n2, iter1->getIndex(iter1, UITER_CURRENT));
- return;
- }
- } while(c1>=0);
-
- if(iter1->hasPrevious(iter1)) {
- log_err("%s->hasPrevious() at the start returns TRUE\n", n1);
- return;
- }
- if(iter2->hasPrevious(iter2)) {
- log_err("%s->hasPrevious() at the start returns TRUE\n", n2);
- return;
- }
-}
-
-/* n2 must have a digit 1 at the end, will be incremented with the normalization mode */
-static void
-testUNormIteratorWithText(const UChar *text, int32_t textLength, int32_t middle,
- const char *name1, const char *n2) {
- UChar buffer[600];
- char name2[40];
-
- UCharIterator iter1, iter2, *iter;
- UNormIterator *uni;
-
- UNormalizationMode mode;
- UErrorCode errorCode;
- int32_t length;
-
- /* open a normalizing iterator */
- errorCode=U_ZERO_ERROR;
- uni=unorm_openIter(NULL, 0, &errorCode);
- if(U_FAILURE(errorCode)) {
- log_err("unorm_openIter() fails: %s\n", u_errorName(errorCode));
- return;
- }
-
- /* set iterator 2 to the original text */
- uiter_setString(&iter2, text, textLength);
-
- strcpy(name2, n2);
-
- /* test the normalizing iterator for each mode */
- for(mode=UNORM_NONE; mode<UNORM_MODE_COUNT; ++mode) {
- length=unorm_normalize(text, textLength, mode, 0, buffer, LENGTHOF(buffer), &errorCode);
- if(U_FAILURE(errorCode)) {
- log_err("unorm_normalize(mode %d) failed: %s\n", mode, u_errorName(errorCode));
- break;
- }
-
- /* set iterator 1 to the normalized text */
- uiter_setString(&iter1, buffer, length);
-
- /* set the normalizing iterator to use iter2 */
- iter=unorm_setIter(uni, &iter2, mode, &errorCode);
- if(U_FAILURE(errorCode)) {
- log_err("unorm_setIter(mode %d) failed: %s\n", mode, u_errorName(errorCode));
- break;
- }
-
- compareIterNoIndexes(&iter1, name1, iter, name2, middle);
- ++name2[strlen(name2)-1];
- }
-
- unorm_closeIter(uni);
-}
-
-static void
-TestUNormIterator() {
- static const UChar text[]={ /* must contain <00C5 0327> see u_strchr() below */
- 0x61, /* 'a' */
- 0xe4, 0x61, 0x308, /* variations of 'a'+umlaut */
- 0xc5, 0x327, 0x41, 0x30a, 0x327, 0x41, 0x327, 0x30a, /* variations of 'A'+ring+cedilla */
- 0xfb03, 0xfb00, 0x69, 0x66, 0x66, 0x69, 0x66, 0xfb01 /* variations of 'ffi' */
- };
- static const UChar surrogateText[]={
- 0x6e, 0xd900, 0x6a, 0xdc00, 0xd900, 0xdc00, 0x61
- };
-
- UChar longText[600];
- int32_t i, middle, length;
-
- length=LENGTHOF(text);
- testUNormIteratorWithText(text, length, length/2, "UCharIter", "UNormIter1");
- testUNormIteratorWithText(text, length, length, "UCharIterEnd", "UNormIterEnd1");
-
- /* test again, this time with an insane string to cause internal buffer overflows */
- middle=(int32_t)(u_strchr(text, 0x327)-text); /* see comment at text[] */
- memcpy(longText, text, middle*U_SIZEOF_UCHAR);
- for(i=0; i<150; ++i) {
- longText[middle+i]=0x30a; /* insert many rings between 'A-ring' and cedilla */
- }
- memcpy(longText+middle+i, text+middle, (LENGTHOF(text)-middle)*U_SIZEOF_UCHAR);
- length=LENGTHOF(text)+i;
-
- /* append another copy of this string for more overflows */
- memcpy(longText+length, longText, length*U_SIZEOF_UCHAR);
- length*=2;
-
- /* the first test of the following two starts at length/4, inside the sea of combining rings */
- testUNormIteratorWithText(longText, length, length/4, "UCharIterLong", "UNormIterLong1");
- testUNormIteratorWithText(longText, length, length, "UCharIterLongEnd", "UNormIterLongEnd1");
-
- length=LENGTHOF(surrogateText);
- testUNormIteratorWithText(surrogateText, length, length/4, "UCharIterSurr", "UNormIterSurr1");
- testUNormIteratorWithText(surrogateText, length, length, "UCharIterSurrEnd", "UNormIterSurrEnd1");
-}
-
-#endif