+
+static void TestBreakIteratorRefresh(void) {
+ /*
+ * RefreshInput changes out the input of a Break Iterator without
+ * changing anything else in the iterator's state. Used with Java JNI,
+ * when Java moves the underlying string storage. This test
+ * runs a ubrk_next() repeatedly, moving the text in the middle of the sequence.
+ * The right set of boundaries should still be found.
+ */
+ UChar testStr[] = {0x20, 0x41, 0x20, 0x42, 0x20, 0x43, 0x20, 0x44, 0x0}; /* = " A B C D" */
+ UChar movedStr[] = {0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0};
+ UErrorCode status = U_ZERO_ERROR;
+ UBreakIterator *bi;
+ UText ut1 = UTEXT_INITIALIZER;
+ UText ut2 = UTEXT_INITIALIZER;
+
+ bi = ubrk_open(UBRK_LINE, "en_US", NULL, 0, &status);
+ TEST_ASSERT_SUCCESS(status);
+ if (U_FAILURE(status)) {
+ return;
+ }
+
+ utext_openUChars(&ut1, testStr, -1, &status);
+ TEST_ASSERT_SUCCESS(status);
+ ubrk_setUText(bi, &ut1, &status);
+ TEST_ASSERT_SUCCESS(status);
+
+ if (U_SUCCESS(status)) {
+ /* Line boundaries will occur before each letter in the original string */
+ TEST_ASSERT(1 == ubrk_next(bi));
+ TEST_ASSERT(3 == ubrk_next(bi));
+
+ /* Move the string, kill the original string. */
+ u_strcpy(movedStr, testStr);
+ u_memset(testStr, 0x20, u_strlen(testStr));
+ utext_openUChars(&ut2, movedStr, -1, &status);
+ TEST_ASSERT_SUCCESS(status);
+ ubrk_refreshUText(bi, &ut2, &status);
+ TEST_ASSERT_SUCCESS(status);
+
+ /* Find the following matches, now working in the moved string. */
+ TEST_ASSERT(5 == ubrk_next(bi));
+ TEST_ASSERT(7 == ubrk_next(bi));
+ TEST_ASSERT(8 == ubrk_next(bi));
+ TEST_ASSERT(UBRK_DONE == ubrk_next(bi));
+ TEST_ASSERT_SUCCESS(status);
+
+ utext_close(&ut1);
+ utext_close(&ut2);
+ }
+ ubrk_close(bi);
+}
+
+
+static void TestBug11665(void) {
+ // The problem was with the incorrect breaking of Japanese text beginning
+ // with Katakana characters when no prior Japanese or Chinese text had been
+ // encountered.
+ //
+ // Tested here in cintltst, rather than in intltest, because only cintltst
+ // tests have the ability to reset ICU, which is needed to get the bug
+ // to manifest itself.
+
+ static UChar japaneseText[] = {0x30A2, 0x30EC, 0x30EB, 0x30AE, 0x30FC, 0x6027, 0x7D50, 0x819C, 0x708E};
+ int32_t boundaries[10] = {0};
+ UBreakIterator *bi = NULL;
+ int32_t brk;
+ int32_t brkIdx = 0;
+ int32_t totalBreaks = 0;
+ UErrorCode status = U_ZERO_ERROR;
+
+ ctest_resetICU();
+ bi = ubrk_open(UBRK_WORD, "en_US", japaneseText, UPRV_LENGTHOF(japaneseText), &status);
+ TEST_ASSERT_SUCCESS(status);
+ if (!bi) {
+ return;
+ }
+ for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
+ boundaries[brkIdx] = brk;
+ if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
+ break;
+ }
+ }
+ if (brkIdx <= 2 || brkIdx >= UPRV_LENGTHOF(boundaries)) {
+ log_err("%s:%d too few or many breaks found.\n", __FILE__, __LINE__);
+ } else {
+ totalBreaks = brkIdx;
+ brkIdx = 0;
+ for (brk=ubrk_first(bi); brk != UBRK_DONE; brk=ubrk_next(bi)) {
+ if (brk != boundaries[brkIdx]) {
+ log_err("%s:%d Break #%d differs between first and second iteration.\n", __FILE__, __LINE__, brkIdx);
+ break;
+ }
+ if (++brkIdx >= UPRV_LENGTHOF(boundaries) - 1) {
+ log_err("%s:%d Too many breaks.\n", __FILE__, __LINE__);
+ break;
+ }
+ }
+ if (totalBreaks != brkIdx) {
+ log_err("%s:%d Number of breaks differ between first and second iteration.\n", __FILE__, __LINE__);
+ }
+ }
+ ubrk_close(bi);
+}
+
+/*
+ * expOffset is the set of expected offsets, ending with '-1'.
+ * "Expected expOffset -1" means "expected the end of the offsets"
+ */
+
+static const char testSentenceSuppressionsEn[] = "Mr. Jones comes home. Dr. Smith Ph.D. is out. In the U.S.A. it is hot.";
+static const int32_t testSentSuppFwdOffsetsEn[] = { 22, 46, 70, -1 }; /* With suppressions */
+static const int32_t testSentFwdOffsetsEn[] = { 4, 22, 26, 46, 70, -1 }; /* Without suppressions */
+static const int32_t testSentSuppRevOffsetsEn[] = { 46, 22, 0, -1 }; /* With suppressions */
+static const int32_t testSentRevOffsetsEn[] = { 46, 26, 22, 4, 0, -1 }; /* Without suppressions */
+
+static const char testSentenceSuppressionsDe[] = "Wenn ich schon h\\u00F6re zu Guttenberg kommt evtl. zur\\u00FCck.";
+// "Wenn ich schon höre zu Guttenberg kommt evtl. zurück."
+static const int32_t testSentSuppFwdOffsetsDe[] = { 53, -1 }; /* With suppressions */
+static const int32_t testSentFwdOffsetsDe[] = { 53, -1 }; /* Without suppressions; no break in evtl. zur due to casing */
+static const int32_t testSentSuppRevOffsetsDe[] = { 0, -1 }; /* With suppressions */
+static const int32_t testSentRevOffsetsDe[] = { 0, -1 }; /* Without suppressions */
+
+static const char testSentenceSuppressionsEs[] = "Te esperamos todos los miercoles en Bravo 416, Col. El Pueblo a las 7 PM.";
+static const int32_t testSentSuppFwdOffsetsEs[] = { 73, -1 }; /* With suppressions */
+static const int32_t testSentFwdOffsetsEs[] = { 52, 73, -1 }; /* Without suppressions */
+static const int32_t testSentSuppRevOffsetsEs[] = { 0, -1 }; /* With suppressions */
+static const int32_t testSentRevOffsetsEs[] = { 52, 0, -1 }; /* Without suppressions */
+
+static const char testSentenceSuppressionsE1[] = "Add or detract. The world will little note.";
+static const char testSentenceSuppressionsE1u[] = "ADD OR DETRACT. THE WORLD WILL LITTLE NOTE.";
+static const int32_t testSentFwdOffsetsE1[] = { 16, 43, -1 }; /* Suppressions and case should make no difference */
+static const int32_t testSentRevOffsetsE1[] = { 16, 0, -1 }; /* Suppressions and case should make no difference */
+
+static const char testSentenceSuppressionsE2[] = "Coming up, the sprints at NCAA. Are you watching?";
+static const char testSentenceSuppressionsE2u[] = "COMING UP, THE SPRINTS AT NCAA. ARE YOU WATCHING?";
+static const int32_t testSentFwdOffsetsE2[] = { 32, 49, -1 }; /* Suppressions and case should make no difference */
+static const int32_t testSentRevOffsetsE2[] = { 32, 0, -1 }; /* Suppressions and case should make no difference */
+
+static const char testSentenceSuppressionsFr[] = "Tr\\u00E8s bonne prise de parole de M. Junod, municipal \\u00E0 la culture de Lausanne.";
+// "Très bonne prise de parole de M. Junod, municipal à la culture de Lausanne."
+static const int32_t testSentFwdOffsetsFr[] = { 33, 75, -1 }; /* Without suppressions */
+static const int32_t testSentSuppFwdOffsetsFr[] = { 75, -1 }; /* With suppressions */
+static const int32_t testSentRevOffsetsFr[] = { 33, 0, -1 }; /* Without suppressions */
+static const int32_t testSentSuppRevOffsetsFr[] = { 0, -1 }; /* With suppressions */
+
+static const char testSentenceSuppressionsE3[] = "G8 countries e.g. U.K., Japan. Sanctions i.e. restrictions. Test E. Xx G. Xx I. Xx.";
+static const char testSentenceSuppressionsE3u[] = "G8 COUNTRIES E.G. U.K., JAPAN. SANCTIONS I.E. RESTRICTIONS. TEST E. XX G. XX I. XX.";
+static const int32_t testSentSuppFwdOffsetsE3[] = { 31, 60, 83, -1 }; /* With suppressions */
+static const int32_t testSentSuppRevOffsetsE3[] = { 60, 31, 0, -1 }; /* With suppressions */
+static const int32_t testSentFwdOffsetsE3[] = { 18, 31, 60, 68, 74, 80, 83, -1 }; /* Without suppressions */
+static const int32_t testSentRevOffsetsE3[] = { 80, 74, 68, 60, 31, 18, 0, -1 }; /* Without suppressions */
+static const int32_t testSentFwdOffsetsE3u[] = { 18, 31, 46, 60, 68, 74, 80, 83, -1 }; /* Without suppressions */
+static const int32_t testSentRevOffsetsE3u[] = { 80, 74, 68, 60, 46, 31, 18, 0, -1 }; /* Without suppressions */
+
+enum { kTextULenMax = 128, kTextBLenMax = 192 };
+
+typedef struct {
+ const char * locale;
+ const char * text;
+ const int32_t * expFwdOffsets;
+ const int32_t * expRevOffsets;
+} TestBISuppressionsItem;
+
+static const TestBISuppressionsItem testBISuppressionsItems[] = {
+ { "en@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn },
+ { "en", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
+ { "fr@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn },
+ { "af@ss=standard", testSentenceSuppressionsEn, testSentSuppFwdOffsetsEn, testSentSuppRevOffsetsEn }, /* no brkiter data => en suppressions? */
+ { "zh@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
+ { "zh_Hant@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
+ { "fi@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
+ { "ja@ss=standard", testSentenceSuppressionsEn, testSentFwdOffsetsEn, testSentRevOffsetsEn }, /* brkiter data, no suppressions data => no suppressions */
+ { "de@ss=standard", testSentenceSuppressionsDe, testSentSuppFwdOffsetsDe, testSentSuppRevOffsetsDe },
+ { "de", testSentenceSuppressionsDe, testSentFwdOffsetsDe, testSentRevOffsetsDe },
+ { "es@ss=standard", testSentenceSuppressionsEs, testSentSuppFwdOffsetsEs, testSentSuppRevOffsetsEs },
+ { "es", testSentenceSuppressionsEs, testSentFwdOffsetsEs, testSentRevOffsetsEs },
+ { "en", testSentenceSuppressionsE1, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
+ { "en@ss=standard", testSentenceSuppressionsE1, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
+ { "en", testSentenceSuppressionsE1u, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
+ { "en@ss=standard", testSentenceSuppressionsE1u, testSentFwdOffsetsE1, testSentRevOffsetsE1 },
+ { "en", testSentenceSuppressionsE2, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
+ { "en@ss=standard", testSentenceSuppressionsE2, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
+ { "en", testSentenceSuppressionsE2u, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
+ { "en@ss=standard", testSentenceSuppressionsE2u, testSentFwdOffsetsE2, testSentRevOffsetsE2 },
+ { "fr", testSentenceSuppressionsFr, testSentFwdOffsetsFr, testSentRevOffsetsFr },
+ { "fr@ss=standard", testSentenceSuppressionsFr, testSentSuppFwdOffsetsFr, testSentSuppRevOffsetsFr },
+ { "en@ss=standard", testSentenceSuppressionsE3, testSentSuppFwdOffsetsE3, testSentSuppRevOffsetsE3 },
+ { "en", testSentenceSuppressionsE3, testSentFwdOffsetsE3, testSentRevOffsetsE3 },
+ { "en@ss=standard", testSentenceSuppressionsE3u, testSentSuppFwdOffsetsE3, testSentSuppRevOffsetsE3 },
+ { "en", testSentenceSuppressionsE3u, testSentFwdOffsetsE3u, testSentRevOffsetsE3u },
+ { NULL, NULL, NULL }
+};
+
+static void TestBreakIteratorSuppressions(void) {
+ const TestBISuppressionsItem * itemPtr;
+
+ for (itemPtr = testBISuppressionsItems; itemPtr->locale != NULL; itemPtr++) {
+ UChar textU[kTextULenMax + 1];
+ char textB[kTextBLenMax];
+ int32_t textULen = u_unescape(itemPtr->text, textU, kTextULenMax);
+ textU[kTextULenMax] = 0; // ensure zero termination
+ UErrorCode status = U_ZERO_ERROR;
+ UBreakIterator *bi = ubrk_open(UBRK_SENTENCE, itemPtr->locale, textU, textULen, &status);
+ log_verbose("#%d: %s\n", (itemPtr-testBISuppressionsItems), itemPtr->locale);
+ if (U_SUCCESS(status)) {
+ int32_t offset, start;
+ const int32_t * expOffsetPtr;
+ const int32_t * expOffsetStart;
+ u_austrcpy(textB, textU);
+
+ expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
+ ubrk_first(bi);
+ for (; (offset = ubrk_next(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
+ if (offset != *expOffsetPtr) {
+ log_err("FAIL: ubrk_next loc \"%s\", expected %d, got %d, text \"%s\"\n",
+ itemPtr->locale, *expOffsetPtr, offset, textB);
+ }
+ }
+ if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
+ log_err("FAIL: ubrk_next loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
+ itemPtr->locale, offset, *expOffsetPtr, textB);
+ }
+
+ expOffsetStart = expOffsetPtr = itemPtr->expFwdOffsets;
+ start = ubrk_first(bi) + 1;
+ for (; (offset = ubrk_following(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
+ if (offset != *expOffsetPtr) {
+ log_err("FAIL: ubrk_following(%d) loc \"%s\", expected %d, got %d, text \"%s\"\n",
+ start, itemPtr->locale, *expOffsetPtr, offset, textB);
+ }
+ start = *expOffsetPtr + 1;
+ }
+ if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
+ log_err("FAIL: ubrk_following(%d) loc \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
+ start, itemPtr->locale, offset, *expOffsetPtr, textB);
+ }
+
+ expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
+ offset = ubrk_last(bi);
+ log_verbose("___ @%d ubrk_last\n", offset);
+ if(offset == 0) {
+ log_err("FAIL: ubrk_last loc \"%s\" unexpected %d\n", itemPtr->locale, offset);
+ }
+ for (; (offset = ubrk_previous(bi)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
+ if (offset != *expOffsetPtr) {
+ log_err("FAIL: ubrk_previous loc \"%s\", expected %d, got %d, text \"%s\"\n",
+ itemPtr->locale, *expOffsetPtr, offset, textB);
+ } else {
+ log_verbose("[%d] @%d ubrk_previous()\n", (expOffsetPtr - expOffsetStart), offset);
+ }
+ }
+ if (offset != UBRK_DONE || *expOffsetPtr >= 0) {
+ log_err("FAIL: ubrk_previous loc \"%s\", expected UBRK_DONE & expOffset[%d] -1, got %d and %d, text \"%s\"\n",
+ itemPtr->locale, expOffsetPtr - expOffsetStart, offset, *expOffsetPtr, textB);
+ }
+
+ expOffsetStart = expOffsetPtr = itemPtr->expRevOffsets;
+ start = ubrk_last(bi) - 1;
+ for (; (offset = ubrk_preceding(bi, start)) != UBRK_DONE && *expOffsetPtr >= 0; expOffsetPtr++) {
+ if (offset != *expOffsetPtr) {
+ log_err("FAIL: ubrk_preceding(%d) loc \"%s\", expected %d, got %d, text \"%s\"\n",
+ start, itemPtr->locale, *expOffsetPtr, offset, textB);
+ }
+ start = *expOffsetPtr - 1;
+ }
+ if (start >=0 && (offset != UBRK_DONE || *expOffsetPtr >= 0)) {
+ log_err("FAIL: ubrk_preceding loc(%d) \"%s\", expected UBRK_DONE & expOffset -1, got %d and %d, text \"%s\"\n",
+ start, itemPtr->locale, offset, *expOffsetPtr, textB);
+ }
+
+ ubrk_close(bi);
+ } else {
+ log_data_err("FAIL: ubrk_open(UBRK_SENTENCE, \"%s\", ...) status %s (Are you missing data?)\n",
+ itemPtr->locale, u_errorName(status));
+ }
+ }
+}
+
+
projects / apple / icu.git / blobdiff