ICU-10998 collation reorder others=Zzzz, recognize only others = reset

X-SVN-Rev: 36017

icu4c/source/i18n/coll.cpp

@@ -301,6 +301,9 @@ int32_t getReorderCode(const char *s) {
             return UCOL_REORDER_CODE_FIRST + i;
         }
     }
+    // Not supporting "others" = UCOL_REORDER_CODE_OTHERS
+    // as a synonym for Zzzz = USCRIPT_UNKNOWN for now:
+    // Avoid introducing synonyms/aliases.
     return -1;
 }
 

icu4c/source/i18n/collationruleparser.cpp

@@ -710,8 +710,7 @@ CollationRuleParser::parseReordering(const UnicodeString &raw, UErrorCode &error
         i = limit;
     }
     int32_t length = reorderCodes.size();
-    if(length == 1 && reorderCodes.elementAti(0) == UCOL_REORDER_CODE_DEFAULT) {
-        // The root collator does not have a reordering, by definition.
+    if(length == 1 && reorderCodes.elementAti(0) == UCOL_REORDER_CODE_NONE) {
         settings->resetReordering();
         return;
     }
@@ -738,10 +737,10 @@ CollationRuleParser::getReorderCode(const char *word) {
     if(script >= 0) {
         return script;
     }
-    if(uprv_stricmp(word, "default") == 0) {
-        return UCOL_REORDER_CODE_DEFAULT;
+    if(uprv_stricmp(word, "others") == 0) {
+        return UCOL_REORDER_CODE_OTHERS;  // same as Zzzz = USCRIPT_UNKNOWN
     }
-    return -2;
+    return -1;
 }
 
 UColAttributeValue

icu4c/source/i18n/collationruleparser.h

@@ -132,8 +132,7 @@ public:
     /**
      * Gets a script or reorder code from its string representation.
      * @return the script/reorder code, or
-     * -1==UCOL_REORDER_CODE_DEFAULT, or
-     * -2 if not recognized
+     * -1 if not recognized
      */
     static int32_t getReorderCode(const char *word);
 

icu4c/source/i18n/rulebasedcollator.cpp

@@ -670,7 +670,7 @@ RuleBasedCollator::setReorderCodes(const int32_t *reorderCodes, int32_t length,
         errorCode = U_MEMORY_ALLOCATION_ERROR;
         return;
     }
-    if(length == 0) {
+    if(length == 0 || (length == 1 && reorderCodes[0] == UCOL_REORDER_CODE_NONE)) {
         ownedSettings->resetReordering();
     } else {
         uint8_t reorderTable[256];

icu4c/source/test/cintltst/cmsccoll.c

@@ -4691,6 +4691,7 @@ static void TestReorderingAPI(void)
     int32_t reorderCodes[3] = {USCRIPT_GREEK, USCRIPT_HAN, UCOL_REORDER_CODE_PUNCTUATION};
     int32_t duplicateReorderCodes[] = {USCRIPT_CUNEIFORM, USCRIPT_GREEK, UCOL_REORDER_CODE_CURRENCY, USCRIPT_EGYPTIAN_HIEROGLYPHS};
     int32_t reorderCodesStartingWithDefault[] = {UCOL_REORDER_CODE_DEFAULT, USCRIPT_GREEK, USCRIPT_HAN, UCOL_REORDER_CODE_PUNCTUATION};
+    int32_t reorderCodeNone = UCOL_REORDER_CODE_NONE;
     UCollationResult collResult;
     int32_t retrievedReorderCodesLength;
     int32_t retrievedReorderCodes[10];
@@ -4768,6 +4769,22 @@ static void TestReorderingAPI(void)
         return;
     }
 
+    /* clear the reordering using [NONE] */
+    ucol_setReorderCodes(myCollation, &reorderCodeNone, 1, &status);    
+    if (U_FAILURE(status)) {
+        log_err_status(status, "ERROR: setting reorder codes to [NONE]: %s\n", myErrorName(status));
+        return;
+    }
+
+    /* get the reordering again */
+    retrievedReorderCodesLength = ucol_getReorderCodes(myCollation, NULL, 0, &status);
+    if (retrievedReorderCodesLength != 0) {
+        log_err_status(status,
+                       "ERROR: [NONE] retrieved reorder codes length was %d but should have been 0\n",
+                       retrievedReorderCodesLength);
+        return;
+    }
+
     /* test for error condition on duplicate reorder codes */
     ucol_setReorderCodes(myCollation, duplicateReorderCodes, LEN(duplicateReorderCodes), &status);
     if (!U_FAILURE(status)) {

icu4c/source/test/intltest/collationtest.cpp

@@ -1173,11 +1173,15 @@ void CollationTest::parseAndSetReorderCodes(int32_t start, IcuTestErrorCode &err
         CharString name;
         name.appendInvariantChars(fileLine.tempSubStringBetween(start, limit), errorCode);
         int32_t code = CollationRuleParser::getReorderCode(name.data());
-        if(code < -1) {
-            errln("invalid reorder code '%s' on line %d", name.data(), (int)fileLineNumber);
-            infoln(fileLine);
-            errorCode.set(U_PARSE_ERROR);
-            return;
+        if(code < 0) {
+            if(uprv_stricmp(name.data(), "default") == 0) {
+                code = UCOL_REORDER_CODE_DEFAULT;  // -1
+            } else {
+                errln("invalid reorder code '%s' on line %d", name.data(), (int)fileLineNumber);
+                infoln(fileLine);
+                errorCode.set(U_PARSE_ERROR);
+                return;
+            }
         }
         reorderCodes.addElement(code, errorCode);
         start = limit;

icu4c/source/test/testdata/collationtest.txt

@@ -2439,3 +2439,28 @@
 <1 a
 <1 ae
 <2 ä
+
+** test: import rules from a language with non-Latin native script, and reset the reordering, ICU ticket 10998
+# Greek should sort Greek first.
+@ rules
+[import el]
+* compare
+<1 4
+<1 Ω
+<1 L
+
+# Import Greek, and then reset the reordering.
+@ rules
+[import el][reorder Zzzz]
+* compare
+<1 4
+<1 L
+<1 Ω
+
+# "others" is a synonym for Zzzz.
+@ rules
+[import el][reorder others]
+* compare
+<1 4
+<1 L
+<1 Ω