ICU-57163.0.1.tar.gz

[apple/icu.git] / icuSources / common / ruleiter.cpp

diff --git a/icuSources/common/ruleiter.cpp b/icuSources/common/ruleiter.cpp
index 30f136c0828d930d3cd9712bd980617659e18ab2..667795efabcf39632c79a8c2480a8656099bce02 100644 (file)
--- a/icuSources/common/ruleiter.cpp
+++ b/icuSources/common/ruleiter.cpp
@@ -1,6 +1,6 @@
 /*
 **********************************************************************
-* Copyright (c) 2003-2005, International Business Machines
+* Copyright (c) 2003-2011, International Business Machines
 * Corporation and others.  All Rights Reserved.
 **********************************************************************
 * Author: Alan Liu
@@ -10,9 +10,13 @@
 */
 #include "ruleiter.h"
 #include "unicode/parsepos.h"
-#include "unicode/unistr.h"
 #include "unicode/symtable.h"
-#include "util.h"
+#include "unicode/unistr.h"
+#include "unicode/utf16.h"
+#include "patternprops.h"
+
+/* \U87654321 or \ud800\udc00 */
+#define MAX_U_NOTATION_LEN 12
 
 U_NAMESPACE_BEGIN
 
@@ -21,7 +25,8 @@ RuleCharacterIterator::RuleCharacterIterator(const UnicodeString& theText, const
     text(theText),
     pos(thePos),
     sym(theSym),
-    buf(0)
+    buf(0),
+    bufPos(0)
 {}
 
 UBool RuleCharacterIterator::atEnd() const {
@@ -36,7 +41,7 @@ UChar32 RuleCharacterIterator::next(int32_t options, UBool& isEscaped, UErrorCod
 
     for (;;) {
         c = _current();
-        _advance(UTF_CHAR_LENGTH(c));
+        _advance(U16_LENGTH(c));
 
         if (c == SymbolTable::SYMBOL_REF && buf == 0 &&
             (options & PARSE_VARIABLES) != 0 && sym != 0) {
@@ -59,15 +64,14 @@ UChar32 RuleCharacterIterator::next(int32_t options, UBool& isEscaped, UErrorCod
             continue;
         }
 
-        if ((options & SKIP_WHITESPACE) != 0 &&
-            uprv_isRuleWhiteSpace(c)) {
+        if ((options & SKIP_WHITESPACE) != 0 && PatternProps::isWhiteSpace(c)) {
             continue;
         }
 
         if (c == 0x5C /*'\\'*/ && (options & PARSE_ESCAPES) != 0) {
-            UnicodeString s;
+            UnicodeString tempEscape;
             int32_t offset = 0;
-            c = lookahead(s).unescapeAt(offset);
+            c = lookahead(tempEscape, MAX_U_NOTATION_LEN).unescapeAt(offset);
             jumpahead(offset);
             isEscaped = TRUE;
             if (c < 0) {
@@ -98,17 +102,20 @@ void RuleCharacterIterator::skipIgnored(int32_t options) {
     if ((options & SKIP_WHITESPACE) != 0) {
         for (;;) {
             UChar32 a = _current();
-            if (!uprv_isRuleWhiteSpace(a)) break;
-            _advance(UTF_CHAR_LENGTH(a));
+            if (!PatternProps::isWhiteSpace(a)) break;
+            _advance(U16_LENGTH(a));
         }
     }
 }
 
-UnicodeString& RuleCharacterIterator::lookahead(UnicodeString& result) const {
+UnicodeString& RuleCharacterIterator::lookahead(UnicodeString& result, int32_t maxLookAhead) const {
+    if (maxLookAhead < 0) {
+        maxLookAhead = 0x7FFFFFFF;
+    }
     if (buf != 0) {
-        buf->extract(bufPos, 0x7FFFFFFF, result);
+        buf->extract(bufPos, maxLookAhead, result);
     } else {
-        text.extract(pos.getIndex(), 0x7FFFFFFF, result);
+        text.extract(pos.getIndex(), maxLookAhead, result);
     }
     return result;
 }