]> git.saurik.com Git - apple/icu.git/blobdiff - icuSources/common/ruleiter.cpp
ICU-6.2.4.tar.gz
[apple/icu.git] / icuSources / common / ruleiter.cpp
diff --git a/icuSources/common/ruleiter.cpp b/icuSources/common/ruleiter.cpp
new file mode 100644 (file)
index 0000000..7a084aa
--- /dev/null
@@ -0,0 +1,151 @@
+/*
+**********************************************************************
+* Copyright (c) 2003-2004, International Business Machines
+* Corporation and others.  All Rights Reserved.
+**********************************************************************
+* Author: Alan Liu
+* Created: September 24 2003
+* Since: ICU 2.8
+**********************************************************************
+*/
+#include "ruleiter.h"
+#include "unicode/parsepos.h"
+#include "unicode/unistr.h"
+#include "unicode/symtable.h"
+#include "util.h"
+
+U_NAMESPACE_BEGIN
+
+RuleCharacterIterator::RuleCharacterIterator(const UnicodeString& theText, const SymbolTable* theSym,
+                      ParsePosition& thePos) :
+    text(theText),
+    pos(thePos),
+    sym(theSym),
+    buf(0)
+{}
+
+UBool RuleCharacterIterator::atEnd() const {
+    return buf == 0 && pos.getIndex() == text.length();
+}
+
+UChar32 RuleCharacterIterator::next(int32_t options, UBool& isEscaped, UErrorCode& ec) {
+    if (U_FAILURE(ec)) return DONE;
+
+    UChar32 c = DONE;
+    isEscaped = FALSE;
+
+    for (;;) {
+        c = _current();
+        _advance(UTF_CHAR_LENGTH(c));
+
+        if (c == SymbolTable::SYMBOL_REF && buf == 0 &&
+            (options & PARSE_VARIABLES) != 0 && sym != 0) {
+            UnicodeString name = sym->parseReference(text, pos, text.length());
+            // If name is empty there was an isolated SYMBOL_REF;
+            // return it.  Caller must be prepared for this.
+            if (name.length() == 0) {
+                break;
+            }
+            bufPos = 0;
+            buf = sym->lookup(name);
+            if (buf == 0) {
+                ec = U_UNDEFINED_VARIABLE;
+                return DONE;
+            }
+            // Handle empty variable value
+            if (buf->length() == 0) {
+                buf = 0;
+            }
+            continue;
+        }
+
+        if ((options & SKIP_WHITESPACE) != 0 &&
+            uprv_isRuleWhiteSpace(c)) {
+            continue;
+        }
+
+        if (c == 0x5C /*'\\'*/ && (options & PARSE_ESCAPES) != 0) {
+            UnicodeString s;
+            int32_t offset = 0;
+            c = lookahead(s).unescapeAt(offset);
+            jumpahead(offset);
+            isEscaped = TRUE;
+            if (c < 0) {
+                ec = U_MALFORMED_UNICODE_ESCAPE;
+                return DONE;
+            }
+        }
+
+        break;
+    }
+
+    return c;
+}
+
+void RuleCharacterIterator::getPos(RuleCharacterIterator::Pos& p) const {
+    p.buf = buf;
+    p.pos = pos.getIndex();
+    p.bufPos = bufPos;
+}
+
+void RuleCharacterIterator::setPos(const RuleCharacterIterator::Pos& p) {
+    buf = p.buf;
+    pos.setIndex(p.pos);
+    bufPos = p.bufPos;
+}
+
+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));
+        }
+    }
+}
+
+UnicodeString& RuleCharacterIterator::lookahead(UnicodeString& result) const {
+    if (buf != 0) {
+        buf->extract(bufPos, 0x7FFFFFFF, result);
+    } else {
+        text.extract(pos.getIndex(), 0x7FFFFFFF, result);
+    }
+    return result;
+}
+
+void RuleCharacterIterator::jumpahead(int32_t count) {
+    _advance(count);
+}
+
+UnicodeString& RuleCharacterIterator::toString(UnicodeString& result) const {
+    int32_t b = pos.getIndex();
+    text.extract(0, b, result);
+    return result.append((UChar) 0x7C /*'|'*/).append(text, b, 0x7FFFFFFF);
+}
+
+UChar32 RuleCharacterIterator::_current() const {
+    if (buf != 0) {
+        return buf->char32At(bufPos);
+    } else {
+        int i = pos.getIndex();
+        return (i < text.length()) ? text.char32At(i) : (UChar32)DONE;
+    }
+}
+
+void RuleCharacterIterator::_advance(int32_t count) {
+    if (buf != 0) {
+        bufPos += count;
+        if (bufPos == buf->length()) {
+            buf = 0;
+        }
+    } else {
+        pos.setIndex(pos.getIndex() + count);
+        if (pos.getIndex() > text.length()) {
+            pos.setIndex(text.length());
+        }
+    }
+}
+
+U_NAMESPACE_END
+
+//eof