Note:
+ * - Most UnicodeSet methods do not take a UErrorCode parameter because
+ * there are usually very few opportunities for failure other than a shortage
+ * of memory, error codes in low-level C++ string methods would be inconvenient,
+ * and the error code as the last parameter (ICU convention) would prevent
+ * the use of default parameter values.
+ * Instead, such methods set the UnicodeSet into a "bogus" state
+ * (see isBogus()) if an error occurs.
*
* @author Alan Liu
* @stable ICU 2.0
@@ -257,11 +277,11 @@ class U_COMMON_API UnicodeSet : public UnicodeFilter {
int32_t len; // length of list used; 0 <= len <= capacity
int32_t capacity; // capacity of list
- int32_t bufferCapacity; // capacity of buffer
UChar32* list; // MUST be terminated with HIGH
+ BMPSet *bmpSet; // The set is frozen iff either bmpSet or stringSpan is not NULL.
UChar32* buffer; // internal buffer, may be NULL
-
- UVector* strings; // maintained in sorted order
+ int32_t bufferCapacity; // capacity of buffer
+ int32_t patLen;
/**
* The pattern representation of this set. This may not be the
@@ -272,29 +292,60 @@ class U_COMMON_API UnicodeSet : public UnicodeFilter {
* indicating that toPattern() must generate a pattern
* representation from the inversion list.
*/
- UnicodeString pat;
+ UChar *pat;
+ UVector* strings; // maintained in sorted order
+ UnicodeSetStringSpan *stringSpan;
+private:
+ enum { // constants
+ kIsBogus = 1 // This set is bogus (i.e. not valid)
+ };
+ uint8_t fFlags; // Bit flag (see constants above)
public:
-
/**
- * Minimum value that can be stored in a UnicodeSet.
- * @stable ICU 2.4
+ * Determine if this object contains a valid set.
+ * A bogus set has no value. It is different from an empty set.
+ * It can be used to indicate that no set value is available.
+ *
+ * @return TRUE if the set is valid, FALSE otherwise
+ * @see setToBogus()
+ * @stable ICU 4.0
*/
-#ifdef U_CYGWIN
- static U_COMMON_API const UChar32 MIN_VALUE;
-#else
- static const UChar32 MIN_VALUE;
-#endif
-
+ inline UBool isBogus(void) const;
+
/**
- * Maximum value that can be stored in a UnicodeSet.
- * @stable ICU 2.4
+ * Make this UnicodeSet object invalid.
+ * The string will test TRUE with isBogus().
+ *
+ * A bogus set has no value. It is different from an empty set.
+ * It can be used to indicate that no set value is available.
+ *
+ * This utility function is used throughout the UnicodeSet
+ * implementation to indicate that a UnicodeSet operation failed,
+ * and may be used in other functions,
+ * especially but not exclusively when such functions do not
+ * take a UErrorCode for simplicity.
+ *
+ * @see isBogus()
+ * @stable ICU 4.0
*/
-#ifdef U_CYGWIN
- static U_COMMON_API const UChar32 MAX_VALUE;
-#else
- static const UChar32 MAX_VALUE;
-#endif
+ void setToBogus();
+
+public:
+
+ enum {
+ /**
+ * Minimum value that can be stored in a UnicodeSet.
+ * @stable ICU 2.4
+ */
+ MIN_VALUE = 0,
+
+ /**
+ * Maximum value that can be stored in a UnicodeSet.
+ * @stable ICU 2.4
+ */
+ MAX_VALUE = 0x10ffff
+ };
//----------------------------------------------------------------
// Constructors &c
@@ -329,6 +380,7 @@ public:
UnicodeSet(const UnicodeString& pattern,
UErrorCode& status);
+#ifndef U_HIDE_INTERNAL_API
/**
* Constructs a set from the given pattern. See the class
* description for the syntax of the pattern language.
@@ -345,6 +397,7 @@ public:
uint32_t options,
const SymbolTable* symbols,
UErrorCode& status);
+#endif /* U_HIDE_INTERNAL_API */
/**
* Constructs a set from the given pattern. See the class description
@@ -357,23 +410,13 @@ public:
* @param symbols a symbol table mapping variable names to values
* and stand-in characters to UnicodeSets; may be NULL
* @param status input-output error code
- * @draft ICU 2.8
+ * @stable ICU 2.8
*/
UnicodeSet(const UnicodeString& pattern, ParsePosition& pos,
uint32_t options,
const SymbolTable* symbols,
UErrorCode& status);
-#ifdef U_USE_UNICODESET_DEPRECATES
- /**
- * Obsolete: Constructs a set from the given Unicode character category.
- * @param category an integer indicating the character category as
- * defined in uchar.h.
- * @obsolete ICU 2.6. Use a pattern with the category instead since this API will be removed in that release.
- */
- UnicodeSet(int8_t category, UErrorCode& status);
-#endif
-
/**
* Constructs a set that is identical to the given UnicodeSet.
* @stable ICU 2.0
@@ -388,6 +431,7 @@ public:
/**
* Assigns this object to be a copy of another.
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
UnicodeSet& operator=(const UnicodeSet& o);
@@ -416,6 +460,9 @@ public:
* Returns a copy of this object. All UnicodeFunctor objects have
* to support cloning in order to allow classes using
* UnicodeFunctors, such as Transliterator, to implement cloning.
+ * If this set is frozen, then the clone will be frozen as well.
+ * Use cloneAsThawed() for a mutable clone of a frozen set.
+ * @see cloneAsThawed
* @stable ICU 2.0
*/
virtual UnicodeFunctor* clone() const;
@@ -429,6 +476,85 @@ public:
*/
virtual int32_t hashCode(void) const;
+ /**
+ * Get a UnicodeSet pointer from a USet
+ *
+ * @param uset a USet (the ICU plain C type for UnicodeSet)
+ * @return the corresponding UnicodeSet pointer.
+ *
+ * @stable ICU 4.2
+ */
+ inline static UnicodeSet *fromUSet(USet *uset);
+
+ /**
+ * Get a UnicodeSet pointer from a const USet
+ *
+ * @param uset a const USet (the ICU plain C type for UnicodeSet)
+ * @return the corresponding UnicodeSet pointer.
+ *
+ * @stable ICU 4.2
+ */
+ inline static const UnicodeSet *fromUSet(const USet *uset);
+
+ /**
+ * Produce a USet * pointer for this UnicodeSet.
+ * USet is the plain C type for UnicodeSet
+ *
+ * @return a USet pointer for this UnicodeSet
+ * @stable ICU 4.2
+ */
+ inline USet *toUSet();
+
+
+ /**
+ * Produce a const USet * pointer for this UnicodeSet.
+ * USet is the plain C type for UnicodeSet
+ *
+ * @return a const USet pointer for this UnicodeSet
+ * @stable ICU 4.2
+ */
+ inline const USet * toUSet() const;
+
+
+ //----------------------------------------------------------------
+ // Freezable API
+ //----------------------------------------------------------------
+
+ /**
+ * Determines whether the set has been frozen (made immutable) or not.
+ * See the ICU4J Freezable interface for details.
+ * @return TRUE/FALSE for whether the set has been frozen
+ * @see freeze
+ * @see cloneAsThawed
+ * @stable ICU 3.8
+ */
+ inline UBool isFrozen() const;
+
+ /**
+ * Freeze the set (make it immutable).
+ * Once frozen, it cannot be unfrozen and is therefore thread-safe
+ * until it is deleted.
+ * See the ICU4J Freezable interface for details.
+ * Freezing the set may also make some operations faster, for example
+ * contains() and span().
+ * A frozen set will not be modified. (It remains frozen.)
+ * @return this set.
+ * @see isFrozen
+ * @see cloneAsThawed
+ * @stable ICU 3.8
+ */
+ UnicodeFunctor *freeze();
+
+ /**
+ * Clone the set and make the clone mutable.
+ * See the ICU4J Freezable interface for details.
+ * @return the mutable clone
+ * @see freeze
+ * @see isFrozen
+ * @stable ICU 3.8
+ */
+ UnicodeFunctor *cloneAsThawed() const;
+
//----------------------------------------------------------------
// Public API
//----------------------------------------------------------------
@@ -437,6 +563,7 @@ public:
* Make this object represent the range start - end.
* If end > start then this object is set to an
* an empty range.
+ * A frozen set will not be modified.
*
* @param start first character in the set, inclusive
* @param end last character in the set, inclusive
@@ -454,8 +581,9 @@ public:
/**
* Modifies this set to represent the set specified by the given
- * pattern, optionally ignoring white space. See the class
- * description for the syntax of the pattern language.
+ * pattern, ignoring Unicode Pattern_White_Space characters.
+ * See the class description for the syntax of the pattern language.
+ * A frozen set will not be modified.
* @param pattern a string specifying what characters are in the set
* @param status returns U_ILLEGAL_ARGUMENT_ERROR if the pattern
* contains a syntax error.
@@ -466,10 +594,12 @@ public:
UnicodeSet& applyPattern(const UnicodeString& pattern,
UErrorCode& status);
+#ifndef U_HIDE_INTERNAL_API
/**
* Modifies this set to represent the set specified by the given
- * pattern, optionally ignoring white space. See the class
- * description for the syntax of the pattern language.
+ * pattern, optionally ignoring Unicode Pattern_White_Space characters.
+ * See the class description for the syntax of the pattern language.
+ * A frozen set will not be modified.
* @param pattern a string specifying what characters are in the set
* @param options bitmask for options to apply to the pattern.
* Valid options are USET_IGNORE_SPACE and USET_CASE_INSENSITIVE.
@@ -485,6 +615,7 @@ public:
uint32_t options,
const SymbolTable* symbols,
UErrorCode& status);
+#endif /* U_HIDE_INTERNAL_API */
/**
* Parses the given pattern, starting at the given position. The
@@ -497,6 +628,7 @@ public:
* pairs list for the parsed pattern is returned. This method calls
* itself recursively to parse embedded subpatterns.
* Empties the set passed before applying the pattern.
+ * A frozen set will not be modified.
*
* @param pattern the string containing the pattern to be parsed.
* The portion of the string from pos.getIndex(), which must be a
@@ -514,7 +646,7 @@ public:
* @param status returns U_ILLEGAL_ARGUMENT_ERROR if the pattern
* contains a syntax error.
* @return a reference to this
- * @draft ICU 2.8
+ * @stable ICU 2.8
*/
UnicodeSet& applyPattern(const UnicodeString& pattern,
ParsePosition& pos,
@@ -526,6 +658,7 @@ public:
* Returns a string representation of this set. If the result of
* calling this function is passed to a UnicodeSet constructor, it
* will produce another set that is equal to this one.
+ * A frozen set will not be modified.
* @param result the string to receive the rules. Previous
* contents will be deleted.
* @param escapeUnprintable if TRUE then convert unprintable
@@ -541,6 +674,7 @@ public:
* Modifies this set to contain those code points which have the given value
* for the given binary or enumerated property, as returned by
* u_getIntPropertyValue. Prior contents of this set are lost.
+ * A frozen set will not be modified.
*
* @param prop a property in the range UCHAR_BIN_START..UCHAR_BIN_LIMIT-1
* or UCHAR_INT_START..UCHAR_INT_LIMIT-1
@@ -566,6 +700,7 @@ public:
* Modifies this set to contain those code points which have the
* given value for the given property. Prior contents of this
* set are lost.
+ * A frozen set will not be modified.
*
* @param prop a property alias, either short or long. The name is matched
* loosely. See PropertyAliases.txt for names and a description of loose
@@ -575,7 +710,8 @@ public:
* correspond to the following sets:
*
* "ANY" = [\\u0000-\\U0010FFFF],
- * "ASCII" = [\\u0000-\\u007F].
+ * "ASCII" = [\\u0000-\\u007F],
+ * "Assigned" = [:^Cn:].
*
* @param value a value alias, either short or long. The name is matched
* loosely. See PropertyValueAliases.txt for names and a description of
@@ -613,6 +749,7 @@ public:
/**
* Returns true if this set contains the given character.
+ * This function works faster with a frozen set.
* @param c character to be checked for containment
* @return true if the test condition is met
* @stable ICU 2.0
@@ -712,6 +849,113 @@ public:
*/
inline UBool containsSome(const UnicodeString& s) const;
+ /**
+ * Returns the length of the initial substring of the input string which
+ * consists only of characters and strings that are contained in this set
+ * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
+ * or only of characters and strings that are not contained
+ * in this set (USET_SPAN_NOT_CONTAINED).
+ * See USetSpanCondition for details.
+ * Similar to the strspn() C library function.
+ * Unpaired surrogates are treated according to contains() of their surrogate code points.
+ * This function works faster with a frozen set and with a non-negative string length argument.
+ * @param s start of the string
+ * @param length of the string; can be -1 for NUL-terminated
+ * @param spanCondition specifies the containment condition
+ * @return the length of the initial substring according to the spanCondition;
+ * 0 if the start of the string does not fit the spanCondition
+ * @stable ICU 3.8
+ * @see USetSpanCondition
+ */
+ int32_t span(const UChar *s, int32_t length, USetSpanCondition spanCondition) const;
+
+ /**
+ * Returns the end of the substring of the input string according to the USetSpanCondition.
+ * Same as start+span(s.getBuffer()+start, s.length()-start, spanCondition)
+ * after pinning start to 0<=start<=s.length().
+ * @param s the string
+ * @param start the start index in the string for the span operation
+ * @param spanCondition specifies the containment condition
+ * @return the exclusive end of the substring according to the spanCondition;
+ * the substring s.tempSubStringBetween(start, end) fulfills the spanCondition
+ * @stable ICU 4.4
+ * @see USetSpanCondition
+ */
+ inline int32_t span(const UnicodeString &s, int32_t start, USetSpanCondition spanCondition) const;
+
+ /**
+ * Returns the start of the trailing substring of the input string which
+ * consists only of characters and strings that are contained in this set
+ * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
+ * or only of characters and strings that are not contained
+ * in this set (USET_SPAN_NOT_CONTAINED).
+ * See USetSpanCondition for details.
+ * Unpaired surrogates are treated according to contains() of their surrogate code points.
+ * This function works faster with a frozen set and with a non-negative string length argument.
+ * @param s start of the string
+ * @param length of the string; can be -1 for NUL-terminated
+ * @param spanCondition specifies the containment condition
+ * @return the start of the trailing substring according to the spanCondition;
+ * the string length if the end of the string does not fit the spanCondition
+ * @stable ICU 3.8
+ * @see USetSpanCondition
+ */
+ int32_t spanBack(const UChar *s, int32_t length, USetSpanCondition spanCondition) const;
+
+ /**
+ * Returns the start of the substring of the input string according to the USetSpanCondition.
+ * Same as spanBack(s.getBuffer(), limit, spanCondition)
+ * after pinning limit to 0<=end<=s.length().
+ * @param s the string
+ * @param limit the exclusive-end index in the string for the span operation
+ * (use s.length() or INT32_MAX for spanning back from the end of the string)
+ * @param spanCondition specifies the containment condition
+ * @return the start of the substring according to the spanCondition;
+ * the substring s.tempSubStringBetween(start, limit) fulfills the spanCondition
+ * @stable ICU 4.4
+ * @see USetSpanCondition
+ */
+ inline int32_t spanBack(const UnicodeString &s, int32_t limit, USetSpanCondition spanCondition) const;
+
+ /**
+ * Returns the length of the initial substring of the input string which
+ * consists only of characters and strings that are contained in this set
+ * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
+ * or only of characters and strings that are not contained
+ * in this set (USET_SPAN_NOT_CONTAINED).
+ * See USetSpanCondition for details.
+ * Similar to the strspn() C library function.
+ * Malformed byte sequences are treated according to contains(0xfffd).
+ * This function works faster with a frozen set and with a non-negative string length argument.
+ * @param s start of the string (UTF-8)
+ * @param length of the string; can be -1 for NUL-terminated
+ * @param spanCondition specifies the containment condition
+ * @return the length of the initial substring according to the spanCondition;
+ * 0 if the start of the string does not fit the spanCondition
+ * @stable ICU 3.8
+ * @see USetSpanCondition
+ */
+ int32_t spanUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const;
+
+ /**
+ * Returns the start of the trailing substring of the input string which
+ * consists only of characters and strings that are contained in this set
+ * (USET_SPAN_CONTAINED, USET_SPAN_SIMPLE),
+ * or only of characters and strings that are not contained
+ * in this set (USET_SPAN_NOT_CONTAINED).
+ * See USetSpanCondition for details.
+ * Malformed byte sequences are treated according to contains(0xfffd).
+ * This function works faster with a frozen set and with a non-negative string length argument.
+ * @param s start of the string (UTF-8)
+ * @param length of the string; can be -1 for NUL-terminated
+ * @param spanCondition specifies the containment condition
+ * @return the start of the trailing substring according to the spanCondition;
+ * the string length if the end of the string does not fit the spanCondition
+ * @stable ICU 3.8
+ * @see USetSpanCondition
+ */
+ int32_t spanBackUTF8(const char *s, int32_t length, USetSpanCondition spanCondition) const;
+
/**
* Implement UnicodeMatcher::matches()
* @stable ICU 2.4
@@ -738,6 +982,7 @@ private:
* @param limit the limit offset for matching, either last+1 in
* the forward direction, or last-1 in the reverse direction,
* where last is the index of the last character to match.
+ * @param s
* @return If part of s matches up to the limit, return |limit -
* start|. If all of s matches before reaching the limit, return
* s.length(). If there is a mismatch between s and text, return
@@ -796,6 +1041,7 @@ public:
* the call leaves this set unchanged. If end > start
* then an empty range is added, leaving the set unchanged.
* This is equivalent to a boolean logic OR, or a set UNION.
+ * A frozen set will not be modified.
*
* @param start first character, inclusive, of range to be added
* to this set.
@@ -809,6 +1055,7 @@ public:
* Adds the specified character to this set if it is not already
* present. If this set already contains the specified character,
* the call leaves this set unchanged.
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
UnicodeSet& add(UChar32 c);
@@ -819,6 +1066,7 @@ public:
* the call leaves this set unchanged.
* Thus "ch" => {"ch"}
*
Warning: you cannot add an empty string ("") to a UnicodeSet.
+ * A frozen set will not be modified.
* @param s the source string
* @return this object, for chaining
* @stable ICU 2.4
@@ -829,7 +1077,7 @@ public:
/**
* @return a code point IF the string consists of a single one.
* otherwise returns -1.
- * @param string to test
+ * @param s string to test
*/
static int32_t getSingleCP(const UnicodeString& s);
@@ -839,6 +1087,7 @@ public:
/**
* Adds each of the characters in this string to the set. Thus "ch" => {"c", "h"}
* If this set already any particular character, it has no effect on that character.
+ * A frozen set will not be modified.
* @param s the source string
* @return this object, for chaining
* @stable ICU 2.4
@@ -848,6 +1097,7 @@ public:
/**
* Retains EACH of the characters in this string. Note: "ch" == {"c", "h"}
* If this set already any particular character, it has no effect on that character.
+ * A frozen set will not be modified.
* @param s the source string
* @return this object, for chaining
* @stable ICU 2.4
@@ -857,6 +1107,7 @@ public:
/**
* Complement EACH of the characters in this string. Note: "ch" == {"c", "h"}
* If this set already any particular character, it has no effect on that character.
+ * A frozen set will not be modified.
* @param s the source string
* @return this object, for chaining
* @stable ICU 2.4
@@ -866,6 +1117,7 @@ public:
/**
* Remove EACH of the characters in this string. Note: "ch" == {"c", "h"}
* If this set already any particular character, it has no effect on that character.
+ * A frozen set will not be modified.
* @param s the source string
* @return this object, for chaining
* @stable ICU 2.4
@@ -897,6 +1149,7 @@ public:
* specified range. If end > start then an empty range is
* retained, leaving the set empty. This is equivalent to
* a boolean logic AND, or a set INTERSECTION.
+ * A frozen set will not be modified.
*
* @param start first character, inclusive, of range to be retained
* to this set.
@@ -909,6 +1162,7 @@ public:
/**
* Retain the specified character from this set if it is present.
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
UnicodeSet& retain(UChar32 c);
@@ -918,6 +1172,7 @@ public:
* The set will not contain the specified range once the call
* returns. If end > start then an empty range is
* removed, leaving the set unchanged.
+ * A frozen set will not be modified.
*
* @param start first character, inclusive, of range to be removed
* from this set.
@@ -931,6 +1186,7 @@ public:
* Removes the specified character from this set if it is present.
* The set will not contain the specified range once the call
* returns.
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
UnicodeSet& remove(UChar32 c);
@@ -939,6 +1195,7 @@ public:
* Removes the specified string from this set if it is present.
* The set will not contain the specified character once the call
* returns.
+ * A frozen set will not be modified.
* @param s the source string
* @return this object, for chaining
* @stable ICU 2.4
@@ -949,6 +1206,7 @@ public:
* Inverts this set. This operation modifies this set so that
* its value is its complement. This is equivalent to
* complement(MIN_VALUE, MAX_VALUE).
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
virtual UnicodeSet& complement(void);
@@ -959,6 +1217,7 @@ public:
* added if it is not in this set. If end > start
* then an empty range is complemented, leaving the set unchanged.
* This is equivalent to a boolean logic XOR.
+ * A frozen set will not be modified.
*
* @param start first character, inclusive, of range to be removed
* from this set.
@@ -972,6 +1231,7 @@ public:
* Complements the specified character in this set. The character
* will be removed if it is in this set, or will be added if it is
* not in this set.
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
UnicodeSet& complement(UChar32 c);
@@ -981,6 +1241,7 @@ public:
* The set will not contain the specified string once the call
* returns.
*
Warning: you cannot add an empty string ("") to a UnicodeSet.
+ * A frozen set will not be modified.
* @param s the string to complement
* @return this object, for chaining
* @stable ICU 2.4
@@ -993,9 +1254,10 @@ public:
* modifies this set so that its value is the union of the two
* sets. The behavior of this operation is unspecified if the specified
* collection is modified while the operation is in progress.
+ * A frozen set will not be modified.
*
* @param c set whose elements are to be added to this set.
- * @see #add(char, char)
+ * @see #add(UChar32, UChar32)
* @stable ICU 2.0
*/
virtual UnicodeSet& addAll(const UnicodeSet& c);
@@ -1006,6 +1268,7 @@ public:
* its elements that are not contained in the specified set. This
* operation effectively modifies this set so that its value is
* the intersection of the two sets.
+ * A frozen set will not be modified.
*
* @param c set that defines which elements this set will retain.
* @stable ICU 2.0
@@ -1017,6 +1280,7 @@ public:
* specified set. This operation effectively modifies this
* set so that its value is the asymmetric set difference of
* the two sets.
+ * A frozen set will not be modified.
*
* @param c set that defines which elements will be removed from
* this set.
@@ -1028,6 +1292,7 @@ public:
* Complements in this set all elements contained in the specified
* set. Any character in the other set will be removed if it is
* in this set, or will be added if it is not in this set.
+ * A frozen set will not be modified.
*
* @param c set that defines which elements will be xor'ed from
* this set.
@@ -1038,6 +1303,7 @@ public:
/**
* Removes all of the elements from this set. This set will be
* empty after this call returns.
+ * A frozen set will not be modified.
* @stable ICU 2.0
*/
virtual UnicodeSet& clear(void);
@@ -1059,14 +1325,24 @@ public:
* == b denotes that the contents are the same, not pointer
* comparison.)
*
+ * A frozen set will not be modified.
+ *
* @param attribute bitmask for attributes to close over.
* Currently only the USET_CASE bit is supported. Any undefined bits
* are ignored.
* @return a reference to this set.
- * @internal
+ * @stable ICU 4.2
*/
UnicodeSet& closeOver(int32_t attribute);
+ /**
+ * Remove all strings from this set.
+ *
+ * @return a reference to this set.
+ * @stable ICU 4.2
+ */
+ virtual UnicodeSet &removeAllStrings();
+
/**
* Iteration method that returns the number of ranges contained in
* this set.
@@ -1147,6 +1423,7 @@ public:
/**
* Reallocate this objects internal structures to take up the least
* possible space, without changing this object's value.
+ * A frozen set will not be modified.
* @stable ICU 2.4
*/
virtual UnicodeSet& compact();
@@ -1198,28 +1475,41 @@ private:
virtual UBool matchesIndexValue(uint8_t v) const;
private:
+ friend class RBBIRuleScanner;
+
+ //----------------------------------------------------------------
+ // Implementation: Clone as thawed (see ICU4J Freezable)
+ //----------------------------------------------------------------
+
+ UnicodeSet(const UnicodeSet& o, UBool /* asThawed */);
//----------------------------------------------------------------
// Implementation: Pattern parsing
//----------------------------------------------------------------
+ void applyPatternIgnoreSpace(const UnicodeString& pattern,
+ ParsePosition& pos,
+ const SymbolTable* symbols,
+ UErrorCode& status);
+
void applyPattern(RuleCharacterIterator& chars,
const SymbolTable* symbols,
UnicodeString& rebuiltPat,
uint32_t options,
+ UnicodeSet& (UnicodeSet::*caseClosure)(int32_t attribute),
UErrorCode& ec);
//----------------------------------------------------------------
// Implementation: Utility methods
//----------------------------------------------------------------
- void ensureCapacity(int32_t newLen);
+ void ensureCapacity(int32_t newLen, UErrorCode& ec);
- void ensureBufferCapacity(int32_t newLen);
+ void ensureBufferCapacity(int32_t newLen, UErrorCode& ec);
void swapBuffers(void);
- UBool allocateStrings();
+ UBool allocateStrings(UErrorCode &status);
UnicodeString& _toPattern(UnicodeString& result,
UBool escapeUnprintable) const;
@@ -1258,7 +1548,7 @@ private:
*
* The original design document is out of date, but still useful.
* Ignore the property and value names:
- * http://oss.software.ibm.com/cvs/icu/~checkout~/icuhtml/design/unicodeset_properties.html
+ * http://source.icu-project.org/repos/icu/icuhtml/trunk/design/unicodeset_properties.html
*
* Recognized syntax:
*
@@ -1266,8 +1556,8 @@ private:
* \\p{foo} \\P{foo} - white space not allowed within "\\p" or "\\P"
* \\N{name} - white space not allowed within "\\N"
*
- * Other than the above restrictions, white space is ignored. Case
- * is ignored except in "\\p" and "\\P" and "\\N". In 'name' leading
+ * Other than the above restrictions, Unicode Pattern_White_Space characters are ignored.
+ * Case is ignored except in "\\p" and "\\P" and "\\N". In 'name' leading
* and trailing space is deleted, and internal runs of whitespace
* are collapsed to a single space.
*
@@ -1288,6 +1578,7 @@ private:
* On return, the position after the last character parsed, that is,
* the locations marked '%'. If the parse fails, ppos is returned
* unchanged.
+ * @param ec status
* @return a reference to this.
*/
UnicodeSet& applyPropertyPattern(const UnicodeString& pattern,
@@ -1298,6 +1589,9 @@ private:
UnicodeString& rebuiltPat,
UErrorCode& ec);
+ friend void U_CALLCONV UnicodeSet_initInclusion(int32_t src, UErrorCode &status);
+ static const UnicodeSet* getInclusions(int32_t src, UErrorCode &status);
+
/**
* A filter that returns TRUE if the given code point should be
* included in the UnicodeSet being constructed.
@@ -1319,31 +1613,27 @@ private:
UErrorCode &status);
/**
- * Return a cached copy of the inclusions list for the property source.
+ * Set the new pattern to cache.
*/
- static const UnicodeSet* getInclusions(int32_t src, UErrorCode &errorCode);
+ void setPattern(const UnicodeString& newPat);
+ /**
+ * Release existing cached pattern.
+ */
+ void releasePattern();
friend class UnicodeSetIterator;
+};
- //----------------------------------------------------------------
- // Implementation: closeOver
- //----------------------------------------------------------------
-
- void caseCloseOne(const UnicodeString& folded);
-
- void caseCloseOne(const CaseEquivClass& c);
-
- void caseCloseOne(UChar folded);
-
- static const CaseEquivClass* getCaseMapOf(const UnicodeString& folded);
- static const CaseEquivClass* getCaseMapOf(UChar folded);
-};
inline UBool UnicodeSet::operator!=(const UnicodeSet& o) const {
return !operator==(o);
}
+inline UBool UnicodeSet::isFrozen() const {
+ return (UBool)(bmpSet!=NULL || stringSpan!=NULL);
+}
+
inline UBool UnicodeSet::containsSome(UChar32 start, UChar32 end) const {
return !containsNone(start, end);
}
@@ -1356,6 +1646,46 @@ inline UBool UnicodeSet::containsSome(const UnicodeString& s) const {
return !containsNone(s);
}
+inline UBool UnicodeSet::isBogus() const {
+ return (UBool)(fFlags & kIsBogus);
+}
+
+inline UnicodeSet *UnicodeSet::fromUSet(USet *uset) {
+ return reinterpret_cast