- * The UCharacter class provides extensions to the
- *
+ * The UCharacter class provides extensions to the
+ *
* java.lang.Character class. These extensions provide support for
* more Unicode properties and together with the UTF16
* class, provide support for supplementary characters (those with code
* points above U+FFFF).
* Each ICU release supports the latest version of Unicode available at that time.
- *
- * Code points are represented in these API using ints. While it would be
+ *
+ * Code points are represented in these API using ints. While it would be
* more convenient in Java to have a separate primitive datatype for them,
* ints suffice in the meantime.
- *
- * To use this class please add the jar file name icu4j.jar to the
+ *
+ * To use this class please add the jar file name icu4j.jar to the
* class path, since it contains data files which supply the information used
* by this file.
- * Aside from the additions for UTF-16 support, and the updated Unicode
+ *
+ * Aside from the additions for UTF-16 support, and the updated Unicode
* properties, the main differences between UCharacter and Character are:
*
- * Further detail differences can be determined from the program
- *
+ * Further detail on differences can be determined using the program
+ *
* com.ibm.icu.dev.test.lang.UCharacterCompare
*
@@ -103,8 +93,11 @@
*
* For more information see
- * "About the Unicode Character Database" (http://www.unicode.org/ucd/)
- * and the ICU User Guide chapter on Properties (http://www.icu-project.org/userguide/properties.html).
+ * "About the Unicode Character Database"
+ * (http://www.unicode.org/ucd/)
+ * and the ICU
+ * User Guide chapter on Properties
+ * (http://www.icu-project.org/userguide/properties.html).
*
* There are also functions that provide easy migration from C/POSIX functions
@@ -128,12 +121,15 @@
* Annex C: Compatibility Properties of UTS #18 Unicode Regular Expressions
* (http://www.unicode.org/reports/tr18/#Compatibility_Properties).
*
* API access for C/POSIX character classes is as follows:
+ *
* E.g. In Windows
@@ -64,9 +54,8 @@
* unames.icu from the icu4j source subdirectory
* $ICU4J_SRC/src/com.ibm.icu.impl.data to your class directory
* $ICU4J_CLASS/com.ibm.icu.impl.data.
- *
*
* {@code
+ * {@code
* - alpha: isUAlphabetic(c) or hasBinaryProperty(c, UProperty.ALPHABETIC)
* - lower: isULowercase(c) or hasBinaryProperty(c, UProperty.LOWERCASE)
* - upper: isUUppercase(c) or hasBinaryProperty(c, UProperty.UPPERCASE)
- * - punct: ((1<
* The C/POSIX character classes are also available in UnicodeSet patterns, * using patterns like [:graph:] or \p{graph}. *
- *- * Note: There are several ICU (and Java) whitespace functions. - * Comparison: - * - isUWhiteSpace=UCHAR_WHITE_SPACE: Unicode White_Space property; + * + * There are several ICU (and Java) whitespace functions. + * Comparison:
- * This class is not subclassable + * This class is not subclassable. *
* @author Syn Wee Quek * @stable ICU 2.1 @@ -168,6 +165,19 @@ { /** + * Joining Group constants. + * @see UProperty#JOINING_GROUP + * @stable ICU 2.4 + */ + public static interface JoiningGroup + { + /** + * @stable ICU 2.4 + */ + public static final int NO_JOINING_GROUP = 0; + } + + /** * Numeric Type constants. * @see UProperty#NUMERIC_TYPE * @stable ICU 2.4 @@ -177,7 +187,61 @@ /** * @stable ICU 2.4 */ + public static final int NONE = 0; + /** + * @stable ICU 2.4 + */ public static final int DECIMAL = 1; + /** + * @stable ICU 2.4 + */ + public static final int DIGIT = 2; + /** + * @stable ICU 2.4 + */ + public static final int NUMERIC = 3; + /** + * @stable ICU 2.4 + */ + public static final int COUNT = 4; + } + + /** + * Hangul Syllable Type constants. + * + * @see UProperty#HANGUL_SYLLABLE_TYPE + * @stable ICU 2.6 + */ + public static interface HangulSyllableType + { + /** + * @stable ICU 2.6 + */ + public static final int NOT_APPLICABLE = 0; /*[NA]*/ /*See note !!*/ + /** + * @stable ICU 2.6 + */ + public static final int LEADING_JAMO = 1; /*[L]*/ + /** + * @stable ICU 2.6 + */ + public static final int VOWEL_JAMO = 2; /*[V]*/ + /** + * @stable ICU 2.6 + */ + public static final int TRAILING_JAMO = 3; /*[T]*/ + /** + * @stable ICU 2.6 + */ + public static final int LV_SYLLABLE = 4; /*[LV]*/ + /** + * @stable ICU 2.6 + */ + public static final int LVT_SYLLABLE = 5; /*[LVT]*/ + /** + * @stable ICU 2.6 + */ + public static final int COUNT = 6; } // public data members ----------------------------------------------- @@ -192,22 +256,15 @@ * The highest Unicode code point value (scalar value) according to the * Unicode Standard. * This is a 21-bit value (21 bits, rounded up).java.lang.Character.digit(). Note that this
* will return positive values for code points for which isDigit
@@ -231,15 +288,54 @@
*/
public static int digit(int ch, int radix)
{
- // when ch is out of bounds getProperty == 0
- int props = getProperty(ch);
- int value;
- if (getNumericType(props) == NumericType.DECIMAL) {
- value = UCharacterProperty.getUnsignedValue(props);
+ if (2 <= radix && radix <= 36) {
+ int value = digit(ch);
+ if (value < 0) {
+ // ch is not a decimal digit, try latin letters
+ value = UCharacterProperty.getEuropeanDigit(ch);
+ }
+ return (value < radix) ? value : -1;
} else {
- value = getEuropeanDigit(ch);
+ return -1; // invalid radix
}
- return (0 <= value && value < radix) ? value : -1;
+ }
+
+ /**
+ * Returns the numeric value of a decimal digit code point.
+ * digit(int, int)
+ * that provides a decimal radix.
+ * Get the "age" of the code point.
+ * Returns the "age" of the code point. *The "age" is the Unicode version when the code point was first * designated (as a non-character or for Private Use) or assigned a * character. @@ -289,143 +445,95 @@ public static VersionInfo getAge(int ch) { if (ch < MIN_VALUE || ch > MAX_VALUE) { - throw new IllegalArgumentException("Codepoint out of bounds"); + throw new IllegalArgumentException("Codepoint out of bounds"); } - return PROPERTY_.getAge(ch); + return UCharacterProperty.INSTANCE.getAge(ch); } - // private variables ------------------------------------------------- - - /** - * Database storing the sets of character property - */ - private static final UCharacterProperty PROPERTY_; /** - * For optimization + * Returns the property value for an Unicode property type of a code point. + * Also returns binary and mask property values.
+ *Unicode, especially in version 3.2, defines many more properties than + * the original set in UnicodeData.txt.
+ *The properties APIs are intended to reflect Unicode properties as + * defined in the Unicode Character Database (UCD) and Unicode Technical + * Reports (UTR). For details about the properties see + * http://www.unicode.org/.
+ *For names of Unicode properties see the UCD file PropertyAliases.txt. + *
+ *+ * Sample usage: + * int ea = UCharacter.getIntPropertyValue(c, UProperty.EAST_ASIAN_WIDTH); + * int ideo = UCharacter.getIntPropertyValue(c, UProperty.IDEOGRAPHIC); + * boolean b = (ideo == 1) ? true : false; + *+ * @param ch code point to test. + * @param type UProperty selector constant, identifies which binary + * property to check. Must be + * UProperty.BINARY_START <= type < UProperty.BINARY_LIMIT or + * UProperty.INT_START <= type < UProperty.INT_LIMIT or + * UProperty.MASK_START <= type < UProperty.MASK_LIMIT. + * @return numeric value that is directly the property value or, + * for enumerated properties, corresponds to the numeric value of + * the enumerated constant of the respective property value + * enumeration type (cast to enum type if necessary). + * Returns 0 or 1 (for false / true) for binary Unicode properties. + * Returns a bit-mask for mask properties. + * Returns 0 if 'type' is out of bounds or if the Unicode version + * does not have data for the property at all, or not for this code + * point. + * @see UProperty + * @see #hasBinaryProperty + * @see #getIntPropertyMinValue + * @see #getIntPropertyMaxValue + * @see #getUnicodeVersion + * @stable ICU 2.4 */ - private static final char[] PROPERTY_TRIE_INDEX_; - private static final char[] PROPERTY_TRIE_DATA_; - private static final int PROPERTY_INITIAL_VALUE_; - - private static final UBiDiProps gBdp; - - // block to initialise character property database - static - { - try - { - PROPERTY_ = UCharacterProperty.getInstance(); - PROPERTY_TRIE_INDEX_ = PROPERTY_.m_trieIndex_; - PROPERTY_TRIE_DATA_ = PROPERTY_.m_trieData_; - PROPERTY_INITIAL_VALUE_ = PROPERTY_.m_trieInitialValue_; - } - catch (Exception e) - { - throw new MissingResourceException(e.getMessage(),"",""); - } - - UBiDiProps bdp; - try { - bdp=UBiDiProps.getSingleton(); - } catch(IOException e) { - bdp=UBiDiProps.getDummy(); - } - gBdp=bdp; + // for BiDiBase.java + public static int getIntPropertyValue(int ch, int type) { + return UCharacterProperty.INSTANCE.getIntPropertyValue(ch, type); } - /** - * Shift to get numeric type - */ - private static final int NUMERIC_TYPE_SHIFT_ = 5; - /** - * Mask to get numeric type - */ - private static final int NUMERIC_TYPE_MASK_ = 0x7 << NUMERIC_TYPE_SHIFT_; - - // private methods --------------------------------------------------- + // private constructor ----------------------------------------------- /** - * Getting the digit values of characters like 'A' - 'Z', normal, - * half-width and full-width. This method assumes that the other digit - * characters are checked by the calling method. - * @param ch character to test - * @return -1 if ch is not a character of the form 'A' - 'Z', otherwise - * its corresponding digit will be returned. + * Private constructor to prevent instantiation */ - private static int getEuropeanDigit(int ch) { - if ((ch > 0x7a && ch < 0xff21) - || ch < 0x41 || (ch > 0x5a && ch < 0x61) - || ch > 0xff5a || (ch > 0xff3a && ch < 0xff41)) { - return -1; - } - if (ch <= 0x7a) { - // ch >= 0x41 or ch < 0x61 - return ch + 10 - ((ch <= 0x5a) ? 0x41 : 0x61); - } - // ch >= 0xff21 - if (ch <= 0xff3a) { - return ch + 10 - 0xff21; - } - // ch >= 0xff41 && ch <= 0xff5a - return ch + 10 - 0xff41; - } + private UCharacter() { } - /** - * Gets the numeric type of the property argument - * @param props 32 bit property - * @return the numeric type - */ - private static int getNumericType(int props) - { - return (props & NUMERIC_TYPE_MASK_) >> NUMERIC_TYPE_SHIFT_; - } + /* + * Copied from UCharacterEnums.java + */ - /** - * Gets the property value at the index. - * This is optimized. - * Note this is alittle different from CharTrie the index m_trieData_ - * is never negative. - * This is a duplicate of UCharacterProperty.getProperty. For optimization - * purposes, this method calls the trie data directly instead of through - * UCharacterProperty.getProperty. - * @param ch code point whose property value is to be retrieved - * @return property value of code point - * @stable ICU 2.6 - */ - private static final int getProperty(int ch) - { - if (ch < UTF16.LEAD_SURROGATE_MIN_VALUE - || (ch > UTF16.LEAD_SURROGATE_MAX_VALUE - && ch < UTF16.SUPPLEMENTARY_MIN_VALUE)) { - // BMP codepoint 0000..D7FF or DC00..FFFF - try { // using try for ch < 0 is faster than using an if statement - return PROPERTY_TRIE_DATA_[ - (PROPERTY_TRIE_INDEX_[ch >> 5] << 2) - + (ch & 0x1f)]; - } catch (ArrayIndexOutOfBoundsException e) { - return PROPERTY_INITIAL_VALUE_; - } - } - if (ch <= UTF16.LEAD_SURROGATE_MAX_VALUE) { - // lead surrogate D800..DBFF - return PROPERTY_TRIE_DATA_[ - (PROPERTY_TRIE_INDEX_[(0x2800 >> 5) + (ch >> 5)] << 2) - + (ch & 0x1f)]; - } - // for optimization - if (ch <= UTF16.CODEPOINT_MAX_VALUE) { - // supplementary code point 10000..10FFFF - // look at the construction of supplementary characters - // trail forms the ends of it. - return PROPERTY_.m_trie_.getSurrogateValue( - UTF16.getLeadSurrogate(ch), - (char)(ch & 0x3ff)); - } - // return m_dataOffset_ if there is an error, in this case we return - // the default value: m_initialValue_ - // we cannot assume that m_initialValue_ is at offset 0 - // this is for optimization. - return PROPERTY_INITIAL_VALUE_; - } + /** + * Character type Mn + * @stable ICU 2.1 + */ + public static final byte NON_SPACING_MARK = 6; + /** + * Character type Me + * @stable ICU 2.1 + */ + public static final byte ENCLOSING_MARK = 7; + /** + * Character type Mc + * @stable ICU 2.1 + */ + public static final byte COMBINING_SPACING_MARK = 8; + /** + * Character type count + * @stable ICU 2.1 + */ + public static final byte CHAR_CATEGORY_COUNT = 30; + /** + * Directional type R + * @stable ICU 2.1 + */ + public static final int RIGHT_TO_LEFT = 1; + /** + * Directional type AL + * @stable ICU 2.1 + */ + public static final int RIGHT_TO_LEFT_ARABIC = 13; }