This is the mail archive of the
java-patches@gcc.gnu.org
mailing list for the Java project.
Patch: FYI: Mark's String merge
- From: Tom Tromey <tromey at redhat dot com>
- To: GCJ-patches <java-patches at gcc dot gnu dot org>
- Date: 05 Mar 2007 10:11:04 -0700
- Subject: Patch: FYI: Mark's String merge
- Reply-to: tromey at redhat dot com
I'm checking this in on the trunk and the RH 4.1 branch.
This is Mark's String merge patch. I tried it against various Mauve
tests with good results. It brings String and a few others up to 1.5.
Tom
Index: ChangeLog
from Mark Wielaard <mark@klomp.org>
* java/lang/Character.java: Re-merged with Classpath.
* java/lang/natString.cc (nativeCompareTo): Renamed from
compareTo.
* java/lang/StringBuilder.java: Re-merged with Classpath.
* java/lang/String.java: Re-merged with Classpath.
(nativeCompareTo): Renamed from compareTo.
* java/lang/StringBuffer.java: Re-merged with Classpath.
* jni.cc (_Jv_JNI_GetAnyMethodID): Split calls to append.
Index: jni.cc
===================================================================
--- jni.cc (revision 122552)
+++ jni.cc (working copy)
@@ -751,7 +751,8 @@
java::lang::StringBuffer *name_sig =
new java::lang::StringBuffer (JvNewStringUTF (name));
- name_sig->append ((jchar) ' ')->append (JvNewStringUTF (s));
+ name_sig->append ((jchar) ' ');
+ name_sig->append (JvNewStringUTF (s));
env->ex = new java::lang::NoSuchMethodError (name_sig->toString ());
}
catch (jthrowable t)
Index: java/lang/StringBuffer.java
===================================================================
--- java/lang/StringBuffer.java (revision 122552)
+++ java/lang/StringBuffer.java (working copy)
@@ -1,5 +1,5 @@
/* StringBuffer.java -- Growable strings
- Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006
+ Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007
Free Software Foundation, Inc.
This file is part of GNU Classpath.
@@ -72,8 +72,12 @@
* @since 1.0
* @status updated to 1.4
*/
-public final class StringBuffer implements Serializable, CharSequence
+public final class StringBuffer
+ implements Serializable, CharSequence, Appendable
{
+ // Implementation note: if you change this class, you usually will
+ // want to change StringBuilder as well.
+
/**
* Compatible with JDK 1.0+.
*/
@@ -152,17 +156,16 @@
* specified <code>CharSequence</code>. Initial capacity will be the
* size of the CharSequence plus 16.
*
- * @param sequence the <code>String</code> to convert
+ * @param seq the <code>String</code> to convert
* @throws NullPointerException if str is null
- *
* @since 1.5
*/
- public StringBuffer(CharSequence sequence)
+ public StringBuffer(CharSequence seq)
{
- count = Math.max(0, sequence.length());
+ count = Math.max(0, seq.length());
value = new char[count + DEFAULT_CAPACITY];
for (int i = 0; i < count; ++i)
- value[i] = sequence.charAt(i);
+ value[i] = seq.charAt(i);
}
/**
@@ -391,46 +394,6 @@
}
/**
- * Append the <code>CharSequence</code> value of the argument to this
- * <code>StringBuffer</code>.
- *
- * @param sequence the <code>CharSequence</code> to append
- * @return this <code>StringBuffer</code>
- * @see #append(Object)
- * @since 1.5
- */
- public synchronized StringBuffer append(CharSequence sequence)
- {
- if (sequence == null)
- sequence = "null";
- return append(sequence, 0, sequence.length());
- }
-
- /**
- * Append the specified subsequence of the <code>CharSequence</code>
- * argument to this <code>StringBuffer</code>.
- *
- * @param sequence the <code>CharSequence</code> to append
- * @param start the starting index
- * @param end one past the ending index
- * @return this <code>StringBuffer</code>
- * @see #append(Object)
- * @since 1.5
- */
- public synchronized StringBuffer append(CharSequence sequence,
- int start, int end)
- {
- if (sequence == null)
- sequence = "null";
- if (start < 0 || end < 0 || start > end || end > sequence.length())
- throw new IndexOutOfBoundsException();
- ensureCapacity_unsynchronized(this.count + end - start);
- for (int i = start; i < end; ++i)
- value[count++] = sequence.charAt(i);
- return this;
- }
-
- /**
* Append the <code>char</code> array to this <code>StringBuffer</code>.
* This is similar (but more efficient) than
* <code>append(new String(data))</code>, except in the case of null.
@@ -470,6 +433,25 @@
}
/**
+ * Append the code point to this <code>StringBuffer</code>.
+ * This is like #append(char), but will append two characters
+ * if a supplementary code point is given.
+ *
+ * @param code the code point to append
+ * @return this <code>StringBuffer</code>
+ * @see Character#toChars(int, char[], int)
+ * @since 1.5
+ */
+ public synchronized StringBuffer appendCodePoint(int code)
+ {
+ int len = Character.charCount(code);
+ ensureCapacity_unsynchronized(count + len);
+ Character.toChars(code, value, count);
+ count += len;
+ return this;
+ }
+
+ /**
* Append the <code>String</code> value of the argument to this
* <code>StringBuffer</code>. Uses <code>String.valueOf()</code> to convert
* to <code>String</code>.
@@ -497,21 +479,41 @@
}
/**
- * Append the code point to this <code>StringBuffer</code>.
- * This is like #append(char), but will append two characters
- * if a supplementary code point is given.
+ * Append the <code>CharSequence</code> value of the argument to this
+ * <code>StringBuffer</code>.
*
- * @param code the code point to append
+ * @param seq the <code>CharSequence</code> to append
* @return this <code>StringBuffer</code>
- * @see Character#toChars(int, char[], int)
+ * @see #append(Object)
* @since 1.5
*/
- public synchronized StringBuffer appendCodePoint(int code)
+ public synchronized StringBuffer append(CharSequence seq)
{
- int len = Character.charCount(code);
- ensureCapacity_unsynchronized(count + len);
- Character.toChars(code, value, count);
- count += len;
+ if (seq == null)
+ seq = "null";
+ return append(seq, 0, seq.length());
+ }
+
+ /**
+ * Append the specified subsequence of the <code>CharSequence</code>
+ * argument to this <code>StringBuffer</code>.
+ *
+ * @param seq the <code>CharSequence</code> to append
+ * @param start the starting index
+ * @param end one past the ending index
+ * @return this <code>StringBuffer</code>
+ * @see #append(Object)
+ * @since 1.5
+ */
+ public synchronized StringBuffer append(CharSequence seq, int start, int end)
+ {
+ if (seq == null)
+ seq = "null";
+ if (start < 0 || end < 0 || start > end || end > seq.length())
+ throw new IndexOutOfBoundsException();
+ ensureCapacity_unsynchronized(this.count + end - start);
+ for (int i = start; i < end; ++i)
+ value[count++] = seq.charAt(i);
return this;
}
Index: java/lang/String.java
===================================================================
--- java/lang/String.java (revision 122552)
+++ java/lang/String.java (working copy)
@@ -1,5 +1,5 @@
/* String.java -- immutable character sequences; the object of string literals
- Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006
+ Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007
Free Software Foundation, Inc.
This file is part of GNU Classpath.
@@ -41,9 +41,11 @@
import java.io.Serializable;
import java.io.UnsupportedEncodingException;
-import java.lang.Comparable;
import java.util.Comparator;
+import java.text.Collator;
+import java.util.Formatter;
import java.util.Locale;
+import java.util.regex.Matcher;
import java.util.regex.Pattern;
import java.util.regex.PatternSyntaxException;
@@ -70,10 +72,13 @@
* @author Paul N. Fisher
* @author Eric Blake (ebb9@email.byu.edu)
* @author Per Bothner (bothner@cygnus.com)
+ * @author Tom Tromey (tromey@redhat.com)
+ * @author Andrew John Hughes (gnu_andrew@member.fsf.org)
* @since 1.0
* @status updated to 1.4
*/
-public final class String implements Serializable, Comparable, CharSequence
+public final class String
+ implements Serializable, Comparable<String>, CharSequence
{
// WARNING: String is a CORE class in the bootstrap cycle. See the comments
// in vm/reference/java/lang/Runtime for implications of this fact.
@@ -113,12 +118,12 @@
private int cachedHashCode;
/**
- * An implementation for {@link CASE_INSENSITIVE_ORDER}.
+ * An implementation for {@link #CASE_INSENSITIVE_ORDER}.
* This must be {@link Serializable}. The class name is dictated by
* compatibility with Sun's JDK.
*/
private static final class CaseInsensitiveComparator
- implements Comparator, Serializable
+ implements Comparator<String>, Serializable
{
/**
* Compatible with JDK 1.2.
@@ -142,9 +147,9 @@
* @throws ClassCastException if either argument is not a String
* @see #compareToIgnoreCase(String)
*/
- public int compare(Object o1, Object o2)
+ public int compare(String o1, String o2)
{
- return ((String) o1).compareToIgnoreCase((String) o2);
+ return o1.compareToIgnoreCase(o2);
}
} // class CaseInsensitiveComparator
@@ -156,7 +161,7 @@
* @see Collator#compare(String, String)
* @since 1.2
*/
- public static final Comparator CASE_INSENSITIVE_ORDER
+ public static final Comparator<String> CASE_INSENSITIVE_ORDER
= new CaseInsensitiveComparator();
/**
@@ -207,6 +212,7 @@
* @param count the number of characters from data to copy
* @throws NullPointerException if data is null
* @throws IndexOutOfBoundsException if (offset < 0 || count < 0
+ * || offset + count < 0 (overflow)
* || offset + count > data.length)
* (while unspecified, this is a StringIndexOutOfBoundsException)
*/
@@ -230,6 +236,7 @@
* @param count the number of characters from ascii to copy
* @throws NullPointerException if ascii is null
* @throws IndexOutOfBoundsException if (offset < 0 || count < 0
+ * || offset + count < 0 (overflow)
* || offset + count > ascii.length)
* (while unspecified, this is a StringIndexOutOfBoundsException)
* @see #String(byte[])
@@ -281,7 +288,7 @@
*
* @param data byte array to copy
* @param offset the offset to start at
- * @param count the number of characters in the array to use
+ * @param count the number of bytes in the array to use
* @param encoding the name of the encoding to use
* @throws NullPointerException if data or encoding is null
* @throws IndexOutOfBoundsException if offset or count is incorrect
@@ -330,7 +337,7 @@
*
* @param data byte array to copy
* @param offset the offset to start at
- * @param count the number of characters in the array to use
+ * @param count the number of bytes in the array to use
* @throws NullPointerException if data is null
* @throws IndexOutOfBoundsException if offset or count is incorrect
* @throws Error if the decoding fails
@@ -644,23 +651,17 @@
* @return the comparison
* @throws NullPointerException if anotherString is null
*/
- public native int compareTo(String anotherString);
+ public int compareTo(String anotherString)
+ {
+ return nativeCompareTo(anotherString);
+ }
/**
- * Behaves like <code>compareTo(java.lang.String)</code> unless the Object
- * is not a <code>String</code>. Then it throws a
- * <code>ClassCastException</code>.
- *
- * @param o the object to compare against
- * @return the comparison
- * @throws NullPointerException if o is null
- * @throws ClassCastException if o is not a <code>String</code>
- * @since 1.2
+ * The native implementation of compareTo(). Must be named different
+ * since cni doesn't understand the bridge method generated from
+ * the compareTo() method because of the Comparable<String> interface.
*/
- public int compareTo(Object o)
- {
- return compareTo((String) o);
- }
+ private native int nativeCompareTo(String anotherString);
/**
* Compares this String and another String (case insensitive). This
@@ -710,7 +711,7 @@
* @param ignoreCase true if case should be ignored in comparision
* @param toffset index to start comparison at for this String
* @param other String to compare region to this String
- * @param oofset index to start comparison at for other
+ * @param ooffset index to start comparison at for other
* @param len number of characters to compare
* @return true if regions match, false otherwise
* @throws NullPointerException if other is null
@@ -1124,7 +1125,7 @@
* Trims all characters less than or equal to <code>'\u0020'</code>
* (<code>' '</code>) from the beginning and end of this String. This
* includes many, but not all, ASCII control characters, and all
- * {@link Character#whitespace(char)}.
+ * {@link Character#isWhitespace(char)}.
*
* @return new trimmed String, or this if nothing trimmed
*/
@@ -1204,6 +1205,7 @@
* @return String containing the chars from data[offset..offset+count]
* @throws NullPointerException if data is null
* @throws IndexOutOfBoundsException if (offset < 0 || count < 0
+ * || offset + count < 0 (overflow)
* || offset + count > data.length)
* (while unspecified, this is a StringIndexOutOfBoundsException)
* @see #String(char[], int, int)
@@ -1294,12 +1296,27 @@
return Double.toString(d);
}
+
+ /** @since 1.5 */
+ public static String format(Locale locale, String format, Object... args)
+ {
+ Formatter f = new Formatter(locale);
+ return f.format(format, args).toString();
+ }
+
+ /** @since 1.5 */
+ public static String format(String format, Object... args)
+ {
+ return format(Locale.getDefault(), format, args);
+ }
+
/**
- * Fetches this String from the intern hashtable. If two Strings are
- * considered equal, by the equals() method, then intern() will return the
- * same String instance. ie. if (s1.equals(s2)) then
- * (s1.intern() == s2.intern()). All string literals and string-valued
- * constant expressions are already interned.
+ * Fetches this String from the intern hashtable.
+ * If two Strings are considered equal, by the equals() method,
+ * then intern() will return the same String instance. ie.
+ * if (s1.equals(s2)) then (s1.intern() == s2.intern()).
+ * All string literals and string-valued constant expressions
+ * are already interned.
*
* @return the interned String
*/
@@ -1345,20 +1362,20 @@
}
return count;
}
-
+
/**
* Returns true iff this String contains the sequence of Characters
* described in s.
* @param s the CharSequence
* @return true iff this String contains s
- *
+ *
* @since 1.5
*/
public boolean contains (CharSequence s)
{
return this.indexOf(s.toString()) != -1;
}
-
+
/**
* Returns a string that is this string with all instances of the sequence
* represented by <code>target</code> replaced by the sequence in
@@ -1386,7 +1403,42 @@
}
return result.toString();
}
+
+ /**
+ * Return the index into this String that is offset from the given index by
+ * <code>codePointOffset</code> code points.
+ * @param index the index at which to start
+ * @param codePointOffset the number of code points to offset
+ * @return the index into this String that is <code>codePointOffset</code>
+ * code points offset from <code>index</code>.
+ *
+ * @throws IndexOutOfBoundsException if index is negative or larger than the
+ * length of this string.
+ * @throws IndexOutOfBoundsException if codePointOffset is positive and the
+ * substring starting with index has fewer than codePointOffset code points.
+ * @throws IndexOutOfBoundsException if codePointOffset is negative and the
+ * substring ending with index has fewer than (-codePointOffset) code points.
+ * @since 1.5
+ */
+ public int offsetByCodePoints(int index, int codePointOffset)
+ {
+ if (index < 0 || index > count)
+ throw new IndexOutOfBoundsException();
+
+ return Character.offsetByCodePoints(this, index, codePointOffset);
+ }
+ /**
+ * Returns true if, and only if, {@link #length()}
+ * is <code>0</code>.
+ *
+ * @return true if the length of the string is zero.
+ * @since 1.6
+ */
+ public boolean isEmpty()
+ {
+ return count == 0;
+ }
private native void init(char[] chars, int offset, int count,
boolean dont_copy);
Index: java/lang/Character.java
===================================================================
--- java/lang/Character.java (revision 122552)
+++ java/lang/Character.java (working copy)
@@ -1,5 +1,6 @@
/* java.lang.Character -- Wrapper class for char, and Unicode subsets
- Copyright (C) 1998, 1999, 2001, 2002, 2005, 2006 Free Software Foundation, Inc.
+ Copyright (C) 1998, 1999, 2001, 2002, 2005, 2006, 2007
+ Free Software Foundation, Inc.
This file is part of GNU Classpath.
@@ -54,7 +55,7 @@
/**
* Wrapper class for the primitive char data type. In addition, this class
* allows one to retrieve property information and perform transformations
- * on the 57,707 defined characters in the Unicode Standard, Version 3.0.0.
+ * on the defined characters in the Unicode Standard, Version 4.0.0.
* java.lang.Character is designed to be very dynamic, and as such, it
* retrieves information on the Unicode character set from a separate
* database, gnu.java.lang.CharData, which can be easily upgraded.
@@ -62,7 +63,7 @@
* <p>For predicates, boundaries are used to describe
* the set of characters for which the method will return true.
* This syntax uses fairly normal regular expression notation.
- * See 5.13 of the Unicode Standard, Version 3.0, for the
+ * See 5.13 of the Unicode Standard, Version 4.0, for the
* boundary specification.
*
* <p>See <a href="http://www.unicode.org";>http://www.unicode.org</a>
@@ -72,10 +73,11 @@
* @author Paul N. Fisher
* @author Jochen Hoenicke
* @author Eric Blake (ebb9@email.byu.edu)
+ * @author Andrew John Hughes (gnu_andrew@member.fsf.org)
* @since 1.0
- * @status updated to 1.4
+ * @status partly updated to 1.5; some things still missing
*/
-public final class Character implements Serializable, Comparable
+public final class Character implements Serializable, Comparable<Character>
{
/**
* A subset of Unicode blocks.
@@ -160,10 +162,8 @@
/** The canonical name of the block according to the Unicode standard. */
private final String canonicalName;
- /** Constants for the <code>forName()</code> method */
- private static final int CANONICAL_NAME = 0;
- private static final int NO_SPACES_NAME = 1;
- private static final int CONSTANT_NAME = 2;
+ /** Enumeration for the <code>forName()</code> method */
+ private enum NameType { CANONICAL, NO_SPACES, CONSTANT; };
/**
* Constructor for strictly defined blocks.
@@ -173,7 +173,7 @@
* @param name the block name
*/
private UnicodeBlock(int start, int end, String name,
- String canonicalName)
+ String canonicalName)
{
super(name);
this.start = start;
@@ -207,8 +207,8 @@
public static UnicodeBlock of(int codePoint)
{
if (codePoint > MAX_CODE_POINT)
- throw new IllegalArgumentException("The supplied integer value is " +
- "too large to be a codepoint.");
+ throw new IllegalArgumentException("The supplied integer value is " +
+ "too large to be a codepoint.");
// Simple binary search for the correct block.
int low = 0;
int hi = sets.length - 1;
@@ -262,60 +262,52 @@
*/
public static final UnicodeBlock forName(String blockName)
{
- int type;
+ NameType type;
if (blockName.indexOf(' ') != -1)
- type = CANONICAL_NAME;
+ type = NameType.CANONICAL;
else if (blockName.indexOf('_') != -1)
- type = CONSTANT_NAME;
+ type = NameType.CONSTANT;
else
- type = NO_SPACES_NAME;
+ type = NameType.NO_SPACES;
Collator usCollator = Collator.getInstance(Locale.US);
usCollator.setStrength(Collator.PRIMARY);
/* Special case for deprecated blocks not in sets */
switch (type)
{
- case CANONICAL_NAME:
+ case CANONICAL:
if (usCollator.compare(blockName, "Surrogates Area") == 0)
return SURROGATES_AREA;
break;
- case NO_SPACES_NAME:
+ case NO_SPACES:
if (usCollator.compare(blockName, "SurrogatesArea") == 0)
return SURROGATES_AREA;
break;
- case CONSTANT_NAME:
+ case CONSTANT:
if (usCollator.compare(blockName, "SURROGATES_AREA") == 0)
return SURROGATES_AREA;
break;
}
/* Other cases */
- int setLength = sets.length;
switch (type)
{
- case CANONICAL_NAME:
- for (int i = 0; i < setLength; i++)
- {
- UnicodeBlock block = sets[i];
- if (usCollator.compare(blockName, block.canonicalName) == 0)
- return block;
- }
+ case CANONICAL:
+ for (UnicodeBlock block : sets)
+ if (usCollator.compare(blockName, block.canonicalName) == 0)
+ return block;
break;
- case NO_SPACES_NAME:
- for (int i = 0; i < setLength; i++)
- {
- UnicodeBlock block = sets[i];
- String nsName = block.canonicalName.replaceAll(" ","");
- if (usCollator.compare(blockName, nsName) == 0)
- return block;
- }
+ case NO_SPACES:
+ for (UnicodeBlock block : sets)
+ {
+ String nsName = block.canonicalName.replaceAll(" ","");
+ if (usCollator.compare(blockName, nsName) == 0)
+ return block;
+ }
+ break;
+ case CONSTANT:
+ for (UnicodeBlock block : sets)
+ if (usCollator.compare(blockName, block.toString()) == 0)
+ return block;
break;
- case CONSTANT_NAME:
- for (int i = 0; i < setLength; i++)
- {
- UnicodeBlock block = sets[i];
- if (usCollator.compare(blockName, block.toString()) == 0)
- return block;
- }
- break;
}
throw new IllegalArgumentException("No Unicode block found for " +
blockName + ".");
@@ -1517,10 +1509,11 @@
* this. These are also returned from calls to <code>of(int)</code>
* and <code>of(char)</code>.
*/
+ @Deprecated
public static final UnicodeBlock SURROGATES_AREA
= new UnicodeBlock(0xD800, 0xDFFF,
"SURROGATES_AREA",
- "Surrogates Area");
+ "Surrogates Area");
/**
* The defined subsets.
@@ -1699,11 +1692,78 @@
public static final char MAX_VALUE = '\uFFFF';
/**
+ * The minimum Unicode 4.0 code point. This value is <code>0</code>.
+ * @since 1.5
+ */
+ public static final int MIN_CODE_POINT = 0;
+
+ /**
+ * The maximum Unicode 4.0 code point, which is greater than the range
+ * of the char data type.
+ * This value is <code>0x10FFFF</code>.
+ * @since 1.5
+ */
+ public static final int MAX_CODE_POINT = 0x10FFFF;
+
+ /**
+ * The minimum Unicode high surrogate code unit, or
+ * <emph>leading-surrogate</emph>, in the UTF-16 character encoding.
+ * This value is <code>'\uD800'</code>.
+ * @since 1.5
+ */
+ public static final char MIN_HIGH_SURROGATE = '\uD800';
+
+ /**
+ * The maximum Unicode high surrogate code unit, or
+ * <emph>leading-surrogate</emph>, in the UTF-16 character encoding.
+ * This value is <code>'\uDBFF'</code>.
+ * @since 1.5
+ */
+ public static final char MAX_HIGH_SURROGATE = '\uDBFF';
+
+ /**
+ * The minimum Unicode low surrogate code unit, or
+ * <emph>trailing-surrogate</emph>, in the UTF-16 character encoding.
+ * This value is <code>'\uDC00'</code>.
+ * @since 1.5
+ */
+ public static final char MIN_LOW_SURROGATE = '\uDC00';
+
+ /**
+ * The maximum Unicode low surrogate code unit, or
+ * <emph>trailing-surrogate</emph>, in the UTF-16 character encoding.
+ * This value is <code>'\uDFFF'</code>.
+ * @since 1.5
+ */
+ public static final char MAX_LOW_SURROGATE = '\uDFFF';
+
+ /**
+ * The minimum Unicode surrogate code unit in the UTF-16 character encoding.
+ * This value is <code>'\uD800'</code>.
+ * @since 1.5
+ */
+ public static final char MIN_SURROGATE = MIN_HIGH_SURROGATE;
+
+ /**
+ * The maximum Unicode surrogate code unit in the UTF-16 character encoding.
+ * This value is <code>'\uDFFF'</code>.
+ * @since 1.5
+ */
+ public static final char MAX_SURROGATE = MAX_LOW_SURROGATE;
+
+ /**
+ * The lowest possible supplementary Unicode code point (the first code
+ * point outside the basic multilingual plane (BMP)).
+ * This value is <code>0x10000</code>.
+ */
+ public static final int MIN_SUPPLEMENTARY_CODE_POINT = 0x10000;
+
+ /**
* Class object representing the primitive char data type.
*
* @since 1.1
*/
- public static final Class TYPE = VMClassLoader.getPrimitiveClass('C');
+ public static final Class<Character> TYPE = (Class<Character>) VMClassLoader.getPrimitiveClass('C');
/**
* The number of bits needed to represent a <code>char</code>.
@@ -2089,71 +2149,6 @@
private static final int MIRROR_MASK = 0x40;
/**
- * Min value for supplementary code point.
- *
- * @since 1.5
- */
- public static final int MIN_SUPPLEMENTARY_CODE_POINT = 0x10000;
-
- /**
- * Min value for code point.
- *
- * @since 1.5
- */
- public static final int MIN_CODE_POINT = 0;
-
-
- /**
- * Max value for code point.
- *
- * @since 1.5
- */
- public static final int MAX_CODE_POINT = 0x010ffff;
-
-
- /**
- * Minimum high surrogate code in UTF-16 encoding.
- *
- * @since 1.5
- */
- public static final char MIN_HIGH_SURROGATE = '\ud800';
-
- /**
- * Maximum high surrogate code in UTF-16 encoding.
- *
- * @since 1.5
- */
- public static final char MAX_HIGH_SURROGATE = '\udbff';
-
- /**
- * Minimum low surrogate code in UTF-16 encoding.
- *
- * @since 1.5
- */
- public static final char MIN_LOW_SURROGATE = '\udc00';
-
- /**
- * Maximum low surrogate code in UTF-16 encoding.
- *
- * @since 1.5
- */
- public static final char MAX_LOW_SURROGATE = '\udfff';
-
- /**
- * Minimum surrogate code in UTF-16 encoding.
- *
- * @since 1.5
- */
- public static final char MIN_SURROGATE = MIN_HIGH_SURROGATE;
-
- /**
- * Maximum low surrogate code in UTF-16 encoding.
- *
- * @since 1.5
- */
- public static final char MAX_SURROGATE = MAX_LOW_SURROGATE;
-
- /**
* Grabs an attribute offset from the Unicode attribute database. The lower
* 5 bits are the character type, the next 2 bits are flags, and the top
* 9 bits are the offset into the attribute tables. Note that the top 9
@@ -2504,8 +2499,211 @@
| (1 << MODIFIER_LETTER)
| (1 << OTHER_LETTER))) != 0;
}
+
+ /**
+ * Returns the index into the given CharSequence that is offset
+ * <code>codePointOffset</code> code points from <code>index</code>.
+ * @param seq the CharSequence
+ * @param index the start position in the CharSequence
+ * @param codePointOffset the number of code points offset from the start
+ * position
+ * @return the index into the CharSequence that is codePointOffset code
+ * points offset from index
+ *
+ * @throws NullPointerException if seq is null
+ * @throws IndexOutOfBoundsException if index is negative or greater than the
+ * length of the sequence.
+ * @throws IndexOutOfBoundsException if codePointOffset is positive and the
+ * subsequence from index to the end of seq has fewer than codePointOffset
+ * code points
+ * @throws IndexOutOfBoundsException if codePointOffset is negative and the
+ * subsequence from the start of seq to index has fewer than
+ * (-codePointOffset) code points
+ * @since 1.5
+ */
+ public static int offsetByCodePoints(CharSequence seq,
+ int index,
+ int codePointOffset)
+ {
+ int len = seq.length();
+ if (index < 0 || index > len)
+ throw new IndexOutOfBoundsException();
+
+ int numToGo = codePointOffset;
+ int offset = index;
+ int adjust = 1;
+ if (numToGo >= 0)
+ {
+ for (; numToGo > 0; offset++)
+ {
+ numToGo--;
+ if (Character.isHighSurrogate(seq.charAt(offset))
+ && (offset + 1) < len
+ && Character.isLowSurrogate(seq.charAt(offset + 1)))
+ offset++;
+ }
+ return offset;
+ }
+ else
+ {
+ numToGo *= -1;
+ for (; numToGo > 0;)
+ {
+ numToGo--;
+ offset--;
+ if (Character.isLowSurrogate(seq.charAt(offset))
+ && (offset - 1) >= 0
+ && Character.isHighSurrogate(seq.charAt(offset - 1)))
+ offset--;
+ }
+ return offset;
+ }
+ }
+
+ /**
+ * Returns the index into the given char subarray that is offset
+ * <code>codePointOffset</code> code points from <code>index</code>.
+ * @param a the char array
+ * @param start the start index of the subarray
+ * @param count the length of the subarray
+ * @param index the index to be offset
+ * @param codePointOffset the number of code points offset from <code>index
+ * </code>
+ * @return the index into the char array
+ *
+ * @throws NullPointerException if a is null
+ * @throws IndexOutOfBoundsException if start or count is negative or if
+ * start + count is greater than the length of the array
+ * @throws IndexOutOfBoundsException if index is less than start or larger
+ * than start + count
+ * @throws IndexOutOfBoundsException if codePointOffset is positive and the
+ * subarray from index to start + count - 1 has fewer than codePointOffset
+ * code points.
+ * @throws IndexOutOfBoundsException if codePointOffset is negative and the
+ * subarray from start to index - 1 has fewer than (-codePointOffset) code
+ * points
+ * @since 1.5
+ */
+ public static int offsetByCodePoints(char[] a,
+ int start,
+ int count,
+ int index,
+ int codePointOffset)
+ {
+ int len = a.length;
+ int end = start + count;
+ if (start < 0 || count < 0 || end > len || index < start || index > end)
+ throw new IndexOutOfBoundsException();
+
+ int numToGo = codePointOffset;
+ int offset = index;
+ int adjust = 1;
+ if (numToGo >= 0)
+ {
+ for (; numToGo > 0; offset++)
+ {
+ numToGo--;
+ if (Character.isHighSurrogate(a[offset])
+ && (offset + 1) < len
+ && Character.isLowSurrogate(a[offset + 1]))
+ offset++;
+ }
+ return offset;
+ }
+ else
+ {
+ numToGo *= -1;
+ for (; numToGo > 0;)
+ {
+ numToGo--;
+ offset--;
+ if (Character.isLowSurrogate(a[offset])
+ && (offset - 1) >= 0
+ && Character.isHighSurrogate(a[offset - 1]))
+ offset--;
+ if (offset < start)
+ throw new IndexOutOfBoundsException();
+ }
+ return offset;
+ }
+
+ }
+
/**
+ * Returns the number of Unicode code points in the specified range of the
+ * given CharSequence. The first char in the range is at position
+ * beginIndex and the last one is at position endIndex - 1. Paired
+ * surrogates (supplementary characters are represented by a pair of chars -
+ * one from the high surrogates and one from the low surrogates)
+ * count as just one code point.
+ * @param seq the CharSequence to inspect
+ * @param beginIndex the beginning of the range
+ * @param endIndex the end of the range
+ * @return the number of Unicode code points in the given range of the
+ * sequence
+ * @throws NullPointerException if seq is null
+ * @throws IndexOutOfBoundsException if beginIndex is negative, endIndex is
+ * larger than the length of seq, or if beginIndex is greater than endIndex.
+ * @since 1.5
+ */
+ public static int codePointCount(CharSequence seq, int beginIndex,
+ int endIndex)
+ {
+ int len = seq.length();
+ if (beginIndex < 0 || endIndex > len || beginIndex > endIndex)
+ throw new IndexOutOfBoundsException();
+
+ int count = 0;
+ for (int i = beginIndex; i < endIndex; i++)
+ {
+ count++;
+ // If there is a pairing, count it only once.
+ if (isHighSurrogate(seq.charAt(i)) && (i + 1) < endIndex
+ && isLowSurrogate(seq.charAt(i + 1)))
+ i ++;
+ }
+ return count;
+ }
+
+ /**
+ * Returns the number of Unicode code points in the specified range of the
+ * given char array. The first char in the range is at position
+ * offset and the length of the range is count. Paired surrogates
+ * (supplementary characters are represented by a pair of chars -
+ * one from the high surrogates and one from the low surrogates)
+ * count as just one code point.
+ * @param a the char array to inspect
+ * @param offset the beginning of the range
+ * @param count the length of the range
+ * @return the number of Unicode code points in the given range of the
+ * array
+ * @throws NullPointerException if a is null
+ * @throws IndexOutOfBoundsException if offset or count is negative or if
+ * offset + countendIndex is larger than the length of a.
+ * @since 1.5
+ */
+ public static int codePointCount(char[] a, int offset,
+ int count)
+ {
+ int len = a.length;
+ int end = offset + count;
+ if (offset < 0 || count < 0 || end > len)
+ throw new IndexOutOfBoundsException();
+
+ int counter = 0;
+ for (int i = offset; i < end; i++)
+ {
+ counter++;
+ // If there is a pairing, count it only once.
+ if (isHighSurrogate(a[i]) && (i + 1) < end
+ && isLowSurrogate(a[i + 1]))
+ i ++;
+ }
+ return counter;
+ }
+
+ /**
* Determines if a character is a Unicode letter or a Unicode digit. This
* is the combination of isLetter and isDigit.
* <br>
@@ -3497,30 +3695,13 @@
}
/**
- * Compares an object to this Character. Assuming the object is a
- * Character object, this method performs the same comparison as
- * compareTo(Character).
- *
- * @param o object to compare
- * @return the comparison value
- * @throws ClassCastException if o is not a Character object
- * @throws NullPointerException if o is null
- * @see #compareTo(Character)
- * @since 1.2
- */
- public int compareTo(Object o)
- {
- return compareTo((Character) o);
- }
-
- /**
* Returns an <code>Character</code> object wrapping the value.
* In contrast to the <code>Character</code> constructor, this method
* will cache some values. It is used by boxing conversion.
*
* @param val the value to wrap
* @return the <code>Character</code>
- *
+ *
* @since 1.5
*/
public static Character valueOf(char val)
@@ -3529,9 +3710,9 @@
return new Character(val);
synchronized (charCache)
{
- if (charCache[val - MIN_VALUE] == null)
- charCache[val - MIN_VALUE] = new Character(val);
- return charCache[val - MIN_VALUE];
+ if (charCache[val - MIN_VALUE] == null)
+ charCache[val - MIN_VALUE] = new Character(val);
+ return charCache[val - MIN_VALUE];
}
}
@@ -3559,6 +3740,9 @@
*/
public static char[] toChars(int codePoint)
{
+ if (!isValidCodePoint(codePoint))
+ throw new IllegalArgumentException("Illegal Unicode code point : "
+ + codePoint);
char[] result = new char[charCount(codePoint)];
int ignore = toChars(codePoint, result, 0);
return result;
@@ -3776,7 +3960,7 @@
*/
public static int codePointAt(char[] chars, int index, int limit)
{
- if (index < 0 || index >= limit || limit < 0 || limit >= chars.length)
+ if (index < 0 || index >= limit || limit < 0 || limit > chars.length)
throw new IndexOutOfBoundsException();
char high = chars[index];
if (! isHighSurrogate(high) || ++index >= limit)
Index: java/lang/StringBuilder.java
===================================================================
--- java/lang/StringBuilder.java (revision 122552)
+++ java/lang/StringBuilder.java (working copy)
@@ -1,5 +1,5 @@
/* StringBuilder.java -- Unsynchronized growable strings
- Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006
+ Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007
Free Software Foundation, Inc.
This file is part of GNU Classpath.
@@ -998,4 +998,65 @@
*/
// GCJ LOCAL: Native to access String internals properly.
private native boolean regionMatches(int toffset, String other);
+
+ /**
+ * Get the code point at the specified index. This is like #charAt(int),
+ * but if the character is the start of a surrogate pair, and the
+ * following character completes the pair, then the corresponding
+ * supplementary code point is returned.
+ * @param index the index of the codepoint to get, starting at 0
+ * @return the codepoint at the specified index
+ * @throws IndexOutOfBoundsException if index is negative or >= length()
+ * @since 1.5
+ */
+ public int codePointAt(int index)
+ {
+ return Character.codePointAt(value, index, count);
+ }
+
+ /**
+ * Get the code point before the specified index. This is like
+ * #codePointAt(int), but checks the characters at <code>index-1</code> and
+ * <code>index-2</code> to see if they form a supplementary code point.
+ * @param index the index just past the codepoint to get, starting at 0
+ * @return the codepoint at the specified index
+ * @throws IndexOutOfBoundsException if index is negative or >= length()
+ * @since 1.5
+ */
+ public int codePointBefore(int index)
+ {
+ // Character.codePointBefore() doesn't perform this check. We
+ // could use the CharSequence overload, but this is just as easy.
+ if (index >= count)
+ throw new IndexOutOfBoundsException();
+ return Character.codePointBefore(value, index, 1);
+ }
+
+ /**
+ * Returns the number of Unicode code points in the specified sub sequence.
+ * Surrogate pairs count as one code point.
+ * @param beginIndex the start of the subarray
+ * @param endIndex the index after the last char in the subarray
+ * @return the number of code points
+ * @throws IndexOutOfBoundsException if beginIndex is less than zero or
+ * greater than endIndex or if endIndex is greater than the length of this
+ * StringBuilder
+ */
+ public int codePointCount(int beginIndex,int endIndex)
+ {
+ if (beginIndex < 0 || beginIndex > endIndex || endIndex > count)
+ throw new IndexOutOfBoundsException("invalid indices: " + beginIndex
+ + ", " + endIndex);
+ return Character.codePointCount(value, beginIndex, endIndex - beginIndex);
+ }
+
+ public void trimToSize()
+ {
+ if (count < value.length)
+ {
+ char[] newValue = new char[count];
+ System.arraycopy(value, 0, newValue, 0, count);
+ value = newValue;
+ }
+ }
}
Index: java/lang/natString.cc
===================================================================
--- java/lang/natString.cc (revision 122552)
+++ java/lang/natString.cc (working copy)
@@ -707,7 +707,7 @@
}
jint
-java::lang::String::compareTo (jstring anotherString)
+java::lang::String::nativeCompareTo (jstring anotherString)
{
jchar *tptr = JvGetStringChars (this);
jchar *optr = JvGetStringChars (anotherString);