This is the mail archive of the
java-patches@gcc.gnu.org
mailing list for the Java project.
[mark@klomp.org: Partial Double/Float merge with Classpath]
- To: java-patches at gcc dot gnu dot org
- Subject: [mark@klomp.org: Partial Double/Float merge with Classpath]
- From: Mark Wielaard <mark at klomp dot org>
- Date: Fri, 12 Oct 2001 18:12:33 +0200
Hi,
Sorry, send this to the wrong list.
Cheers,
Mark
----- Forwarded message from Mark Wielaard <mark@klomp.org> -----
Date: Fri, 12 Oct 2001 17:24:58 +0200
From: Mark Wielaard <mark@klomp.org>
To: gcc-patches@gcc.gnu.org
Subject: Partial Double/Float merge with Classpath
Hi,
Most of the actual merge work has been done by Brian Jones a couple of months
ago. This patches just adds the javadoc comments from Classpath. The actual
code changes are small. The removal of byteValue/shortValue seem to fix
two failures in mauve (although I am not completely sure why). The addition
of the compare() method introduces a small bugfix for comparing two NaN
values (this used to return 1, but now returns 0 which is the correct value).
Interested readers are encouraged to read the Classpath mailing list were
the remaining differences between the libgcj and Classpath versions are
discussed.
2001-10-12 Mark Wielaard <mark@klomp.org>
* java/lang/Double.java: Partial merge with Classpath
(TYPE): initialized through VMClassLoader.getPrimitiveClass()
(value): made final
(static): new static block to load native libary (not used in libgcj)
(Double (String)): call parseDouble()
(byteValue): removed, already defined in superclass Number
(shortValue): likewise
(valueOf (String)): call new Double(String) directly
(compare (double,double)): new 1.4 method
(compareTo (Double)): call new compare(double,double) method
(initIDs): new private method (not used in libgcj)
* java/lang/Float.java: Partial merge with Classpath
(TYPE): initialized through VMClassLoader.getPrimitiveClass()
(value): made final
(static): new static block to load native libary (not used in libgcj)
(Float (String)): call parseFloat()
(byteValue): removed, already defined in superclass Number
(shortValue): likewise
(valueOf (String)): call new Float(String) directly
(compare (float,float)): new 1.4 method
(compareTo (Float)): call new compare(double,double) method
OK to commit?
Cheers,
Mark
--
Stuff to read:
<http://www.toad.com/gnu/whatswrong.html>
What's Wrong with Copy Protection, by John Gilmore
Index: java/lang/Double.java
===================================================================
RCS file: /cvs/gcc/gcc/libjava/java/lang/Double.java,v
retrieving revision 1.9
diff -u -r1.9 Double.java
--- Double.java 2001/02/09 22:13:33 1.9
+++ Double.java 2001/10/12 15:03:11
@@ -1,57 +1,155 @@
-/* Copyright (C) 1998, 1999, 2000, 2001 Free Software Foundation
+/* Double.java -- object wrapper for double primitive
+ Copyright (C) 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
- This file is part of libgcj.
+This file is part of GNU Classpath.
-This software is copyrighted work licensed under the terms of the
-Libgcj License. Please consult the file "LIBGCJ_LICENSE" for
-details. */
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING. If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+As a special exception, if you link this library with other files to
+produce an executable, this library does not by itself cause the
+resulting executable to be covered by the GNU General Public License.
+This exception does not however invalidate any other reasons why the
+executable file might be covered by the GNU General Public License. */
+
+
package java.lang;
-/**
- * @author Andrew Haley <aph@cygnus.com>
- * @date September 25, 1998.
- */
+import gnu.classpath.Configuration;
+
/* Written using "Java Class Libraries", 2nd edition, ISBN 0-201-31002-3
* "The Java Language Specification", ISBN 0-201-63451-1
* plus online API docs for JDK 1.2 beta from http://www.javasoft.com.
* Status: Believed complete and correct.
*/
+/**
+ * Instances of class <code>Double</code> represent primitive
+ * <code>double</code> values.
+ *
+ * Additionally, this class provides various helper functions and variables
+ * related to doubles.
+ *
+ * @author Paul Fisher
+ * @author Andrew Haley <aph@cygnus.com>
+ * @since JDK 1.0
+ */
public final class Double extends Number implements Comparable
{
+ /**
+ * The minimum positive value a <code>double</code> may represent
+ * is 5e-324.
+ */
public static final double MIN_VALUE = 5e-324;
+
+ /**
+ * The maximum positive value a <code>double</code> may represent
+ * is 1.7976931348623157e+308.
+ */
public static final double MAX_VALUE = 1.7976931348623157e+308;
+
+ /**
+ * The value of a double representation -1.0/0.0, negative
+ * infinity.
+ */
public static final double NEGATIVE_INFINITY = -1.0d/0.0d;
+
+ /**
+ * The value of a double representing 1.0/0.0, positive infinity.
+ */
public static final double POSITIVE_INFINITY = 1.0d/0.0d;
- public static final double NaN = 0.0d/0.0d;
- // This initialization is seemingly circular, but it is accepted
- // by javac, and is handled specially by gcc.
- public static final Class TYPE = double.class;
+ /**
+ * All IEEE 754 values of NaN have the same value in Java.
+ */
+ public static final double NaN = 0.0d/0.0d;
- private double value;
+ /**
+ * The primitive type <code>double</code> is represented by this
+ * <code>Class</code> object.
+ */
+ public static final Class TYPE = VMClassLoader.getPrimitiveClass ("double");
+
+ /**
+ * The immutable value of this Double.
+ */
+ private final double value;
private static final long serialVersionUID = -9172774392245257468L;
-
- public native static double parseDouble (String s)
- throws NumberFormatException;
- public Double (double v)
- {
- value = v;
- }
-
+ /**
+ * Load native routines necessary for this class.
+ */
+ static
+ {
+ if (Configuration.INIT_LOAD_LIBRARY)
+ {
+ System.loadLibrary ("javalang");
+ initIDs ();
+ }
+ }
+
+ /**
+ * Create a <code>Double</code> from the primitive <code>double</code>
+ * specified.
+ *
+ * @param value the <code>double</code> argument
+ */
+ public Double (double value)
+ {
+ this.value = value;
+ }
+
+ /**
+ * Create a <code>Double</code> from the specified
+ * <code>String</code>.
+ *
+ * This method calls <code>Double.parseDouble()</code>.
+ *
+ * @exception NumberFormatException when the <code>String</code> cannot
+ * be parsed into a <code>Float</code>.
+ * @param s the <code>String</code> to convert
+ * @see #parseDouble(java.lang.String)
+ */
public Double (String s) throws NumberFormatException
{
- value = valueOf (s).doubleValue ();
+ value = parseDouble (s);
}
+ /**
+ * Convert the <code>double</code> value of this <code>Double</code>
+ * to a <code>String</code>. This method calls
+ * <code>Double.toString(double)</code> to do its dirty work.
+ *
+ * @return the <code>String</code> representation of this <code>Double</code>.
+ * @see #toString(double)
+ */
public String toString ()
{
return toString (value);
}
+ /**
+ * If the <code>Object</code> is not <code>null</code>, is an
+ * <code>instanceof</code> <code>Double</code>, and represents
+ * the same primitive <code>double</code> value return
+ * <code>true</code>. Otherwise <code>false</code> is returned.
+ *
+ * @param obj the object to compare to
+ * @return whether the objects are semantically equal.
+ */
public boolean equals (Object obj)
{
if (!(obj instanceof Double))
@@ -62,59 +160,92 @@
return doubleToLongBits (value) == doubleToLongBits (d.doubleValue ());
}
+ /**
+ * The hashcode is the value of the expression: <br>
+ * <br>
+ * <code>(int)(v^(v>>>32))</code><br>
+ * <br>
+ * where v is defined by: <br>
+ * <code>long v = Double.doubleToLongBits(this.longValue());</code><br>
+ */
public int hashCode ()
{
long v = doubleToLongBits (value);
return (int) (v ^ (v >>> 32));
}
+ /**
+ * Return the value of this <code>Double</code> when cast to an
+ * <code>int</code>.
+ */
public int intValue ()
{
return (int) value;
}
+ /**
+ * Return the value of this <code>Double</code> when cast to a
+ * <code>long</code>.
+ */
public long longValue ()
{
return (long) value;
}
+ /**
+ * Return the value of this <code>Double</code> when cast to a
+ * <code>float</code>.
+ */
public float floatValue ()
{
return (float) value;
}
+ /**
+ * Return the primitive <code>double</code> value represented by this
+ * <code>Double</code>.
+ */
public double doubleValue ()
{
return value;
}
-
- public byte byteValue ()
- {
- return (byte) value;
- }
- public short shortValue ()
- {
- return (short) value;
- }
-
- native static String toString (double v, boolean isFloat);
-
- public static String toString (double v)
- {
- return toString (v, false);
- }
-
+ /**
+ * Return the result of calling <code>new Double(java.lang.String)</code>.
+ *
+ * @param s the <code>String</code> to convert to a <code>Double</code>.
+ * @return a new <code>Double</code> representing the <code>String</code>'s
+ * numeric value.
+ *
+ * @exception NullPointerException thrown if <code>String</code> is
+ * <code>null</code>.
+ * @exception NumberFormatException thrown if <code>String</code> cannot
+ * be parsed as a <code>double</code>.
+ * @see #Double(java.lang.String)
+ * @see #parseDouble(java.lang.String)
+ */
public static Double valueOf (String s) throws NumberFormatException
{
- return new Double (parseDouble (s));
+ return new Double (s);
}
+ /**
+ * Return <code>true</code> if the value of this <code>Double</code>
+ * is the same as <code>NaN</code>, otherwise return <code>false</code>.
+ * @return whether this <code>Double</code> is <code>NaN</code>.
+ */
public boolean isNaN ()
{
return isNaN (value);
}
+ /**
+ * Return <code>true</code> if the <code>double</code> has the same
+ * value as <code>NaN</code>, otherwise return <code>false</code>.
+ *
+ * @param v the <code>double</code> to compare
+ * @return whether the argument is <code>NaN</code>.
+ */
public static boolean isNaN (double v)
{
long bits = doubleToLongBits (v);
@@ -124,11 +255,26 @@
return e == 0x7ff0000000000000L && f != 0L;
}
+ /**
+ * Return <code>true</code> if the value of this <code>Double</code>
+ * is the same as <code>NEGATIVE_INFINITY</code> or
+ * <code>POSITIVE_INFINITY</code>, otherwise return <code>false</code>.
+ *
+ * @return whether this <code>Double</code> is (-/+) infinity.
+ */
public boolean isInfinite ()
{
return isInfinite (value);
}
+ /**
+ * Return <code>true</code> if the <code>double</code> has a value
+ * equal to either <code>NEGATIVE_INFINITY</code> or
+ * <code>POSITIVE_INFINITY</code>, otherwise return <code>false</code>.
+ *
+ * @param v the <code>double</code> to compare
+ * @return whether the argument is (-/+) infinity.
+ */
public static boolean isInfinite (double v)
{
long bits = doubleToLongBits (v);
@@ -136,30 +282,231 @@
return f == 0x7ff0000000000000L;
}
-
- public static native long doubleToLongBits (double value);
- public static native long doubleToRawLongBits (double value);
- public static native double longBitsToDouble (long bits);
+ /**
+ * Returns 0 if the <code>double</code> value of the argument is
+ * equal to the value of this <code>Double</code>. Returns a number
+ * less than zero if the value of this <code>Double</code> is less
+ * than the <code>double</code> value of the argument, and returns a
+ * number greater than zero if the value of this <code>Double</code>
+ * is greater than the <code>double</code> value of the argument.
+ * <br>
+ * <code>Double.NaN</code> is greater than any number other than itself,
+ * even <code>Double.POSITIVE_INFINITY</code>.
+ * <br>
+ * <code>0.0d</code> is greater than <code>-0.0d</code>.
+ *
+ * @param d the Double to compare to.
+ * @return 0 if the <code>Double</code>s are the same, < 0 if this
+ * <code>Double</code> is less than the <code>Double</code> in
+ * in question, or > 0 if it is greater.
+ * @since 1.2
+ */
public int compareTo (Double d)
{
- double v = d.value;
- if (isNaN (value))
- return isNaN (v) ? 1 : 0;
- else if (isNaN (v))
+ return compare (value, d.value);
+ }
+
+ /**
+ * Returns 0 if the first argument is equal to the second argument.
+ * Returns a number less than zero if the first argument is less than the
+ * second argument, and returns a number greater than zero if the first
+ * argument is greater than the second argument.
+ * <br>
+ * <code>Double.NaN</code> is greater than any number other than itself,
+ * even <code>Double.POSITIVE_INFINITY</code>.
+ * <br>
+ * <code>0.0d</code> is greater than <code>-0.0d</code>.
+ *
+ * @param x the first double to compare.
+ * @param y the second double to compare.
+ * @return 0 if the arguments are the same, < 0 if the
+ * first argument is less than the second argument in
+ * in question, or > 0 if it is greater.
+ * @since 1.4
+ */
+ public static int compare (double x, double y)
+ {
+ if (isNaN (x))
+ return isNaN (y) ? 0 : 1;
+ if (isNaN (y))
return -1;
- else if (value == 0.0 && v == -0.0)
+ if ((x == 0.0d) && (y == -0.0d))
return 1;
- else if (value == -0.0 && v == 0.0)
+ if ((x == -0.0d) && (y == 0.0d))
return -1;
- else if (value == v)
+ if (x == y)
return 0;
- return value > v ? 1 : -1;
+ return (x > y) ? 1 : -1;
}
+ /**
+ * Compares the specified <code>Object</code> to this <code>Double</code>
+ * if and only if the <code>Object</code> is an instanceof
+ * <code>Double</code>.
+ *
+ * @param o the Object to compare to.
+ * @return 0 if the <code>Double</code>s are the same, < 0 if this
+ * <code>Double</code> is less than the <code>Double</code> in
+ * in question, or > 0 if it is greater.
+ * @throws ClassCastException if the argument is not a <code>Double</code>
+ */
public int compareTo (Object o)
{
return compareTo ((Double) o);
}
+
+ /**
+ * Convert the <code>double</code> to a <code>String</code>.
+ * <P>
+ *
+ * Floating-point string representation is fairly complex: here is a
+ * rundown of the possible values. "<CODE>[-]</CODE>" indicates that a
+ * negative sign will be printed if the value (or exponent) is negative.
+ * "<CODE><number></CODE>" means a string of digits (0-9).
+ * "<CODE><digit></CODE>" means a single digit (0-9).
+ * <P>
+ *
+ * <TABLE BORDER=1>
+ * <TR><TH>Value of Float</TH><TH>String Representation</TH></TR>
+ * <TR>
+ * <TD>[+-] 0</TD>
+ * <TD>[<CODE>-</CODE>]<CODE>0.0</CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>Between [+-] 10<SUP>-3</SUP> and 10<SUP>7</SUP></TD>
+ * <TD><CODE>[-]number.number</CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>Other numeric value</TD>
+ * <TD><CODE>[-]<digit>.<number>E[-]<number></CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>[+-] infinity</TD>
+ * <TD><CODE>[-]Infinity</CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>NaN</TD>
+ * <TD><CODE>NaN</CODE></TD>
+ * </TR>
+ * </TABLE>
+ *
+ * Yes, negative zero <EM>is</EM> a possible value. Note that there is
+ * <EM>always</EM> a <CODE>.</CODE> and at least one digit printed after
+ * it: even if the number is 3, it will be printed as <CODE>3.0</CODE>.
+ * After the ".", all digits will be printed except trailing zeros. No
+ * truncation or rounding is done by this function.
+ *
+ *
+ * @XXX specify where we are not in accord with the spec.
+ *
+ * @param d the <code>double</code> to convert
+ * @return the <code>String</code> representing the <code>double</code>.
+ */
+ public static String toString (double d)
+ {
+ return toString (d, false);
+ }
+
+ static native String toString (double d, boolean isFloat);
+
+ /**
+ * Return the long bits of the specified <code>double</code>.
+ * The result of this function can be used as the argument to
+ * <code>Double.longBitsToDouble(long)</code> to obtain the
+ * original <code>double</code> value.
+ *
+ * @param value the <code>double</code> to convert
+ * @return the bits of the <code>double</code>.
+ */
+ public static native long doubleToLongBits (double value);
+
+ /**
+ * Return the long bits of the specified <code>double</code>.
+ * The result of this function can be used as the argument to
+ * <code>Double.longBitsToDouble(long)</code> to obtain the
+ * original <code>double</code> value. This method differs from
+ * <code>doubleToLongBits</code> in that it does not collapse
+ * NaN values.
+ *
+ * @param value the <code>double</code> to convert
+ * @return the bits of the <code>double</code>.
+ */
+ public static native long doubleToRawLongBits (double value);
+
+ /**
+ * Return the <code>double</code> represented by the long
+ * bits specified.
+ *
+ * @param bits the long bits representing a <code>double</code>
+ * @return the <code>double</code> represented by the bits.
+ */
+ public static native double longBitsToDouble (long bits);
+
+ /**
+ * Parse the specified <code>String</code> as a <code>double</code>.
+ *
+ * The number is really read as <em>n * 10<sup>exponent</sup></em>. The
+ * first number is <em>n</em>, and if there is an "<code>E</code>"
+ * ("<code>e</code>" is also acceptable), then the integer after that is
+ * the exponent.
+ * <P>
+ * Here are the possible forms the number can take:
+ * <BR>
+ * <TABLE BORDER=1>
+ * <TR><TH>Form</TH><TH>Examples</TH></TR>
+ * <TR><TD><CODE>[+-]<number>[.]</CODE></TD><TD>345., -10, 12</TD></TR>
+ * <TR><TD><CODE>[+-]<number>.<number></CODE></TD><TD>40.2, 80.00, -12.30</TD></TR>
+ * <TR><TD><CODE>[+-]<number>[.]E[+-]<number></CODE></TD><TD>80E12, -12e+7, 4.E-123</TD></TR>
+ * <TR><TD><CODE>[+-]<number>.<number>E[+-]<number></CODE></TD><TD>6.02e-22, -40.2E+6, 12.3e9</TD></TR>
+ * </TABLE>
+ *
+ * "<code>[+-]</code>" means either a plus or minus sign may go there, or
+ * neither, in which case + is assumed.
+ * <BR>
+ * "<code>[.]</code>" means a dot may be placed here, but is optional.
+ * <BR>
+ * "<code><number></code>" means a string of digits (0-9), basically
+ * an integer. "<code><number>.<number></code>" is basically
+ * a real number, a floating-point value.
+ * <P>
+ *
+ * Remember that a <code>double</code> has a limited range. If the
+ * number you specify is greater than <code>Double.MAX_VALUE</code> or less
+ * than <code>-Double.MAX_VALUE</code>, it will be set at
+ * <code>Double.POSITIVE_INFINITY</code> or
+ * <code>Double.NEGATIVE_INFINITY</code>, respectively.
+ * <P>
+ * Note also that <code>double</code> does not have perfect precision. Many
+ * numbers cannot be precisely represented. The number you specify
+ * will be rounded to the nearest representable value.
+ * <code>Double.MIN_VALUE</code> is the margin of error for
+ * <code>double</code> values.
+ * <P>
+ * If an unexpected character is found in the <code>String</code>, a
+ * <code>NumberFormatException</code> will be thrown. Spaces are not
+ * allowed, and will cause the same exception.
+ *
+ * @XXX specify where/how we are not in accord with the spec.
+ *
+ * @param str the <code>String</code> to convert
+ * @return the value of the <code>String</code> as a <code>double</code>.
+ * @exception NumberFormatException when the string cannot be parsed to a
+ * <code>double</code>.
+ * @exception NullPointerException when the string is null.
+ * @see #MIN_VALUE
+ * @see #MAX_VALUE
+ * @see #POSITIVE_INFINITY
+ * @see #NEGATIVE_INFINITY
+ * @since 1.2
+ */
+ public native static double parseDouble (String s)
+ throws NumberFormatException;
+
+ /**
+ * Initialize JNI cache. This method is called only by the
+ * static initializer when using JNI.
+ */
+ private static void initIDs () { /* Not used in libgcj */ };
}
Index: java/lang/Float.java
===================================================================
RCS file: /cvs/gcc/gcc/libjava/java/lang/Float.java,v
retrieving revision 1.8
diff -u -r1.8 Float.java
--- Float.java 2001/02/09 22:13:33 1.8
+++ Float.java 2001/10/12 15:02:52
@@ -1,54 +1,194 @@
-/* Copyright (C) 1998, 1999, 2000, 2001 Free Software Foundation
+/* java.lang.Float
+ Copyright (C) 1998, 1999, 2000, 2001 Free Software Foundation, Inc.
- This file is part of libgcj.
+This file is part of GNU Classpath.
-This software is copyrighted work licensed under the terms of the
-Libgcj License. Please consult the file "LIBGCJ_LICENSE" for
-details. */
+GNU Classpath is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 2, or (at your option)
+any later version.
+GNU Classpath is distributed in the hope that it will be useful, but
+WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GNU Classpath; see the file COPYING. If not, write to the
+Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
+02111-1307 USA.
+
+As a special exception, if you link this library with other files to
+produce an executable, this library does not by itself cause the
+resulting executable to be covered by the GNU General Public License.
+This exception does not however invalidate any other reasons why the
+executable file might be covered by the GNU General Public License. */
+
+
package java.lang;
-/**
- * @author Andrew Haley <aph@cygnus.com>
- * @date September 25, 1998.
- */
+import gnu.classpath.Configuration;
+
/* Written using "Java Class Libraries", 2nd edition, ISBN 0-201-31002-3
* "The Java Language Specification", ISBN 0-201-63451-1
* plus online API docs for JDK 1.2 beta from http://www.javasoft.com.
* Status: Believed complete and correct.
*/
+/**
+ * Instances of class <code>Float</code> represent primitive
+ * <code>float</code> values.
+ *
+ * Additionally, this class provides various helper functions and variables
+ * related to floats.
+ *
+ * @author Paul Fisher
+ * @author Andrew Haley <aph@cygnus.com>
+ * @since JDK 1.0
+ */
public final class Float extends Number implements Comparable
{
+ /**
+ * The maximum positive value a <code>double</code> may represent
+ * is 3.4028235e+38f.
+ */
public static final float MAX_VALUE = 3.4028235e+38f;
+
+ /**
+ * The minimum positive value a <code>float</code> may represent
+ * is 1.4e-45.
+ */
public static final float MIN_VALUE = 1.4e-45f;
+
+ /**
+ * The value of a float representation -1.0/0.0, negative infinity.
+ */
public static final float NEGATIVE_INFINITY = -1.0f/0.0f;
+
+ /**
+ * The value of a float representation 1.0/0.0, positive infinity.
+ */
public static final float POSITIVE_INFINITY = 1.0f/0.0f;
- public static final float NaN = 0.0f/0.0f;
- // This initialization is seemingly circular, but it is accepted
- // by javac, and is handled specially by gcc.
- public static final Class TYPE = float.class;
+ /**
+ * All IEEE 754 values of NaN have the same value in Java.
+ */
+ public static final float NaN = 0.0f/0.0f;
- private float value;
+ /**
+ * The primitive type <code>float</code> is represented by this
+ * <code>Class</code> object.
+ */
+ public static final Class TYPE = VMClassLoader.getPrimitiveClass ("float");
+
+ /**
+ * The immutable value of this Float.
+ */
+ private final float value;
private static final long serialVersionUID = -2671257302660747028L;
+ static
+ {
+ if (Configuration.INIT_LOAD_LIBRARY)
+ {
+ System.loadLibrary ("javalang");
+ }
+ }
+
+ /**
+ * Create a <code>float</code> from the primitive <code>Float</code>
+ * specified.
+ *
+ * @param value the <code>Float</code> argument
+ */
public Float (float value)
{
this.value = value;
}
+ /**
+ * Create a <code>Float</code> from the primitive <code>double</code>
+ * specified.
+ *
+ * @param value the <code>double</code> argument
+ */
public Float (double value)
{
this.value = (float)value;
}
+ /**
+ * Create a <code>Float</code> from the specified <code>String</code>.
+ *
+ * This method calls <code>Float.parseFloat()</code>.
+ *
+ * @exception NumberFormatException when the <code>String</code> cannot
+ * be parsed into a <code>Float</code>.
+ * @param s the <code>String</code> to convert
+ * @see #parseFloat(java.lang.String)
+ */
public Float (String s) throws NumberFormatException
{
- this.value = valueOf (s).floatValue ();
+ this.value = parseFloat (s);
}
+ /**
+ * Parse the specified <code>String</code> as a <code>float</code>.
+ *
+ * The number is really read as <em>n * 10<sup>exponent</sup></em>. The
+ * first number is <em>n</em>, and if there is an "<code>E</code>"
+ * ("<code>e</code>" is also acceptable), then the integer after that is
+ * the exponent.
+ * <P>
+ * Here are the possible forms the number can take:
+ * <BR>
+ * <TABLE BORDER=1>
+ * <TR><TH>Form</TH><TH>Examples</TH></TR>
+ * <TR><TD><CODE>[+-]<number>[.]</CODE></TD><TD>345., -10, 12</TD></TR>
+ * <TR><TD><CODE>[+-]<number>.<number></CODE></TD><TD>40.2, 80.00, -12.30</TD></TR>
+ * <TR><TD><CODE>[+-]<number>[.]E[+-]<number></CODE></TD><TD>80E12, -12e+7, 4.E-123</TD></TR>
+ * <TR><TD><CODE>[+-]<number>.<number>E[+-]<number></CODE></TD><TD>6.02e-22, -40.2E+6, 12.3e9</TD></TR>
+ * </TABLE>
+ *
+ * "<code>[+-]</code>" means either a plus or minus sign may go there, or
+ * neither, in which case + is assumed.
+ * <BR>
+ * "<code>[.]</code>" means a dot may be placed here, but is optional.
+ * <BR>
+ * "<code><number></code>" means a string of digits (0-9), basically
+ * an integer. "<code><number>.<number></code>" is basically
+ * a real number, a floating-point value.
+ * <P>
+ * Remember that a <code>float</code> has a limited range. If the
+ * number you specify is greater than <code>Float.MAX_VALUE</code> or less
+ * than <code>-Float.MAX_VALUE</code>, it will be set at
+ * <code>Float.POSITIVE_INFINITY</code> or
+ * <code>Float.NEGATIVE_INFINITY</code>, respectively.
+ * <P>
+ *
+ * Note also that <code>float</code> does not have perfect precision. Many
+ * numbers cannot be precisely represented. The number you specify
+ * will be rounded to the nearest representable value.
+ * <code>Float.MIN_VALUE</code> is the margin of error for <code>float</code>
+ * values.
+ * <P>
+ * If an unexpected character is found in the <code>String</code>, a
+ * <code>NumberFormatException</code> will be thrown. Spaces are not
+ * allowed and will cause this exception to be thrown.
+ *
+ * @XXX specify where/how we are not in accord with the spec.
+ *
+ * @param str the <code>String</code> to convert
+ * @return the value of the <code>String</code> as a <code>float</code>.
+ * @exception NumberFormatException when the string cannot be parsed to a
+ * <code>float</code>.
+ * @since JDK 1.2
+ * @see #MIN_VALUE
+ * @see #MAX_VALUE
+ * @see #POSITIVE_INFINITY
+ * @see #NEGATIVE_INFINITY
+ */
public static float parseFloat (String s) throws NumberFormatException
{
// The spec says that parseFloat() should work like
@@ -57,11 +197,28 @@
return (float) Double.parseDouble (s);
}
+ /**
+ * Convert the <code>float</code> value of this <code>Float</code>
+ * to a <code>String</code>. This method calls
+ * <code>Float.toString(float)</code> to do its dirty work.
+ *
+ * @return the <code>String</code> representation of this <code>Float</code>.
+ * @see #toString(float)
+ */
public String toString ()
{
return toString (value);
}
+ /**
+ * If the <code>Object</code> is not <code>null</code>, is an
+ * <code>instanceof</code> <code>Float</code>, and represents
+ * the same primitive <code>float</code> value return
+ * <code>true</code>. Otherwise <code>false</code> is returned.
+ *
+ * @param obj the object to compare to
+ * @return whether the objects are semantically equal.
+ */
public boolean equals (Object obj)
{
if (!(obj instanceof Float))
@@ -72,56 +229,139 @@
return floatToIntBits (value) == floatToIntBits (f.floatValue ());
}
+ /**
+ * Return a hashcode representing this Object.
+ * <code>Float</code>'s hash code is calculated by calling the
+ * <code>floatToIntBits()</code> function.
+ * @return this Object's hash code.
+ * @see java.lang.Float.floatToIntBits(float)
+ */
public int hashCode ()
{
return floatToIntBits (value);
}
+ /**
+ * Return the value of this <code>Double</code> when cast to an
+ * <code>int</code>.
+ */
public int intValue ()
{
return (int) value;
}
+ /**
+ * Return the value of this <code>Double</code> when cast to a
+ * <code>long</code>.
+ */
public long longValue ()
{
return (long) value;
}
+ /**
+ * Return the value of this <code>Double</code> when cast to a
+ * <code>float</code>.
+ */
public float floatValue ()
{
return (float) value;
}
+ /**
+ * Return the primitive <code>double</code> value represented by this
+ * <code>Double</code>.
+ */
public double doubleValue ()
{
return (double) value;
}
-
- public byte byteValue ()
- {
- return (byte) value;
- }
-
- public short shortValue ()
- {
- return (short) value;
- }
-
- public static String toString (float v)
- {
- return Double.toString ((double) v, true);
- }
+ /**
+ * Convert the <code>float</code> to a <code>String</code>.
+ * <P>
+ *
+ * Floating-point string representation is fairly complex: here is a
+ * rundown of the possible values. "<CODE>[-]</CODE>" indicates that a
+ * negative sign will be printed if the value (or exponent) is negative.
+ * "<CODE><number></CODE>" means a string of digits (0-9).
+ * "<CODE><digit></CODE>" means a single digit (0-9).
+ * <P>
+ *
+ * <TABLE BORDER=1>
+ * <TR><TH>Value of Float</TH><TH>String Representation</TH></TR>
+ * <TR>
+ * <TD>[+-] 0</TD>
+ * <TD>[<CODE>-</CODE>]<CODE>0.0</CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>Between [+-] 10<SUP>-3</SUP> and 10<SUP>7</SUP></TD>
+ * <TD><CODE>[-]number.number</CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>Other numeric value</TD>
+ * <TD><CODE>[-]<digit>.<number>E[-]<number></CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>[+-] infinity</TD>
+ * <TD><CODE>[-]Infinity</CODE></TD>
+ * </TR>
+ * <TR>
+ * <TD>NaN</TD>
+ * <TD><CODE>NaN</CODE></TD>
+ * </TR>
+ * </TABLE>
+ *
+ * Yes, negative zero <EM>is</EM> a possible value. Note that there is
+ * <EM>always</EM> a <CODE>.</CODE> and at least one digit printed after
+ * it: even if the number is 3, it will be printed as <CODE>3.0</CODE>.
+ * After the ".", all digits will be printed except trailing zeros. No
+ * truncation or rounding is done by this function.
+ *
+ * @XXX specify where we are not in accord with the spec.
+ *
+ * @param f the <code>float</code> to convert
+ * @return the <code>String</code> representing the <code>float</code>.
+ */
+ public static String toString (float f)
+ {
+ return Double.toString ((double) f, true);
+ }
+
+ /**
+ * Return the result of calling <code>new Float(java.lang.String)</code>.
+ *
+ * @param s the <code>String</code> to convert to a <code>Float</code>.
+ * @return a new <code>Float</code> representing the <code>String</code>'s
+ * numeric value.
+ *
+ * @exception NumberFormatException thrown if <code>String</code> cannot
+ * be parsed as a <code>double</code>.
+ * @see #Float(java.lang.String)
+ * @see #parseFloat(java.lang.String)
+ */
public static Float valueOf (String s) throws NumberFormatException
{
- return new Float (Double.valueOf (s).floatValue ());
+ return new Float (s);
}
+ /**
+ * Return <code>true</code> if the value of this <code>Float</code>
+ * is the same as <code>NaN</code>, otherwise return <code>false</code>.
+ * @return whether this <code>Float</code> is <code>NaN</code>.
+ */
public boolean isNaN ()
{
return isNaN (value);
}
+ /**
+ * Return <code>true</code> if the <code>float</code> has the same
+ * value as <code>NaN</code>, otherwise return <code>false</code>.
+ *
+ * @param v the <code>float</code> to compare
+ * @return whether the argument is <code>NaN</code>.
+ */
public static boolean isNaN (float v)
{
int bits = floatToIntBits (v);
@@ -131,11 +371,26 @@
return e == 0x7f800000 && f != 0;
}
+ /**
+ * Return <code>true</code> if the value of this <code>Float</code>
+ * is the same as <code>NEGATIVE_INFINITY</code> or
+ * <code>POSITIVE_INFINITY</code>, otherwise return <code>false</code>.
+ *
+ * @return whether this <code>Float</code> is (-/+) infinity.
+ */
public boolean isInfinite ()
{
return isInfinite (value);
}
+ /**
+ * Return <code>true</code> if the <code>float</code> has a value
+ * equal to either <code>NEGATIVE_INFINITY</code> or
+ * <code>POSITIVE_INFINITY</code>, otherwise return <code>false</code>.
+ *
+ * @param v the <code>float</code> to compare
+ * @return whether the argument is (-/+) infinity.
+ */
public static boolean isInfinite (float v)
{
int bits = floatToIntBits (v);
@@ -144,27 +399,112 @@
return f == 0x7f800000;
}
+ /**
+ * Return the int bits of the specified <code>float</code>.
+ * The result of this function can be used as the argument to
+ * <code>Float.intBitsToFloat(long)</code> to obtain the
+ * original <code>float</code> value.
+ *
+ * @param value the <code>float</code> to convert
+ * @return the bits of the <code>float</code>.
+ */
public static native int floatToIntBits (float value);
+
+ /**
+ * Return the int bits of the specified <code>float</code>.
+ * The result of this function can be used as the argument to
+ * <code>Float.intBitsToFloat(long)</code> to obtain the
+ * original <code>float</code> value. The difference between
+ * this function and <code>floatToIntBits</code> is that this
+ * function does not collapse NaN values.
+ *
+ * @param value the <code>float</code> to convert
+ * @return the bits of the <code>float</code>.
+ */
public static native int floatToRawIntBits (float value);
+
+ /**
+ * Return the <code>float</code> represented by the long
+ * bits specified.
+ *
+ * @param bits the long bits representing a <code>double</code>
+ * @return the <code>float</code> represented by the bits.
+ */
public static native float intBitsToFloat (int bits);
- public int compareTo (Float d)
- {
- float v = d.value;
- if (isNaN (value))
- return isNaN (v) ? 1 : 0;
- else if (isNaN (v))
+ /**
+ * Returns 0 if the <code>float</code> value of the argument is
+ * equal to the value of this <code>Float</code>. Returns a number
+ * less than zero if the value of this <code>Float</code> is less
+ * than the <code>Float</code> value of the argument, and returns a
+ * number greater than zero if the value of this <code>Float</code>
+ * is greater than the <code>float</code> value of the argument.
+ * <br>
+ * <code>Float.NaN</code> is greater than any number other than itself,
+ * even <code>Float.POSITIVE_INFINITY</code>.
+ * <br>
+ * <code>0.0</code> is greater than <code>-0.0</code>.
+ *
+ * @param f the Float to compare to.
+ * @return 0 if the <code>Float</code>s are the same, < 0 if this
+ * <code>Float</code> is less than the <code>Float</code> in
+ * in question, or > 0 if it is greater.
+ *
+ * @since 1.2
+ */
+ public int compareTo (Float f)
+ {
+ return compare (value, f.value);
+ }
+
+ /**
+ * Returns 0 if the first argument is equal to the second argument.
+ * Returns a number less than zero if the first argument is less than the
+ * second argument, and returns a number greater than zero if the first
+ * argument is greater than the second argument.
+ * <br>
+ * <code>Float.NaN</code> is greater than any number other than itself,
+ * even <code>Float.POSITIVE_INFINITY</code>.
+ * <br>
+ * <code>0.0</code> is greater than <code>-0.0</code>.
+ *
+ * @param x the first float to compare.
+ * @param y the second float to compare.
+ * @return 0 if the arguments are the same, < 0 if the
+ * first argument is less than the second argument in
+ * in question, or > 0 if it is greater.
+ * @since 1.4
+ */
+ public static int compare (float x, float y)
+ {
+ if (isNaN (x))
+ return isNaN (y) ? 0 : 1;
+ if (isNaN (y))
return -1;
- else if (value == 0.0 && v == -0.0)
+ if ((x == 0.0) && (y == -0.0))
return 1;
- else if (value == -0.0 && v == 0.0)
+ if ((x == -0.0) && (y == 0.0))
return -1;
- else if (value == v)
+ if (x == y)
return 0;
- return value > v ? 1 : -1;
+ return (x > y) ? 1 : -1;
}
+ /**
+ * Compares the specified <code>Object</code> to this <code>Float</code>
+ * if and only if the <code>Object</code> is an instanceof
+ * <code>Float</code>.
+ * Otherwise it throws a <code>ClassCastException</code>
+ *
+ * @param o the Object to compare to.
+ * @return 0 if the <code>Float</code>s are the same, < 0 if this
+ * <code>Float</code> is less than the <code>Float</code> in
+ * in question, or > 0 if it is greater.
+ * @throws ClassCastException if the argument is not a <code>Float</code>
+ *
+ * @since 1.2
+ */
public int compareTo (Object o)
{
return compareTo ((Float) o);
----- End forwarded message -----
--
Stuff to read:
<http://www.toad.com/gnu/whatswrong.html>
What's Wrong with Copy Protection, by John Gilmore