In order to support fields of the same name in subclasses we are now prefixing them with name of the class that defines them. To provide convenient way to access them from generated bytecode and also directly from JavaScript, there is a getter/setter function for each field. It starts with _ followed by the field name. If called with a parameter, it sets the field, with a parameter it just returns it.
2 * Copyright (c) 1994, 2010, Oracle and/or its affiliates. All rights reserved.
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
5 * This code is free software; you can redistribute it and/or modify it
6 * under the terms of the GNU General Public License version 2 only, as
7 * published by the Free Software Foundation. Oracle designates this
8 * particular file as subject to the "Classpath" exception as provided
9 * by Oracle in the LICENSE file that accompanied this code.
11 * This code is distributed in the hope that it will be useful, but WITHOUT
12 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
13 * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
14 * version 2 for more details (a copy is included in the LICENSE file that
15 * accompanied this code).
17 * You should have received a copy of the GNU General Public License version
18 * 2 along with this work; if not, write to the Free Software Foundation,
19 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
21 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
22 * or visit www.oracle.com if you need additional information or have any
28 import org.apidesign.bck2brwsr.core.JavaScriptBody;
31 * The {@code Float} class wraps a value of primitive type
32 * {@code float} in an object. An object of type
33 * {@code Float} contains a single field whose type is
36 * <p>In addition, this class provides several methods for converting a
37 * {@code float} to a {@code String} and a
38 * {@code String} to a {@code float}, as well as other
39 * constants and methods useful when dealing with a
43 * @author Arthur van Hoff
44 * @author Joseph D. Darcy
47 public final class Float extends Number implements Comparable<Float> {
49 * A constant holding the positive infinity of type
50 * {@code float}. It is equal to the value returned by
51 * {@code Float.intBitsToFloat(0x7f800000)}.
53 public static final float POSITIVE_INFINITY = 1.0f / 0.0f;
56 * A constant holding the negative infinity of type
57 * {@code float}. It is equal to the value returned by
58 * {@code Float.intBitsToFloat(0xff800000)}.
60 public static final float NEGATIVE_INFINITY = -1.0f / 0.0f;
63 * A constant holding a Not-a-Number (NaN) value of type
64 * {@code float}. It is equivalent to the value returned by
65 * {@code Float.intBitsToFloat(0x7fc00000)}.
67 public static final float NaN = 0.0f / 0.0f;
70 * A constant holding the largest positive finite value of type
71 * {@code float}, (2-2<sup>-23</sup>)·2<sup>127</sup>.
72 * It is equal to the hexadecimal floating-point literal
73 * {@code 0x1.fffffeP+127f} and also equal to
74 * {@code Float.intBitsToFloat(0x7f7fffff)}.
76 public static final float MAX_VALUE = 0x1.fffffeP+127f; // 3.4028235e+38f
79 * A constant holding the smallest positive normal value of type
80 * {@code float}, 2<sup>-126</sup>. It is equal to the
81 * hexadecimal floating-point literal {@code 0x1.0p-126f} and also
82 * equal to {@code Float.intBitsToFloat(0x00800000)}.
86 public static final float MIN_NORMAL = 0x1.0p-126f; // 1.17549435E-38f
89 * A constant holding the smallest positive nonzero value of type
90 * {@code float}, 2<sup>-149</sup>. It is equal to the
91 * hexadecimal floating-point literal {@code 0x0.000002P-126f}
92 * and also equal to {@code Float.intBitsToFloat(0x1)}.
94 public static final float MIN_VALUE = 0x0.000002P-126f; // 1.4e-45f
97 * Maximum exponent a finite {@code float} variable may have. It
98 * is equal to the value returned by {@code
99 * Math.getExponent(Float.MAX_VALUE)}.
103 public static final int MAX_EXPONENT = 127;
106 * Minimum exponent a normalized {@code float} variable may have.
107 * It is equal to the value returned by {@code
108 * Math.getExponent(Float.MIN_NORMAL)}.
112 public static final int MIN_EXPONENT = -126;
115 * The number of bits used to represent a {@code float} value.
119 public static final int SIZE = 32;
122 * The {@code Class} instance representing the primitive type
127 public static final Class<Float> TYPE = Class.getPrimitiveClass("float");
130 * Returns a string representation of the {@code float}
131 * argument. All characters mentioned below are ASCII characters.
133 * <li>If the argument is NaN, the result is the string
135 * <li>Otherwise, the result is a string that represents the sign and
136 * magnitude (absolute value) of the argument. If the sign is
137 * negative, the first character of the result is
138 * '{@code -}' (<code>'\u002D'</code>); if the sign is
139 * positive, no sign character appears in the result. As for
140 * the magnitude <i>m</i>:
142 * <li>If <i>m</i> is infinity, it is represented by the characters
143 * {@code "Infinity"}; thus, positive infinity produces
144 * the result {@code "Infinity"} and negative infinity
145 * produces the result {@code "-Infinity"}.
146 * <li>If <i>m</i> is zero, it is represented by the characters
147 * {@code "0.0"}; thus, negative zero produces the result
148 * {@code "-0.0"} and positive zero produces the result
150 * <li> If <i>m</i> is greater than or equal to 10<sup>-3</sup> but
151 * less than 10<sup>7</sup>, then it is represented as the
152 * integer part of <i>m</i>, in decimal form with no leading
153 * zeroes, followed by '{@code .}'
154 * (<code>'\u002E'</code>), followed by one or more
155 * decimal digits representing the fractional part of
157 * <li> If <i>m</i> is less than 10<sup>-3</sup> or greater than or
158 * equal to 10<sup>7</sup>, then it is represented in
159 * so-called "computerized scientific notation." Let <i>n</i>
160 * be the unique integer such that 10<sup><i>n</i> </sup>≤
161 * <i>m</i> {@literal <} 10<sup><i>n</i>+1</sup>; then let <i>a</i>
162 * be the mathematically exact quotient of <i>m</i> and
163 * 10<sup><i>n</i></sup> so that 1 ≤ <i>a</i> {@literal <} 10.
164 * The magnitude is then represented as the integer part of
165 * <i>a</i>, as a single decimal digit, followed by
166 * '{@code .}' (<code>'\u002E'</code>), followed by
167 * decimal digits representing the fractional part of
168 * <i>a</i>, followed by the letter '{@code E}'
169 * (<code>'\u0045'</code>), followed by a representation
170 * of <i>n</i> as a decimal integer, as produced by the
171 * method {@link java.lang.Integer#toString(int)}.
175 * How many digits must be printed for the fractional part of
176 * <i>m</i> or <i>a</i>? There must be at least one digit
177 * to represent the fractional part, and beyond that as many, but
178 * only as many, more digits as are needed to uniquely distinguish
179 * the argument value from adjacent values of type
180 * {@code float}. That is, suppose that <i>x</i> is the
181 * exact mathematical value represented by the decimal
182 * representation produced by this method for a finite nonzero
183 * argument <i>f</i>. Then <i>f</i> must be the {@code float}
184 * value nearest to <i>x</i>; or, if two {@code float} values are
185 * equally close to <i>x</i>, then <i>f</i> must be one of
186 * them and the least significant bit of the significand of
187 * <i>f</i> must be {@code 0}.
189 * <p>To create localized string representations of a floating-point
190 * value, use subclasses of {@link java.text.NumberFormat}.
192 * @param f the float to be converted.
193 * @return a string representation of the argument.
195 public static String toString(float f) {
196 return Double.toString(f);
200 * Returns a hexadecimal string representation of the
201 * {@code float} argument. All characters mentioned below are
205 * <li>If the argument is NaN, the result is the string
207 * <li>Otherwise, the result is a string that represents the sign and
208 * magnitude (absolute value) of the argument. If the sign is negative,
209 * the first character of the result is '{@code -}'
210 * (<code>'\u002D'</code>); if the sign is positive, no sign character
211 * appears in the result. As for the magnitude <i>m</i>:
214 * <li>If <i>m</i> is infinity, it is represented by the string
215 * {@code "Infinity"}; thus, positive infinity produces the
216 * result {@code "Infinity"} and negative infinity produces
217 * the result {@code "-Infinity"}.
219 * <li>If <i>m</i> is zero, it is represented by the string
220 * {@code "0x0.0p0"}; thus, negative zero produces the result
221 * {@code "-0x0.0p0"} and positive zero produces the result
224 * <li>If <i>m</i> is a {@code float} value with a
225 * normalized representation, substrings are used to represent the
226 * significand and exponent fields. The significand is
227 * represented by the characters {@code "0x1."}
228 * followed by a lowercase hexadecimal representation of the rest
229 * of the significand as a fraction. Trailing zeros in the
230 * hexadecimal representation are removed unless all the digits
231 * are zero, in which case a single zero is used. Next, the
232 * exponent is represented by {@code "p"} followed
233 * by a decimal string of the unbiased exponent as if produced by
234 * a call to {@link Integer#toString(int) Integer.toString} on the
237 * <li>If <i>m</i> is a {@code float} value with a subnormal
238 * representation, the significand is represented by the
239 * characters {@code "0x0."} followed by a
240 * hexadecimal representation of the rest of the significand as a
241 * fraction. Trailing zeros in the hexadecimal representation are
242 * removed. Next, the exponent is represented by
243 * {@code "p-126"}. Note that there must be at
244 * least one nonzero digit in a subnormal significand.
251 * <caption><h3>Examples</h3></caption>
252 * <tr><th>Floating-point Value</th><th>Hexadecimal String</th>
253 * <tr><td>{@code 1.0}</td> <td>{@code 0x1.0p0}</td>
254 * <tr><td>{@code -1.0}</td> <td>{@code -0x1.0p0}</td>
255 * <tr><td>{@code 2.0}</td> <td>{@code 0x1.0p1}</td>
256 * <tr><td>{@code 3.0}</td> <td>{@code 0x1.8p1}</td>
257 * <tr><td>{@code 0.5}</td> <td>{@code 0x1.0p-1}</td>
258 * <tr><td>{@code 0.25}</td> <td>{@code 0x1.0p-2}</td>
259 * <tr><td>{@code Float.MAX_VALUE}</td>
260 * <td>{@code 0x1.fffffep127}</td>
261 * <tr><td>{@code Minimum Normal Value}</td>
262 * <td>{@code 0x1.0p-126}</td>
263 * <tr><td>{@code Maximum Subnormal Value}</td>
264 * <td>{@code 0x0.fffffep-126}</td>
265 * <tr><td>{@code Float.MIN_VALUE}</td>
266 * <td>{@code 0x0.000002p-126}</td>
268 * @param f the {@code float} to be converted.
269 * @return a hex string representation of the argument.
271 * @author Joseph D. Darcy
273 public static String toHexString(float f) {
274 throw new UnsupportedOperationException();
275 // if (Math.abs(f) < FloatConsts.MIN_NORMAL
276 // && f != 0.0f ) {// float subnormal
277 // // Adjust exponent to create subnormal double, then
278 // // replace subnormal double exponent with subnormal float
280 // String s = Double.toHexString(FpUtils.scalb((double)f,
282 // DoubleConsts.MIN_EXPONENT-
283 // FloatConsts.MIN_EXPONENT));
284 // return s.replaceFirst("p-1022$", "p-126");
286 // else // double string will be the same as float string
287 // return Double.toHexString(f);
291 * Returns a {@code Float} object holding the
292 * {@code float} value represented by the argument string
295 * <p>If {@code s} is {@code null}, then a
296 * {@code NullPointerException} is thrown.
298 * <p>Leading and trailing whitespace characters in {@code s}
299 * are ignored. Whitespace is removed as if by the {@link
300 * String#trim} method; that is, both ASCII space and control
301 * characters are removed. The rest of {@code s} should
302 * constitute a <i>FloatValue</i> as described by the lexical
307 * <dt><i>FloatValue:</i>
308 * <dd><i>Sign<sub>opt</sub></i> {@code NaN}
309 * <dd><i>Sign<sub>opt</sub></i> {@code Infinity}
310 * <dd><i>Sign<sub>opt</sub> FloatingPointLiteral</i>
311 * <dd><i>Sign<sub>opt</sub> HexFloatingPointLiteral</i>
312 * <dd><i>SignedInteger</i>
318 * <dt><i>HexFloatingPointLiteral</i>:
319 * <dd> <i>HexSignificand BinaryExponent FloatTypeSuffix<sub>opt</sub></i>
325 * <dt><i>HexSignificand:</i>
326 * <dd><i>HexNumeral</i>
327 * <dd><i>HexNumeral</i> {@code .}
328 * <dd>{@code 0x} <i>HexDigits<sub>opt</sub>
329 * </i>{@code .}<i> HexDigits</i>
330 * <dd>{@code 0X}<i> HexDigits<sub>opt</sub>
331 * </i>{@code .} <i>HexDigits</i>
337 * <dt><i>BinaryExponent:</i>
338 * <dd><i>BinaryExponentIndicator SignedInteger</i>
344 * <dt><i>BinaryExponentIndicator:</i>
351 * where <i>Sign</i>, <i>FloatingPointLiteral</i>,
352 * <i>HexNumeral</i>, <i>HexDigits</i>, <i>SignedInteger</i> and
353 * <i>FloatTypeSuffix</i> are as defined in the lexical structure
355 * <cite>The Java™ Language Specification</cite>,
356 * except that underscores are not accepted between digits.
357 * If {@code s} does not have the form of
358 * a <i>FloatValue</i>, then a {@code NumberFormatException}
359 * is thrown. Otherwise, {@code s} is regarded as
360 * representing an exact decimal value in the usual
361 * "computerized scientific notation" or as an exact
362 * hexadecimal value; this exact numerical value is then
363 * conceptually converted to an "infinitely precise"
364 * binary value that is then rounded to type {@code float}
365 * by the usual round-to-nearest rule of IEEE 754 floating-point
366 * arithmetic, which includes preserving the sign of a zero
369 * Note that the round-to-nearest rule also implies overflow and
370 * underflow behaviour; if the exact value of {@code s} is large
371 * enough in magnitude (greater than or equal to ({@link
372 * #MAX_VALUE} + {@link Math#ulp(float) ulp(MAX_VALUE)}/2),
373 * rounding to {@code float} will result in an infinity and if the
374 * exact value of {@code s} is small enough in magnitude (less
375 * than or equal to {@link #MIN_VALUE}/2), rounding to float will
378 * Finally, after rounding a {@code Float} object representing
379 * this {@code float} value is returned.
381 * <p>To interpret localized string representations of a
382 * floating-point value, use subclasses of {@link
383 * java.text.NumberFormat}.
385 * <p>Note that trailing format specifiers, specifiers that
386 * determine the type of a floating-point literal
387 * ({@code 1.0f} is a {@code float} value;
388 * {@code 1.0d} is a {@code double} value), do
389 * <em>not</em> influence the results of this method. In other
390 * words, the numerical value of the input string is converted
391 * directly to the target floating-point type. In general, the
392 * two-step sequence of conversions, string to {@code double}
393 * followed by {@code double} to {@code float}, is
394 * <em>not</em> equivalent to converting a string directly to
395 * {@code float}. For example, if first converted to an
396 * intermediate {@code double} and then to
397 * {@code float}, the string<br>
398 * {@code "1.00000017881393421514957253748434595763683319091796875001d"}<br>
399 * results in the {@code float} value
400 * {@code 1.0000002f}; if the string is converted directly to
401 * {@code float}, <code>1.000000<b>1</b>f</code> results.
403 * <p>To avoid calling this method on an invalid string and having
404 * a {@code NumberFormatException} be thrown, the documentation
405 * for {@link Double#valueOf Double.valueOf} lists a regular
406 * expression which can be used to screen the input.
408 * @param s the string to be parsed.
409 * @return a {@code Float} object holding the value
410 * represented by the {@code String} argument.
411 * @throws NumberFormatException if the string does not contain a
414 public static Float valueOf(String s) throws NumberFormatException {
415 throw new UnsupportedOperationException();
416 // return new Float(FloatingDecimal.readJavaFormatString(s).floatValue());
420 * Returns a {@code Float} instance representing the specified
421 * {@code float} value.
422 * If a new {@code Float} instance is not required, this method
423 * should generally be used in preference to the constructor
424 * {@link #Float(float)}, as this method is likely to yield
425 * significantly better space and time performance by caching
426 * frequently requested values.
428 * @param f a float value.
429 * @return a {@code Float} instance representing {@code f}.
432 public static Float valueOf(float f) {
437 * Returns a new {@code float} initialized to the value
438 * represented by the specified {@code String}, as performed
439 * by the {@code valueOf} method of class {@code Float}.
441 * @param s the string to be parsed.
442 * @return the {@code float} value represented by the string
444 * @throws NullPointerException if the string is null
445 * @throws NumberFormatException if the string does not contain a
446 * parsable {@code float}.
447 * @see java.lang.Float#valueOf(String)
450 public static float parseFloat(String s) throws NumberFormatException {
451 throw new UnsupportedOperationException();
452 // return FloatingDecimal.readJavaFormatString(s).floatValue();
456 * Returns {@code true} if the specified number is a
457 * Not-a-Number (NaN) value, {@code false} otherwise.
459 * @param v the value to be tested.
460 * @return {@code true} if the argument is NaN;
461 * {@code false} otherwise.
463 static public boolean isNaN(float v) {
468 * Returns {@code true} if the specified number is infinitely
469 * large in magnitude, {@code false} otherwise.
471 * @param v the value to be tested.
472 * @return {@code true} if the argument is positive infinity or
473 * negative infinity; {@code false} otherwise.
475 static public boolean isInfinite(float v) {
476 return (v == POSITIVE_INFINITY) || (v == NEGATIVE_INFINITY);
480 * The value of the Float.
484 private final float value;
487 * Constructs a newly allocated {@code Float} object that
488 * represents the primitive {@code float} argument.
490 * @param value the value to be represented by the {@code Float}.
492 public Float(float value) {
497 * Constructs a newly allocated {@code Float} object that
498 * represents the argument converted to type {@code float}.
500 * @param value the value to be represented by the {@code Float}.
502 public Float(double value) {
503 this.value = (float)value;
507 * Constructs a newly allocated {@code Float} object that
508 * represents the floating-point value of type {@code float}
509 * represented by the string. The string is converted to a
510 * {@code float} value as if by the {@code valueOf} method.
512 * @param s a string to be converted to a {@code Float}.
513 * @throws NumberFormatException if the string does not contain a
515 * @see java.lang.Float#valueOf(java.lang.String)
517 public Float(String s) throws NumberFormatException {
518 // REMIND: this is inefficient
519 this(valueOf(s).floatValue());
523 * Returns {@code true} if this {@code Float} value is a
524 * Not-a-Number (NaN), {@code false} otherwise.
526 * @return {@code true} if the value represented by this object is
527 * NaN; {@code false} otherwise.
529 public boolean isNaN() {
534 * Returns {@code true} if this {@code Float} value is
535 * infinitely large in magnitude, {@code false} otherwise.
537 * @return {@code true} if the value represented by this object is
538 * positive infinity or negative infinity;
539 * {@code false} otherwise.
541 public boolean isInfinite() {
542 return isInfinite(value);
546 * Returns a string representation of this {@code Float} object.
547 * The primitive {@code float} value represented by this object
548 * is converted to a {@code String} exactly as if by the method
549 * {@code toString} of one argument.
551 * @return a {@code String} representation of this object.
552 * @see java.lang.Float#toString(float)
554 public String toString() {
555 return Float.toString(value);
559 * Returns the value of this {@code Float} as a {@code byte} (by
560 * casting to a {@code byte}).
562 * @return the {@code float} value represented by this object
563 * converted to type {@code byte}
565 public byte byteValue() {
570 * Returns the value of this {@code Float} as a {@code short} (by
571 * casting to a {@code short}).
573 * @return the {@code float} value represented by this object
574 * converted to type {@code short}
577 public short shortValue() {
582 * Returns the value of this {@code Float} as an {@code int} (by
583 * casting to type {@code int}).
585 * @return the {@code float} value represented by this object
586 * converted to type {@code int}
588 public int intValue() {
593 * Returns value of this {@code Float} as a {@code long} (by
594 * casting to type {@code long}).
596 * @return the {@code float} value represented by this object
597 * converted to type {@code long}
599 public long longValue() {
604 * Returns the {@code float} value of this {@code Float} object.
606 * @return the {@code float} value represented by this object
608 public float floatValue() {
613 * Returns the {@code double} value of this {@code Float} object.
615 * @return the {@code float} value represented by this
616 * object is converted to type {@code double} and the
617 * result of the conversion is returned.
619 public double doubleValue() {
620 return (double)value;
624 * Returns a hash code for this {@code Float} object. The
625 * result is the integer bit representation, exactly as produced
626 * by the method {@link #floatToIntBits(float)}, of the primitive
627 * {@code float} value represented by this {@code Float}
630 * @return a hash code value for this object.
632 public int hashCode() {
633 return floatToIntBits(value);
638 * Compares this object against the specified object. The result
639 * is {@code true} if and only if the argument is not
640 * {@code null} and is a {@code Float} object that
641 * represents a {@code float} with the same value as the
642 * {@code float} represented by this object. For this
643 * purpose, two {@code float} values are considered to be the
644 * same if and only if the method {@link #floatToIntBits(float)}
645 * returns the identical {@code int} value when applied to
648 * <p>Note that in most cases, for two instances of class
649 * {@code Float}, {@code f1} and {@code f2}, the value
650 * of {@code f1.equals(f2)} is {@code true} if and only if
653 * f1.floatValue() == f2.floatValue()
654 * </pre></blockquote>
656 * <p>also has the value {@code true}. However, there are two exceptions:
658 * <li>If {@code f1} and {@code f2} both represent
659 * {@code Float.NaN}, then the {@code equals} method returns
660 * {@code true}, even though {@code Float.NaN==Float.NaN}
661 * has the value {@code false}.
662 * <li>If {@code f1} represents {@code +0.0f} while
663 * {@code f2} represents {@code -0.0f}, or vice
664 * versa, the {@code equal} test has the value
665 * {@code false}, even though {@code 0.0f==-0.0f}
666 * has the value {@code true}.
669 * This definition allows hash tables to operate properly.
671 * @param obj the object to be compared
672 * @return {@code true} if the objects are the same;
673 * {@code false} otherwise.
674 * @see java.lang.Float#floatToIntBits(float)
676 public boolean equals(Object obj) {
677 return (obj instanceof Float)
678 && (floatToIntBits(((Float)obj).value) == floatToIntBits(value));
682 * Returns a representation of the specified floating-point value
683 * according to the IEEE 754 floating-point "single format" bit
686 * <p>Bit 31 (the bit that is selected by the mask
687 * {@code 0x80000000}) represents the sign of the floating-point
689 * Bits 30-23 (the bits that are selected by the mask
690 * {@code 0x7f800000}) represent the exponent.
691 * Bits 22-0 (the bits that are selected by the mask
692 * {@code 0x007fffff}) represent the significand (sometimes called
693 * the mantissa) of the floating-point number.
695 * <p>If the argument is positive infinity, the result is
696 * {@code 0x7f800000}.
698 * <p>If the argument is negative infinity, the result is
699 * {@code 0xff800000}.
701 * <p>If the argument is NaN, the result is {@code 0x7fc00000}.
703 * <p>In all cases, the result is an integer that, when given to the
704 * {@link #intBitsToFloat(int)} method, will produce a floating-point
705 * value the same as the argument to {@code floatToIntBits}
706 * (except all NaN values are collapsed to a single
707 * "canonical" NaN value).
709 * @param value a floating-point number.
710 * @return the bits that represent the floating-point number.
712 public static int floatToIntBits(float value) {
713 throw new UnsupportedOperationException();
714 // int result = floatToRawIntBits(value);
715 // // Check for NaN based on values of bit fields, maximum
716 // // exponent and nonzero significand.
717 // if ( ((result & FloatConsts.EXP_BIT_MASK) ==
718 // FloatConsts.EXP_BIT_MASK) &&
719 // (result & FloatConsts.SIGNIF_BIT_MASK) != 0)
720 // result = 0x7fc00000;
725 * Returns a representation of the specified floating-point value
726 * according to the IEEE 754 floating-point "single format" bit
727 * layout, preserving Not-a-Number (NaN) values.
729 * <p>Bit 31 (the bit that is selected by the mask
730 * {@code 0x80000000}) represents the sign of the floating-point
732 * Bits 30-23 (the bits that are selected by the mask
733 * {@code 0x7f800000}) represent the exponent.
734 * Bits 22-0 (the bits that are selected by the mask
735 * {@code 0x007fffff}) represent the significand (sometimes called
736 * the mantissa) of the floating-point number.
738 * <p>If the argument is positive infinity, the result is
739 * {@code 0x7f800000}.
741 * <p>If the argument is negative infinity, the result is
742 * {@code 0xff800000}.
744 * <p>If the argument is NaN, the result is the integer representing
745 * the actual NaN value. Unlike the {@code floatToIntBits}
746 * method, {@code floatToRawIntBits} does not collapse all the
747 * bit patterns encoding a NaN to a single "canonical"
750 * <p>In all cases, the result is an integer that, when given to the
751 * {@link #intBitsToFloat(int)} method, will produce a
752 * floating-point value the same as the argument to
753 * {@code floatToRawIntBits}.
755 * @param value a floating-point number.
756 * @return the bits that represent the floating-point number.
759 public static native int floatToRawIntBits(float value);
762 * Returns the {@code float} value corresponding to a given
763 * bit representation.
764 * The argument is considered to be a representation of a
765 * floating-point value according to the IEEE 754 floating-point
766 * "single format" bit layout.
768 * <p>If the argument is {@code 0x7f800000}, the result is positive
771 * <p>If the argument is {@code 0xff800000}, the result is negative
774 * <p>If the argument is any value in the range
775 * {@code 0x7f800001} through {@code 0x7fffffff} or in
776 * the range {@code 0xff800001} through
777 * {@code 0xffffffff}, the result is a NaN. No IEEE 754
778 * floating-point operation provided by Java can distinguish
779 * between two NaN values of the same type with different bit
780 * patterns. Distinct values of NaN are only distinguishable by
781 * use of the {@code Float.floatToRawIntBits} method.
783 * <p>In all other cases, let <i>s</i>, <i>e</i>, and <i>m</i> be three
784 * values that can be computed from the argument:
787 * int s = ((bits >> 31) == 0) ? 1 : -1;
788 * int e = ((bits >> 23) & 0xff);
790 * (bits & 0x7fffff) << 1 :
791 * (bits & 0x7fffff) | 0x800000;
792 * </pre></blockquote>
794 * Then the floating-point result equals the value of the mathematical
795 * expression <i>s</i>·<i>m</i>·2<sup><i>e</i>-150</sup>.
797 * <p>Note that this method may not be able to return a
798 * {@code float} NaN with exactly same bit pattern as the
799 * {@code int} argument. IEEE 754 distinguishes between two
800 * kinds of NaNs, quiet NaNs and <i>signaling NaNs</i>. The
801 * differences between the two kinds of NaN are generally not
802 * visible in Java. Arithmetic operations on signaling NaNs turn
803 * them into quiet NaNs with a different, but often similar, bit
804 * pattern. However, on some processors merely copying a
805 * signaling NaN also performs that conversion. In particular,
806 * copying a signaling NaN to return it to the calling method may
807 * perform this conversion. So {@code intBitsToFloat} may
808 * not be able to return a {@code float} with a signaling NaN
809 * bit pattern. Consequently, for some {@code int} values,
810 * {@code floatToRawIntBits(intBitsToFloat(start))} may
811 * <i>not</i> equal {@code start}. Moreover, which
812 * particular bit patterns represent signaling NaNs is platform
813 * dependent; although all NaN bit patterns, quiet or signaling,
814 * must be in the NaN range identified above.
816 * @param bits an integer.
817 * @return the {@code float} floating-point value with the same bit
820 @JavaScriptBody(args = "bits",
822 "if (bits === 0x7f800000) return Number.POSITIVE_INFINITY;\n"
823 + "if (bits === 0xff800000) return Number.NEGATIVE_INFINITY;\n"
824 + "if (bits >= 0x7f800001 && bits <= 0xffffffff) return Number.NaN;\n"
825 + "var s = ((bits >> 31) == 0) ? 1 : -1;\n"
826 + "var e = ((bits >> 23) & 0xff);\n"
827 + "var m = (e == 0) ?\n"
828 + " (bits & 0x7fffff) << 1 :\n"
829 + " (bits & 0x7fffff) | 0x800000;\n"
830 + "return s * m * Math.pow(2.0, e - 150);\n"
832 public static native float intBitsToFloat(int bits);
835 * Compares two {@code Float} objects numerically. There are
836 * two ways in which comparisons performed by this method differ
837 * from those performed by the Java language numerical comparison
838 * operators ({@code <, <=, ==, >=, >}) when
839 * applied to primitive {@code float} values:
842 * {@code Float.NaN} is considered by this method to
843 * be equal to itself and greater than all other
844 * {@code float} values
845 * (including {@code Float.POSITIVE_INFINITY}).
847 * {@code 0.0f} is considered by this method to be greater
848 * than {@code -0.0f}.
851 * This ensures that the <i>natural ordering</i> of {@code Float}
852 * objects imposed by this method is <i>consistent with equals</i>.
854 * @param anotherFloat the {@code Float} to be compared.
855 * @return the value {@code 0} if {@code anotherFloat} is
856 * numerically equal to this {@code Float}; a value
857 * less than {@code 0} if this {@code Float}
858 * is numerically less than {@code anotherFloat};
859 * and a value greater than {@code 0} if this
860 * {@code Float} is numerically greater than
861 * {@code anotherFloat}.
864 * @see Comparable#compareTo(Object)
866 public int compareTo(Float anotherFloat) {
867 return Float.compare(value, anotherFloat.value);
871 * Compares the two specified {@code float} values. The sign
872 * of the integer value returned is the same as that of the
873 * integer that would be returned by the call:
875 * new Float(f1).compareTo(new Float(f2))
878 * @param f1 the first {@code float} to compare.
879 * @param f2 the second {@code float} to compare.
880 * @return the value {@code 0} if {@code f1} is
881 * numerically equal to {@code f2}; a value less than
882 * {@code 0} if {@code f1} is numerically less than
883 * {@code f2}; and a value greater than {@code 0}
884 * if {@code f1} is numerically greater than
888 public static int compare(float f1, float f2) {
890 return -1; // Neither val is NaN, thisVal is smaller
892 return 1; // Neither val is NaN, thisVal is larger
894 // Cannot use floatToRawIntBits because of possibility of NaNs.
895 int thisBits = Float.floatToIntBits(f1);
896 int anotherBits = Float.floatToIntBits(f2);
898 return (thisBits == anotherBits ? 0 : // Values are equal
899 (thisBits < anotherBits ? -1 : // (-0.0, 0.0) or (!NaN, NaN)
900 1)); // (0.0, -0.0) or (NaN, !NaN)
903 /** use serialVersionUID from JDK 1.0.2 for interoperability */
904 private static final long serialVersionUID = -2671257302660747028L;