3202N/A * Copyright (c) 1994, 2010, Oracle and/or its affiliates. All rights reserved. 0N/A * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. 0N/A * This code is free software; you can redistribute it and/or modify it 0N/A * under the terms of the GNU General Public License version 2 only, as 2362N/A * published by the Free Software Foundation. Oracle designates this 0N/A * particular file as subject to the "Classpath" exception as provided 2362N/A * by Oracle in the LICENSE file that accompanied this code. 0N/A * This code is distributed in the hope that it will be useful, but WITHOUT 0N/A * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or 0N/A * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 0N/A * version 2 for more details (a copy is included in the LICENSE file that 0N/A * accompanied this code). 0N/A * You should have received a copy of the GNU General Public License version 0N/A * 2 along with this work; if not, write to the Free Software Foundation, 0N/A * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. 2362N/A * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA 2362N/A * or visit www.oracle.com if you need additional information or have any 0N/A * The {@code Double} class wraps a value of the primitive type 0N/A * {@code double} in an object. An object of type 0N/A * {@code Double} contains a single field whose type is 0N/A * <p>In addition, this class provides several methods for converting a 0N/A * {@code double} to a {@code String} and a 0N/A * {@code String} to a {@code double}, as well as other 0N/A * constants and methods useful when dealing with a 0N/A * @author Lee Boynton 0N/A * @author Arthur van Hoff 0N/A * @author Joseph D. Darcy 0N/A * A constant holding the positive infinity of type 0N/A * {@code double}. It is equal to the value returned by 0N/A * {@code Double.longBitsToDouble(0x7ff0000000000000L)}. 0N/A * A constant holding the negative infinity of type 0N/A * {@code double}. It is equal to the value returned by 0N/A * {@code Double.longBitsToDouble(0xfff0000000000000L)}. 0N/A * A constant holding a Not-a-Number (NaN) value of type 0N/A * {@code double}. It is equivalent to the value returned by 0N/A * {@code Double.longBitsToDouble(0x7ff8000000000000L)}. 0N/A public static final double NaN =
0.0d /
0.0;
0N/A * A constant holding the largest positive finite value of type 0N/A * (2-2<sup>-52</sup>)·2<sup>1023</sup>. It is equal to 0N/A * the hexadecimal floating-point literal 0N/A * {@code 0x1.fffffffffffffP+1023} and also equal to 0N/A * {@code Double.longBitsToDouble(0x7fefffffffffffffL)}. 0N/A * A constant holding the smallest positive normal value of type 0N/A * {@code double}, 2<sup>-1022</sup>. It is equal to the 0N/A * hexadecimal floating-point literal {@code 0x1.0p-1022} and also 0N/A * equal to {@code Double.longBitsToDouble(0x0010000000000000L)}. 0N/A public static final double MIN_NORMAL =
0x1.
0p-
1022;
// 2.2250738585072014E-308 0N/A * A constant holding the smallest positive nonzero value of type 0N/A * {@code double}, 2<sup>-1074</sup>. It is equal to the 0N/A * hexadecimal floating-point literal 0N/A * {@code 0x0.0000000000001P-1022} and also equal to 0N/A * {@code Double.longBitsToDouble(0x1L)}. 0N/A public static final double MIN_VALUE =
0x0.
0000000000001P-
1022;
// 4.9e-324 0N/A * Maximum exponent a finite {@code double} variable may have. 0N/A * It is equal to the value returned by 0N/A * {@code Math.getExponent(Double.MAX_VALUE)}. 0N/A * Minimum exponent a normalized {@code double} variable may 0N/A * have. It is equal to the value returned by 0N/A * {@code Math.getExponent(Double.MIN_NORMAL)}. 0N/A * The number of bits used to represent a {@code double} value. 0N/A * The {@code Class} instance representing the primitive type 0N/A * Returns a string representation of the {@code double} 0N/A * argument. All characters mentioned below are ASCII characters. 0N/A * <li>If the argument is NaN, the result is the string 0N/A * <li>Otherwise, the result is a string that represents the sign and 0N/A * magnitude (absolute value) of the argument. If the sign is negative, 0N/A * the first character of the result is '{@code -}' 0N/A * (<code>'\u002D'</code>); if the sign is positive, no sign character 0N/A * appears in the result. As for the magnitude <i>m</i>: 0N/A * <li>If <i>m</i> is infinity, it is represented by the characters 0N/A * {@code "Infinity"}; thus, positive infinity produces the result 0N/A * {@code "Infinity"} and negative infinity produces the result 0N/A * {@code "-Infinity"}. 0N/A * <li>If <i>m</i> is zero, it is represented by the characters 0N/A * {@code "0.0"}; thus, negative zero produces the result 0N/A * {@code "-0.0"} and positive zero produces the result 0N/A * <li>If <i>m</i> is greater than or equal to 10<sup>-3</sup> but less 0N/A * than 10<sup>7</sup>, then it is represented as the integer part of 0N/A * <i>m</i>, in decimal form with no leading zeroes, followed by 0N/A * '{@code .}' (<code>'\u002E'</code>), followed by one or 0N/A * more decimal digits representing the fractional part of <i>m</i>. 0N/A * <li>If <i>m</i> is less than 10<sup>-3</sup> or greater than or 0N/A * equal to 10<sup>7</sup>, then it is represented in so-called 0N/A * "computerized scientific notation." Let <i>n</i> be the unique 0N/A * integer such that 10<sup><i>n</i></sup> ≤ <i>m</i> {@literal <} 0N/A * 10<sup><i>n</i>+1</sup>; then let <i>a</i> be the 0N/A * mathematically exact quotient of <i>m</i> and 0N/A * 10<sup><i>n</i></sup> so that 1 ≤ <i>a</i> {@literal <} 10. The 0N/A * magnitude is then represented as the integer part of <i>a</i>, 0N/A * as a single decimal digit, followed by '{@code .}' 0N/A * (<code>'\u002E'</code>), followed by decimal digits 0N/A * representing the fractional part of <i>a</i>, followed by the 0N/A * letter '{@code E}' (<code>'\u0045'</code>), followed 0N/A * by a representation of <i>n</i> as a decimal integer, as 0N/A * produced by the method {@link Integer#toString(int)}. 0N/A * How many digits must be printed for the fractional part of 0N/A * <i>m</i> or <i>a</i>? There must be at least one digit to represent 0N/A * the fractional part, and beyond that as many, but only as many, more 0N/A * digits as are needed to uniquely distinguish the argument value from 0N/A * adjacent values of type {@code double}. That is, suppose that 0N/A * <i>x</i> is the exact mathematical value represented by the decimal 0N/A * representation produced by this method for a finite nonzero argument 0N/A * <i>d</i>. Then <i>d</i> must be the {@code double} value nearest 0N/A * to <i>x</i>; or if two {@code double} values are equally close 0N/A * to <i>x</i>, then <i>d</i> must be one of them and the least 0N/A * significant bit of the significand of <i>d</i> must be {@code 0}. 0N/A * <p>To create localized string representations of a floating-point 0N/A * value, use subclasses of {@link java.text.NumberFormat}. 0N/A * @param d the {@code double} to be converted. 0N/A * @return a string representation of the argument. 0N/A * Returns a hexadecimal string representation of the 0N/A * {@code double} argument. All characters mentioned below 0N/A * are ASCII characters. 0N/A * <li>If the argument is NaN, the result is the string 0N/A * <li>Otherwise, the result is a string that represents the sign 0N/A * and magnitude of the argument. If the sign is negative, the 0N/A * first character of the result is '{@code -}' 0N/A * (<code>'\u002D'</code>); if the sign is positive, no sign 0N/A * character appears in the result. As for the magnitude <i>m</i>: 0N/A * <li>If <i>m</i> is infinity, it is represented by the string 0N/A * {@code "Infinity"}; thus, positive infinity produces the 0N/A * result {@code "Infinity"} and negative infinity produces 0N/A * the result {@code "-Infinity"}. 0N/A * <li>If <i>m</i> is zero, it is represented by the string 0N/A * {@code "0x0.0p0"}; thus, negative zero produces the result 0N/A * {@code "-0x0.0p0"} and positive zero produces the result 0N/A * {@code "0x0.0p0"}. 0N/A * <li>If <i>m</i> is a {@code double} value with a 0N/A * normalized representation, substrings are used to represent the 0N/A * significand and exponent fields. The significand is 0N/A * represented by the characters {@code "0x1."} 0N/A * followed by a lowercase hexadecimal representation of the rest 0N/A * of the significand as a fraction. Trailing zeros in the 0N/A * hexadecimal representation are removed unless all the digits 0N/A * are zero, in which case a single zero is used. Next, the 0N/A * exponent is represented by {@code "p"} followed 0N/A * by a decimal string of the unbiased exponent as if produced by 0N/A * a call to {@link Integer#toString(int) Integer.toString} on the 0N/A * <li>If <i>m</i> is a {@code double} value with a subnormal 0N/A * representation, the significand is represented by the 0N/A * characters {@code "0x0."} followed by a 0N/A * hexadecimal representation of the rest of the significand as a 0N/A * fraction. Trailing zeros in the hexadecimal representation are 0N/A * removed. Next, the exponent is represented by 0N/A * {@code "p-1022"}. Note that there must be at 0N/A * least one nonzero digit in a subnormal significand. 0N/A * <caption><h3>Examples</h3></caption> 0N/A * <tr><th>Floating-point Value</th><th>Hexadecimal String</th> 0N/A * <tr><td>{@code 1.0}</td> <td>{@code 0x1.0p0}</td> 0N/A * <tr><td>{@code -1.0}</td> <td>{@code -0x1.0p0}</td> 0N/A * <tr><td>{@code 2.0}</td> <td>{@code 0x1.0p1}</td> 0N/A * <tr><td>{@code 3.0}</td> <td>{@code 0x1.8p1}</td> 0N/A * <tr><td>{@code 0.5}</td> <td>{@code 0x1.0p-1}</td> 0N/A * <tr><td>{@code 0.25}</td> <td>{@code 0x1.0p-2}</td> 0N/A * <tr><td>{@code Double.MAX_VALUE}</td> 0N/A * <td>{@code 0x1.fffffffffffffp1023}</td> 0N/A * <tr><td>{@code Minimum Normal Value}</td> 0N/A * <td>{@code 0x1.0p-1022}</td> 0N/A * <tr><td>{@code Maximum Subnormal Value}</td> 0N/A * <td>{@code 0x0.fffffffffffffp-1022}</td> 0N/A * <tr><td>{@code Double.MIN_VALUE}</td> 0N/A * <td>{@code 0x0.0000000000001p-1022}</td> 0N/A * @param d the {@code double} to be converted. 0N/A * @return a hex string representation of the argument. 0N/A * @author Joseph D. Darcy 0N/A * Modeled after the "a" conversion specifier in C99, section 0N/A * 7.19.6.1; however, the output of this method is more 0N/A * tightly specified. 0N/A // For infinity and NaN, use the decimal output. 0N/A // Initialized to maximum size of output. 0N/A // Isolate significand bits and OR in a high-order bit 0N/A // so that the string representation has a known 0N/A 0x1000000000000000L;
0N/A // Subnormal values have a 0 implicit bit; normal 0N/A // values have a 1 implicit bit. 0N/A // Isolate the low-order 13 digits of the hex 0N/A // representation. If all the digits are zero, 0N/A // replace with a single 0; otherwise, remove all 0N/A // If the value is subnormal, use the E_min exponent 0N/A // value for double; otherwise, extract and report d's 0N/A // exponent (the representation of a subnormal uses 0N/A * Returns a {@code Double} object holding the 0N/A * {@code double} value represented by the argument string 0N/A * <p>If {@code s} is {@code null}, then a 0N/A * {@code NullPointerException} is thrown. 0N/A * <p>Leading and trailing whitespace characters in {@code s} 0N/A * are ignored. Whitespace is removed as if by the {@link 0N/A * String#trim} method; that is, both ASCII space and control 0N/A * characters are removed. The rest of {@code s} should 0N/A * constitute a <i>FloatValue</i> as described by the lexical 0N/A * <dt><i>FloatValue:</i> 0N/A * <dd><i>Sign<sub>opt</sub></i> {@code NaN} 0N/A * <dd><i>Sign<sub>opt</sub></i> {@code Infinity} 0N/A * <dd><i>Sign<sub>opt</sub> FloatingPointLiteral</i> 0N/A * <dd><i>Sign<sub>opt</sub> HexFloatingPointLiteral</i> 0N/A * <dd><i>SignedInteger</i> 0N/A * <dt><i>HexFloatingPointLiteral</i>: 0N/A * <dd> <i>HexSignificand BinaryExponent FloatTypeSuffix<sub>opt</sub></i> 0N/A * <dt><i>HexSignificand:</i> 0N/A * <dd><i>HexNumeral</i> 0N/A * <dd><i>HexNumeral</i> {@code .} 0N/A * <dd>{@code 0x} <i>HexDigits<sub>opt</sub> 0N/A * </i>{@code .}<i> HexDigits</i> 0N/A * <dd>{@code 0X}<i> HexDigits<sub>opt</sub> 0N/A * </i>{@code .} <i>HexDigits</i> 0N/A * <dt><i>BinaryExponent:</i> 0N/A * <dd><i>BinaryExponentIndicator SignedInteger</i> 0N/A * <dt><i>BinaryExponentIndicator:</i> 0N/A * where <i>Sign</i>, <i>FloatingPointLiteral</i>, 0N/A * <i>HexNumeral</i>, <i>HexDigits</i>, <i>SignedInteger</i> and 0N/A * <i>FloatTypeSuffix</i> are as defined in the lexical structure 4008N/A * <cite>The Java™ Language Specification</cite>, 4008N/A * except that underscores are not accepted between digits. 4008N/A * If {@code s} does not have the form of 0N/A * a <i>FloatValue</i>, then a {@code NumberFormatException} 0N/A * is thrown. Otherwise, {@code s} is regarded as 0N/A * representing an exact decimal value in the usual 0N/A * "computerized scientific notation" or as an exact 0N/A * hexadecimal value; this exact numerical value is then 0N/A * conceptually converted to an "infinitely precise" 0N/A * binary value that is then rounded to type {@code double} 0N/A * by the usual round-to-nearest rule of IEEE 754 floating-point 0N/A * arithmetic, which includes preserving the sign of a zero 807N/A * Note that the round-to-nearest rule also implies overflow and 807N/A * underflow behaviour; if the exact value of {@code s} is large 807N/A * enough in magnitude (greater than or equal to ({@link 807N/A * #MAX_VALUE} + {@link Math#ulp(double) ulp(MAX_VALUE)}/2), 807N/A * rounding to {@code double} will result in an infinity and if the 807N/A * exact value of {@code s} is small enough in magnitude (less 807N/A * than or equal to {@link #MIN_VALUE}/2), rounding to float will 807N/A * Finally, after rounding a {@code Double} object representing 807N/A * this {@code double} value is returned. 0N/A * <p> To interpret localized string representations of a 0N/A * floating-point value, use subclasses of {@link 0N/A * java.text.NumberFormat}. 0N/A * <p>Note that trailing format specifiers, specifiers that 0N/A * determine the type of a floating-point literal 0N/A * ({@code 1.0f} is a {@code float} value; 0N/A * {@code 1.0d} is a {@code double} value), do 0N/A * <em>not</em> influence the results of this method. In other 0N/A * words, the numerical value of the input string is converted 0N/A * directly to the target floating-point type. The two-step 0N/A * sequence of conversions, string to {@code float} followed 0N/A * by {@code float} to {@code double}, is <em>not</em> 0N/A * equivalent to converting a string directly to 0N/A * {@code double}. For example, the {@code float} 0N/A * literal {@code 0.1f} is equal to the {@code double} 0N/A * value {@code 0.10000000149011612}; the {@code float} 0N/A * literal {@code 0.1f} represents a different numerical 0N/A * value than the {@code double} literal 0N/A * {@code 0.1}. (The numerical value 0.1 cannot be exactly 0N/A * represented in a binary floating-point number.) 0N/A * <p>To avoid calling this method on an invalid string and having 0N/A * a {@code NumberFormatException} be thrown, the regular 0N/A * expression below can be used to screen the input string: 0N/A * final String Digits = "(\\p{Digit}+)"; 0N/A * final String HexDigits = "(\\p{XDigit}+)"; 0N/A * // an exponent is 'e' or 'E' followed by an optionally 0N/A * // signed decimal integer. 0N/A * final String Exp = "[eE][+-]?"+Digits; 0N/A * final String fpRegex = 0N/A * ("[\\x00-\\x20]*"+ // Optional leading "whitespace" 0N/A * "[+-]?(" + // Optional sign character 0N/A * "NaN|" + // "NaN" string 0N/A * "Infinity|" + // "Infinity" string 0N/A * // A decimal floating-point string representing a finite positive 0N/A * // number without a leading sign has at most five basic pieces: 0N/A * // Digits . Digits ExponentPart FloatTypeSuffix 0N/A * // Since this method allows integer-only strings as input 0N/A * // in addition to strings of floating-point literals, the 0N/A * // two sub-patterns below are simplifications of the grammar 4008N/A * // productions from section 3.10.2 of 4008N/A * // <cite>The Java™ Language Specification</cite>. 0N/A * // Digits ._opt Digits_opt ExponentPart_opt FloatTypeSuffix_opt 0N/A * "((("+Digits+"(\\.)?("+Digits+"?)("+Exp+")?)|"+ 0N/A * // . Digits ExponentPart_opt FloatTypeSuffix_opt 0N/A * "(\\.("+Digits+")("+Exp+")?)|"+ 0N/A * // Hexadecimal strings 0N/A * // 0[xX] HexDigits ._opt BinaryExponent FloatTypeSuffix_opt 0N/A * "(0[xX]" + HexDigits + "(\\.)?)|" + 0N/A * // 0[xX] HexDigits_opt . HexDigits BinaryExponent FloatTypeSuffix_opt 0N/A * "(0[xX]" + HexDigits + "?(\\.)" + HexDigits + ")" + 0N/A * ")[pP][+-]?" + Digits + "))" + 0N/A * "[\\x00-\\x20]*");// Optional trailing "whitespace" 0N/A * if (Pattern.matches(fpRegex, myString)) 0N/A * Double.valueOf(myString); // Will not throw NumberFormatException 0N/A * // Perform suitable alternative action 0N/A * @param s the string to be parsed. 0N/A * @return a {@code Double} object holding the value 0N/A * represented by the {@code String} argument. 0N/A * @throws NumberFormatException if the string does not contain a 0N/A * Returns a {@code Double} instance representing the specified 0N/A * {@code double} value. 0N/A * If a new {@code Double} instance is not required, this method 0N/A * should generally be used in preference to the constructor 0N/A * {@link #Double(double)}, as this method is likely to yield 0N/A * significantly better space and time performance by caching 0N/A * frequently requested values. 0N/A * @param d a double value. 0N/A * @return a {@code Double} instance representing {@code d}. 0N/A * Returns a new {@code double} initialized to the value 0N/A * represented by the specified {@code String}, as performed 0N/A * by the {@code valueOf} method of class 0N/A * @param s the string to be parsed. 0N/A * @return the {@code double} value represented by the string 1419N/A * @throws NullPointerException if the string is null 0N/A * @throws NumberFormatException if the string does not contain 0N/A * a parsable {@code double}. 0N/A * @see java.lang.Double#valueOf(String) 0N/A * Returns {@code true} if the specified number is a 0N/A * Not-a-Number (NaN) value, {@code false} otherwise. 0N/A * @param v the value to be tested. 0N/A * @return {@code true} if the value of the argument is NaN; 0N/A * {@code false} otherwise. 0N/A * Returns {@code true} if the specified number is infinitely 0N/A * large in magnitude, {@code false} otherwise. 0N/A * @param v the value to be tested. 0N/A * @return {@code true} if the value of the argument is positive 0N/A * infinity or negative infinity; {@code false} otherwise. 0N/A * The value of the Double. 0N/A * Constructs a newly allocated {@code Double} object that 0N/A * represents the primitive {@code double} argument. 0N/A * @param value the value to be represented by the {@code Double}. 0N/A * Constructs a newly allocated {@code Double} object that 0N/A * represents the floating-point value of type {@code double} 0N/A * represented by the string. The string is converted to a 0N/A * {@code double} value as if by the {@code valueOf} method. 0N/A * @param s a string to be converted to a {@code Double}. 0N/A * @throws NumberFormatException if the string does not contain a 0N/A * @see java.lang.Double#valueOf(java.lang.String) 0N/A // REMIND: this is inefficient 0N/A * Returns {@code true} if this {@code Double} value is 0N/A * a Not-a-Number (NaN), {@code false} otherwise. 0N/A * @return {@code true} if the value represented by this object is 0N/A * NaN; {@code false} otherwise. 0N/A * Returns {@code true} if this {@code Double} value is 0N/A * infinitely large in magnitude, {@code false} otherwise. 0N/A * @return {@code true} if the value represented by this object is 0N/A * positive infinity or negative infinity; 0N/A * {@code false} otherwise. 0N/A * Returns a string representation of this {@code Double} object. 0N/A * The primitive {@code double} value represented by this 0N/A * object is converted to a string exactly as if by the method 0N/A * {@code toString} of one argument. 0N/A * @return a {@code String} representation of this object. 0N/A * @see java.lang.Double#toString(double) 0N/A * Returns the value of this {@code Double} as a {@code byte} (by 0N/A * casting to a {@code byte}). 0N/A * @return the {@code double} value represented by this object 0N/A * converted to type {@code byte} 0N/A * Returns the value of this {@code Double} as a 0N/A * {@code short} (by casting to a {@code short}). 0N/A * @return the {@code double} value represented by this object 0N/A * converted to type {@code short} 0N/A * Returns the value of this {@code Double} as an 0N/A * {@code int} (by casting to type {@code int}). 0N/A * @return the {@code double} value represented by this object 0N/A * converted to type {@code int} 0N/A * Returns the value of this {@code Double} as a 0N/A * {@code long} (by casting to type {@code long}). 0N/A * @return the {@code double} value represented by this object 0N/A * converted to type {@code long} 0N/A * Returns the {@code float} value of this 0N/A * {@code Double} object. 0N/A * @return the {@code double} value represented by this object 0N/A * converted to type {@code float} 0N/A * Returns the {@code double} value of this 0N/A * {@code Double} object. 0N/A * @return the {@code double} value represented by this object 0N/A * Returns a hash code for this {@code Double} object. The 0N/A * result is the exclusive OR of the two halves of the 0N/A * {@code long} integer bit representation, exactly as 0N/A * produced by the method {@link #doubleToLongBits(double)}, of 0N/A * the primitive {@code double} value represented by this 0N/A * {@code Double} object. That is, the hash code is the value 0N/A * of the expression: 0N/A * {@code (int)(v^(v>>>32))} 0N/A * where {@code v} is defined by: 0N/A * {@code long v = Double.doubleToLongBits(this.doubleValue());} 0N/A * @return a {@code hash code} value for this object. 0N/A * Compares this object against the specified object. The result 0N/A * is {@code true} if and only if the argument is not 0N/A * {@code null} and is a {@code Double} object that 0N/A * represents a {@code double} that has the same value as the 0N/A * {@code double} represented by this object. For this 0N/A * purpose, two {@code double} values are considered to be 0N/A * the same if and only if the method {@link 0N/A * #doubleToLongBits(double)} returns the identical 0N/A * {@code long} value when applied to each. 0N/A * <p>Note that in most cases, for two instances of class 0N/A * {@code Double}, {@code d1} and {@code d2}, the 0N/A * value of {@code d1.equals(d2)} is {@code true} if and 0N/A * {@code d1.doubleValue() == d2.doubleValue()} 0N/A * <p>also has the value {@code true}. However, there are two 0N/A * <li>If {@code d1} and {@code d2} both represent 0N/A * {@code Double.NaN}, then the {@code equals} method 0N/A * returns {@code true}, even though 0N/A * {@code Double.NaN==Double.NaN} has the value 0N/A * <li>If {@code d1} represents {@code +0.0} while 0N/A * {@code d2} represents {@code -0.0}, or vice versa, 0N/A * the {@code equal} test has the value {@code false}, 0N/A * even though {@code +0.0==-0.0} has the value {@code true}. 0N/A * This definition allows hash tables to operate properly. 0N/A * @param obj the object to compare with. 0N/A * @return {@code true} if the objects are the same; 0N/A * {@code false} otherwise. 0N/A * @see java.lang.Double#doubleToLongBits(double) 0N/A * Returns a representation of the specified floating-point value 0N/A * according to the IEEE 754 floating-point "double 0N/A * format" bit layout. 0N/A * <p>Bit 63 (the bit that is selected by the mask 0N/A * {@code 0x8000000000000000L}) represents the sign of the 0N/A * floating-point number. Bits 0N/A * 62-52 (the bits that are selected by the mask 0N/A * {@code 0x7ff0000000000000L}) represent the exponent. Bits 51-0 0N/A * (the bits that are selected by the mask 0N/A * {@code 0x000fffffffffffffL}) represent the significand 0N/A * (sometimes called the mantissa) of the floating-point number. 0N/A * <p>If the argument is positive infinity, the result is 0N/A * {@code 0x7ff0000000000000L}. 0N/A * <p>If the argument is negative infinity, the result is 0N/A * {@code 0xfff0000000000000L}. 0N/A * <p>If the argument is NaN, the result is 0N/A * {@code 0x7ff8000000000000L}. 0N/A * <p>In all cases, the result is a {@code long} integer that, when 0N/A * given to the {@link #longBitsToDouble(long)} method, will produce a 0N/A * floating-point value the same as the argument to 0N/A * {@code doubleToLongBits} (except all NaN values are 0N/A * collapsed to a single "canonical" NaN value). 0N/A * @param value a {@code double} precision floating-point number. 0N/A * @return the bits that represent the floating-point number. 0N/A // Check for NaN based on values of bit fields, maximum 0N/A // exponent and nonzero significand. 0N/A * Returns a representation of the specified floating-point value 0N/A * according to the IEEE 754 floating-point "double 0N/A * format" bit layout, preserving Not-a-Number (NaN) values. 0N/A * <p>Bit 63 (the bit that is selected by the mask 0N/A * {@code 0x8000000000000000L}) represents the sign of the 0N/A * floating-point number. Bits 0N/A * 62-52 (the bits that are selected by the mask 0N/A * {@code 0x7ff0000000000000L}) represent the exponent. Bits 51-0 0N/A * (the bits that are selected by the mask 0N/A * {@code 0x000fffffffffffffL}) represent the significand 0N/A * (sometimes called the mantissa) of the floating-point number. 0N/A * <p>If the argument is positive infinity, the result is 0N/A * {@code 0x7ff0000000000000L}. 0N/A * <p>If the argument is negative infinity, the result is 0N/A * {@code 0xfff0000000000000L}. 0N/A * <p>If the argument is NaN, the result is the {@code long} 0N/A * integer representing the actual NaN value. Unlike the 0N/A * {@code doubleToLongBits} method, 0N/A * {@code doubleToRawLongBits} does not collapse all the bit 0N/A * patterns encoding a NaN to a single "canonical" NaN 0N/A * <p>In all cases, the result is a {@code long} integer that, 0N/A * when given to the {@link #longBitsToDouble(long)} method, will 0N/A * produce a floating-point value the same as the argument to 0N/A * {@code doubleToRawLongBits}. 0N/A * @param value a {@code double} precision floating-point number. 0N/A * @return the bits that represent the floating-point number. 0N/A * Returns the {@code double} value corresponding to a given 0N/A * bit representation. 0N/A * The argument is considered to be a representation of a 0N/A * floating-point value according to the IEEE 754 floating-point 0N/A * "double format" bit layout. 0N/A * <p>If the argument is {@code 0x7ff0000000000000L}, the result 0N/A * is positive infinity. 0N/A * <p>If the argument is {@code 0xfff0000000000000L}, the result 0N/A * is negative infinity. 0N/A * <p>If the argument is any value in the range 0N/A * {@code 0x7ff0000000000001L} through 0N/A * {@code 0x7fffffffffffffffL} or in the range 0N/A * {@code 0xfff0000000000001L} through 0N/A * {@code 0xffffffffffffffffL}, the result is a NaN. No IEEE 0N/A * 754 floating-point operation provided by Java can distinguish 0N/A * between two NaN values of the same type with different bit 0N/A * patterns. Distinct values of NaN are only distinguishable by 0N/A * use of the {@code Double.doubleToRawLongBits} method. 0N/A * <p>In all other cases, let <i>s</i>, <i>e</i>, and <i>m</i> be three 0N/A * values that can be computed from the argument: 0N/A * int s = ((bits >> 63) == 0) ? 1 : -1; 0N/A * int e = (int)((bits >> 52) & 0x7ffL); 0N/A * long m = (e == 0) ? 0N/A * (bits & 0xfffffffffffffL) << 1 : 0N/A * (bits & 0xfffffffffffffL) | 0x10000000000000L; 0N/A * </pre></blockquote> 0N/A * Then the floating-point result equals the value of the mathematical 0N/A * expression <i>s</i>·<i>m</i>·2<sup><i>e</i>-1075</sup>. 0N/A * <p>Note that this method may not be able to return a 0N/A * {@code double} NaN with exactly same bit pattern as the 0N/A * {@code long} argument. IEEE 754 distinguishes between two 0N/A * kinds of NaNs, quiet NaNs and <i>signaling NaNs</i>. The 0N/A * differences between the two kinds of NaN are generally not 0N/A * visible in Java. Arithmetic operations on signaling NaNs turn 0N/A * them into quiet NaNs with a different, but often similar, bit 0N/A * pattern. However, on some processors merely copying a 0N/A * signaling NaN also performs that conversion. In particular, 0N/A * copying a signaling NaN to return it to the calling method 0N/A * may perform this conversion. So {@code longBitsToDouble} 0N/A * may not be able to return a {@code double} with a 0N/A * signaling NaN bit pattern. Consequently, for some 0N/A * {@code long} values, 0N/A * {@code doubleToRawLongBits(longBitsToDouble(start))} may 0N/A * <i>not</i> equal {@code start}. Moreover, which 0N/A * particular bit patterns represent signaling NaNs is platform 0N/A * dependent; although all NaN bit patterns, quiet or signaling, 0N/A * must be in the NaN range identified above. 0N/A * @param bits any {@code long} integer. 0N/A * @return the {@code double} floating-point value with the same 0N/A * Compares two {@code Double} objects numerically. There 0N/A * are two ways in which comparisons performed by this method 0N/A * differ from those performed by the Java language numerical 0N/A * comparison operators ({@code <, <=, ==, >=, >}) 0N/A * when applied to primitive {@code double} values: 0N/A * {@code Double.NaN} is considered by this method 0N/A * to be equal to itself and greater than all other 0N/A * {@code double} values (including 0N/A * {@code Double.POSITIVE_INFINITY}). 0N/A * {@code 0.0d} is considered by this method to be greater 0N/A * than {@code -0.0d}. 0N/A * This ensures that the <i>natural ordering</i> of 0N/A * {@code Double} objects imposed by this method is <i>consistent 0N/A * @param anotherDouble the {@code Double} to be compared. 0N/A * @return the value {@code 0} if {@code anotherDouble} is 0N/A * numerically equal to this {@code Double}; a value 0N/A * less than {@code 0} if this {@code Double} 0N/A * is numerically less than {@code anotherDouble}; 0N/A * and a value greater than {@code 0} if this 0N/A * {@code Double} is numerically greater than 0N/A * {@code anotherDouble}. 0N/A * Compares the two specified {@code double} values. The sign 0N/A * of the integer value returned is the same as that of the 0N/A * integer that would be returned by the call: 0N/A * new Double(d1).compareTo(new Double(d2)) 0N/A * @param d1 the first {@code double} to compare 0N/A * @param d2 the second {@code double} to compare 0N/A * @return the value {@code 0} if {@code d1} is 0N/A * numerically equal to {@code d2}; a value less than 0N/A * {@code 0} if {@code d1} is numerically less than 0N/A * {@code d2}; and a value greater than {@code 0} 0N/A * if {@code d1} is numerically greater than 0N/A return -
1;
// Neither val is NaN, thisVal is smaller 0N/A return 1;
// Neither val is NaN, thisVal is larger 3202N/A // Cannot use doubleToRawLongBits because of possibility of NaNs. 0N/A 1));
// (0.0, -0.0) or (NaN, !NaN) 0N/A /** use serialVersionUID from JDK 1.0.2 for interoperability */