home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Programming Languages Suite
/
ProgLangD.iso
/
VCAFE.3.0A
/
Main.bin
/
DigitList.java
< prev
next >
Wrap
Text File
|
1998-09-22
|
21KB
|
537 lines
/*
* @(#)DigitList.java 1.13 98/01/12
*
* (C) Copyright Taligent, Inc. 1996 - All Rights Reserved
* (C) Copyright IBM Corp. 1996 - All Rights Reserved
*
* Portions copyright (c) 1996 Sun Microsystems, Inc. All Rights Reserved.
*
* The original version of this source code and documentation is copyrighted
* and owned by Taligent, Inc., a wholly-owned subsidiary of IBM. These
* materials are provided under terms of a License Agreement between Taligent
* and Sun. This technology is protected by multiple US and International
* patents. This notice and attribution to Taligent may not be removed.
* Taligent is a registered trademark of Taligent, Inc.
*
* Permission to use, copy, modify, and distribute this software
* and its documentation for NON-COMMERCIAL purposes and without
* fee is hereby granted provided that this copyright notice
* appears in all copies. Please refer to the file "copyright.html"
* for further important copyright and licensing information.
*
* SUN MAKES NO REPRESENTATIONS OR WARRANTIES ABOUT THE SUITABILITY OF
* THE SOFTWARE, EITHER EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED
* TO THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
* PARTICULAR PURPOSE, OR NON-INFRINGEMENT. SUN SHALL NOT BE LIABLE FOR
* ANY DAMAGES SUFFERED BY LICENSEE AS A RESULT OF USING, MODIFYING OR
* DISTRIBUTING THIS SOFTWARE OR ITS DERIVATIVES.
*
*/
package java.text;
/**
* Digit List. Private to DecimalFormat.
* Handles the transcoding
* between numeric values and strings of characters. Only handles
* non-negative numbers. The division of labor between DigitList and
* DecimalFormat is that DigitList handles the radix 10 representation
* issues; DecimalFormat handles the locale-specific issues such as
* positive/negative, grouping, decimal point, currency, and so on.
*
* A DigitList is really a representation of a floating point value.
* It may be an integer value; we assume that a double has sufficient
* precision to represent all digits of a long.
*
* The DigitList representation consists of a string of characters,
* which are the digits radix 10, from '0' to '9'. It also has a radix
* 10 exponent associated with it. The value represented by a DigitList
* object can be computed by mulitplying the fraction f, where 0 <= f < 1,
* derived by placing all the digits of the list to the right of the
* decimal point, by 10^exponent.
*
* @see Locale
* @see Format
* @see NumberFormat
* @see DecimalFormat
* @see ChoiceFormat
* @see MessageFormat
* @version 1.13 01/12/98
* @author Mark Davis, Alan Liu
*/
final class DigitList implements Cloneable {
/**
* The maximum number of significant digits in an IEEE 754 double, that
* is, in a Java double. This must not be increased, or garbage digits
* will be generated, and should not be decreased, or accuracy will be lost.
*/
public static final int MAX_COUNT = 19; // == Long.toString(Long.MAX_VALUE).length()
public static final int DBL_DIG = 17;
/**
* These data members are intentionally public and can be set directly.
*
* The value represented is given by placing the decimal point before
* digits[decimalAt]. If decimalAt is < 0, then leading zeros between
* the decimal point and the first nonzero digit are implied. If decimalAt
* is > count, then trailing zeros between the digits[count-1] and the
* decimal point are implied.
*
* Equivalently, the represented value is given by f * 10^decimalAt. Here
* f is a value 0.1 <= f < 1 arrived at by placing the digits in Digits to
* the right of the decimal.
*
* DigitList is normalized, so if it is non-zero, figits[0] is non-zero. We
* don't allow denormalized numbers because our exponent is effectively of
* unlimited magnitude. The count value contains the number of significant
* digits present in digits[].
*
* Zero is represented by any DigitList with count == 0 or with each digits[i]
* for all i <= count == '0'.
*/
public int decimalAt = 0;
public int count = 0;
public byte[] digits = new byte[MAX_COUNT];
/**
* Return true if the represented number is zero.
*/
boolean isZero()
{
for (int i=0; i<count; ++i) if (digits[i] != '0') return false;
return true;
}
/**
* Clears out the digits.
* Use before appending them.
* Typically, you set a series of digits with append, then at the point
* you hit the decimal point, you set myDigitList.decimalAt = myDigitList.count;
* then go on appending digits.
*/
public void clear () {
decimalAt = 0;
count = 0;
}
/**
* Appends digits to the list. Ignores all digits over MAX_COUNT,
* since they are not significant for either longs or doubles.
*/
public void append (int digit) {
if (count < MAX_COUNT)
digits[count++] = (byte) digit;
}
/**
* Utility routine to get the value of the digit list
* If (count == 0) this throws a NumberFormatException, which
* mimics Long.parseLong().
*/
public final double getDouble() {
if (count == 0) return 0.0;
StringBuffer temp = new StringBuffer(count);
temp.append('.');
for (int i = 0; i < count; ++i) temp.append((char)(digits[i]));
temp.append('E');
temp.append(Integer.toString(decimalAt));
return Double.valueOf(temp.toString()).doubleValue();
// long value = Long.parseLong(temp.toString());
// return (value * Math.pow(10, decimalAt - count));
}
/**
* Utility routine to get the value of the digit list.
* If (count == 0) this returns 0, unlike Long.parseLong().
*/
public final long getLong() {
// for now, simple implementation; later, do proper IEEE native stuff
if (count == 0) return 0;
// We have to check for this, because this is the one NEGATIVE value
// we represent. If we tried to just pass the digits off to parseLong,
// we'd get a parse failure.
if (isLongMIN_VALUE()) return Long.MIN_VALUE;
StringBuffer temp = new StringBuffer(count);
for (int i = 0; i < decimalAt; ++i)
{
temp.append((i < count) ? (char)(digits[i]) : '0');
}
return Long.parseLong(temp.toString());
}
/**
* Return true if the number represented by this object can fit into
* a long.
*/
boolean fitsIntoLong(boolean isPositive)
{
// Figure out if the result will fit in a long. We have to
// first look for nonzero digits after the decimal point;
// then check the size. If the digit count is 18 or less, then
// the value can definitely be represented as a long. If it is 19
// then it may be too large.
// Trim trailing zeros. This does not change the represented value.
while (count > 0 && digits[count - 1] == (byte)'0') --count;
if (count == 0) return true;
if (decimalAt < count || decimalAt > MAX_COUNT) return false;
if (decimalAt < MAX_COUNT) return true;
// At this point we have decimalAt == count, and count == MAX_COUNT.
// The number will overflow if it is larger than 9223372036854775807
// or smaller than -9223372036854775808.
for (int i=0; i<count; ++i)
{
byte dig = digits[i], max = LONG_MIN_REP[i];
if (dig > max) return false;
if (dig < max) return true;
}
// At this point the first count digits match. If decimalAt is less
// than count, then the remaining digits are zero, and we return true.
if (count < decimalAt) return true;
// Now we have a representation of Long.MIN_VALUE, without the leading
// negative sign. If this represents a positive value, then it does
// not fit; otherwise it fits.
return !isPositive;
}
private static final boolean DEBUG = false;
/**
* Set the digit list to a representation of the given double value.
* This method supports fixed-point notation.
* @param source Value to be converted; must not be Inf, -Inf, Nan,
* or a value <= 0.
* @param maximumFractionDigits The most fractional digits which should
* be converted.
*/
public final void set(double source, int maximumFractionDigits)
{
set(source, maximumFractionDigits, true);
}
/**
* Set the digit list to a representation of the given double value.
* This method supports both fixed-point and exponential notation.
* @param source Value to be converted; must not be Inf, -Inf, Nan,
* or a value <= 0.
* @param maximumDigits The most fractional or total digits which should
* be converted.
* @param fixedPoint If true, then maximumDigits is the maximum
* fractional digits to be converted. If false, total digits.
*/
final void set(double source, int maximumDigits, boolean fixedPoint)
{
// Generate a representation of the form DDDDD, DDDDD.DDDDD, or
// DDDDDE+/-DDDDD.
String rep = Double.toString(source);
decimalAt = -1;
count = 0;
int exponent = 0;
// Number of zeros between decimal point and first non-zero digit after
// decimal point, for numbers < 1.
int leadingZerosAfterDecimal = 0;
boolean nonZeroDigitSeen = false;
for (int i=0; i < rep.length(); ++i)
{
char c = rep.charAt(i);
if (c == '.')
{
decimalAt = count;
}
else if (c == 'e' || c == 'E')
{
exponent = Integer.valueOf(rep.substring(i+1)).intValue();
break;
}
else if (count < MAX_COUNT)
{
if (!nonZeroDigitSeen)
{
nonZeroDigitSeen = (c != '0');
if (!nonZeroDigitSeen && decimalAt != -1) ++leadingZerosAfterDecimal;
}
if (nonZeroDigitSeen) digits[count++] = (byte)c;
}
}
if (decimalAt == -1) decimalAt = count;
decimalAt += exponent - leadingZerosAfterDecimal;
if (fixedPoint)
{
// The negative of the exponent represents the number of leading
// zeros between the decimal and the first non-zero digit, for
// a value < 0.1 (e.g., for 0.00123, -decimalAt == 2). If this
// is more than the maximum fraction digits, then we have an underflow
// for the printed representation. We recognize this here and set
// the DigitList representation to zero in this situation.
if (-decimalAt >= maximumDigits) count = 0;
}
// Eliminate trailing zeros.
while (count > 1 && digits[count - 1] == '0')
--count;
if (DEBUG)
{
System.out.print("Before rounding 0.");
for (int i=0; i<count; ++i) System.out.print((char)digits[i]);
System.out.println("x10^" + decimalAt);
}
// Eliminate digits beyond maximum digits to be displayed.
// Round up if appropriate.
round(fixedPoint ? (maximumDigits + decimalAt) : maximumDigits);
if (DEBUG)
{
System.out.print("After rounding 0.");
for (int i=0; i<count; ++i) System.out.print((char)digits[i]);
System.out.println("x10^" + decimalAt);
}
// The following method also works, and does not rely on the specific
// format generated by Double.toString(). However, it introduces significant
// errors in the least-significant digits, which cause round-trip parse and
// format operations to fail. We retain this code for future reference;
// the compiler will ignore it.
if (false)
{
// Find the exponent for this value. Our convention is 0.mmmm * 10^decimalAt,
// so we need to add one.
decimalAt = log10(source) + 1;
// Compute the number of digits to generate based on the maximum fraction
// digits and the exponent. For example, if the exponent is -95 and the
// maximum fraction digits is 100, then we'll have 95 leading zeros and only
// 5 significant digits.
count = maximumDigits + decimalAt;
if (count > DBL_DIG) count = DBL_DIG;
if (count < 0) count = 0;
if (count == 0) return; // Return if we've underflowed to zero
// Put the mantissa into a long. We create a mantissa value in the
// range 10^n-1 <= mantissa < 10^n, where n is the desired number of
// digits. If this is a small number << 1, decimalAt may be negative,
// indicating leading zeros between the decimal point an digits[0]. A
// decimalAt value of 0 indicates that the decimal point is before
// digits[0].
//System.out.println("d = " + source + " log = " + (Math.log(source) / LOG10));
//System.out.println("d == 0.1 " + (source == 0.1));
long mantissa = Math.round(source * Math.pow(10, count - decimalAt));
String longRep = Long.toString(mantissa);
// At this point we have a representation of exactly maxDecimalCount
// characters.
// FOLLOWING LINE FOR DEBUGGING ONLY. THIS catches problems with log10 computation.
if (longRep.length() != count)
throw new Error("Rep=" + longRep + " rep.length=" + longRep.length() +
" exp.len=" + count + " " +
"val=" + source + " mant=" + mantissa +
" decimalAt=" + decimalAt);
// Eliminate trailing zeros.
while (count > 1 && longRep.charAt(count - 1) == '0')
--count;
// Copy digits over
for (int i=0; i<count; ++i)
digits[i] = (byte)longRep.charAt(i);
}
}
/**
* Round the representation to the given number of digits.
* @param maximumDigits The maximum number of digits to be shown.
* Upon return, count will be less than or equal to maximumDigits.
*/
private final void round(int maximumDigits)
{
// Eliminate digits beyond maximum digits to be displayed.
// Round up if appropriate.
if (maximumDigits >= 0 && maximumDigits < count)
{
// Check for round to the nearest even. HShih
if (digits[maximumDigits] == '5' && digits[maximumDigits-1] != '9' &&
(maximumDigits+1 >= count || digits[maximumDigits+1] == '0')) {
if (digits[maximumDigits-1] % 2 != 0)
++digits[maximumDigits-1];
} else if (digits[maximumDigits] >= '5')
{
// Rounding up involved incrementing digits from LSD to MSD.
// In most cases this is simple, but in a worst case situation
// (9999..99) we have to adjust the decimalAt value.
for (;;)
{
--maximumDigits;
if (maximumDigits < 0)
{
// We have all 9's, so we increment to a single digit
// of one and adjust the exponent.
digits[0] = '1';
++decimalAt;
maximumDigits = 0; // Adjust the count
break;
}
++digits[maximumDigits];
if (digits[maximumDigits] <= '9') break;
// digits[maximumDigits] = '0'; // Unnecessary since we'll truncate this
}
++maximumDigits; // Increment for use as count
}
count = maximumDigits;
}
}
/**
* Utility routine to set the value of the digit list from a long
*/
public final void set(long source)
{
set(source, 0);
}
/**
* Set the digit list to a representation of the given long value.
* @param source Value to be converted; must be >= 0 or ==
* Long.MIN_VALUE.
* @param maximumDigits The most digits which should be converted.
* If maximumDigits is lower than the number of significant digits
* in source, the representation will be rounded. Ignored if <= 0.
*/
public final void set(long source, int maximumDigits)
{
// for now, simple implementation; later, do proper IEEE stuff
// String stringDigits = Long.toString(source);
String stringDigits = Long.toString(source);
// This method does not expect a negative number. However,
// "source" can be a Long.MIN_VALUE (-9223372036854775808),
// if the number being formatted is a Long.MIN_VALUE. In that
// case, it will be formatted as -Long.MIN_VALUE, a number
// which is outside the legal range of a long, but which can
// be represented by DigitList.
if (stringDigits.charAt(0) == '-') stringDigits = stringDigits.substring(1);
count = decimalAt = stringDigits.length();
// Don't copy trailing zeros
while (count > 1 && stringDigits.charAt(count - 1) == '0') --count;
for (int i = 0; i < count; ++i)
digits[i] = (byte) stringDigits.charAt(i);
if (maximumDigits > 0) round(maximumDigits);
}
/**
* equality test between two digit lists.
*/
public boolean equals(Object obj) {
if (this == obj) // quick check
return true;
if (!(obj instanceof DigitList)) // (1) same object?
return false;
DigitList other = (DigitList) obj;
if (count != other.count ||
decimalAt != other.decimalAt)
return false;
for (int i = 0; i < count; i++)
if (digits[i] != other.digits[i])
return false;
return true;
}
/**
* Generates the hash code for the digit list.
*/
public int hashCode() {
int hashcode = decimalAt;
for (int i = 0; i < count; i++)
hashcode = hashcode * 37 + digits[i];
return hashcode;
}
/**
* Returns true if this DigitList represents Long.MIN_VALUE;
* false, otherwise. This is required so that getLong() works.
*/
private boolean isLongMIN_VALUE()
{
if (decimalAt != count || count != MAX_COUNT)
return false;
for (int i = 0; i < count; ++i)
{
if (digits[i] != LONG_MIN_REP[i]) return false;
}
return true;
}
private static byte[] LONG_MIN_REP;
static
{
// Store the representation of LONG_MIN without the leading '-'
String s = Long.toString(Long.MIN_VALUE);
LONG_MIN_REP = new byte[MAX_COUNT];
for (int i=0; i < MAX_COUNT; ++i)
{
LONG_MIN_REP[i] = (byte)s.charAt(i + 1);
}
}
/**
* Return the floor of the log base 10 of a given double.
* This method compensates for inaccuracies which arise naturally when
* computing logs, and always give the correct value. The parameter
* must be positive and finite.
*/
private static final int log10(double d)
{
// The reason this routine is needed is that simply taking the
// log and dividing by log10 yields a result which may be off
// by 1 due to rounding errors. For example, the naive log10
// of 1.0e300 taken this way is 299, rather than 300.
double log10 = Math.log(d) / LOG10;
int ilog10 = (int)Math.floor(log10);
// Positive logs could be too small, e.g. 0.99 instead of 1.0
if (log10 > 0 && d >= Math.pow(10, ilog10 + 1))
{
++ilog10;
}
// Negative logs could be too big, e.g. -0.99 instead of -1.0
else if (log10 < 0 && d < Math.pow(10, ilog10))
{
--ilog10;
}
return ilog10;
}
private static final double LOG10 = Math.log(10.0);
public String toString()
{
if (isZero()) return "0";
StringBuffer buf = new StringBuffer("0.");
for (int i=0; i<count; ++i) buf.append((char)digits[i]);
buf.append("x10^");
buf.append(decimalAt);
return buf.toString();
}
}