1 /*
   2  * Copyright (c) 1999, 2013, Oracle and/or its affiliates. All rights reserved.
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   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.
  10  *
  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).
  16  *
  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.
  20  *
  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
  23  * questions.
  24  */
  25 
  26 package java.math;
  27 
  28 /**
  29  * A class used to represent multiprecision integers that makes efficient
  30  * use of allocated space by allowing a number to occupy only part of
  31  * an array so that the arrays do not have to be reallocated as often.
  32  * When performing an operation with many iterations the array used to
  33  * hold a number is only reallocated when necessary and does not have to
  34  * be the same size as the number it represents. A mutable number allows
  35  * calculations to occur on the same number without having to create
  36  * a new number for every step of the calculation as occurs with
  37  * BigIntegers.
  38  *
  39  * @see     BigInteger
  40  * @author  Michael McCloskey
  41  * @author  Timothy Buktu
  42  * @since   1.3
  43  */
  44 
  45 import static java.math.BigDecimal.INFLATED;
  46 import static java.math.BigInteger.LONG_MASK;
  47 import java.util.Arrays;
  48 
  49 class MutableBigInteger {
  50     /**
  51      * Holds the magnitude of this MutableBigInteger in big endian order.
  52      * The magnitude may start at an offset into the value array, and it may
  53      * end before the length of the value array.
  54      */
  55     int[] value;
  56 
  57     /**
  58      * The number of ints of the value array that are currently used
  59      * to hold the magnitude of this MutableBigInteger. The magnitude starts
  60      * at an offset and offset + intLen may be less than value.length.
  61      */
  62     int intLen;
  63 
  64     /**
  65      * The offset into the value array where the magnitude of this
  66      * MutableBigInteger begins.
  67      */
  68     int offset = 0;
  69 
  70     // Constants
  71     /**
  72      * MutableBigInteger with one element value array with the value 1. Used by
  73      * BigDecimal divideAndRound to increment the quotient. Use this constant
  74      * only when the method is not going to modify this object.
  75      */
  76     static final MutableBigInteger ONE = new MutableBigInteger(1);
  77 
  78     /**
  79      * The minimum {@code intLen} for cancelling powers of two before
  80      * dividing.
  81      * If the number of ints is less than this threshold,
  82      * {@code divideKnuth} does not eliminate common powers of two from
  83      * the dividend and divisor.
  84      */
  85     static final int KNUTH_POW2_THRESH_LEN = 6;
  86 
  87     /**
  88      * The minimum number of trailing zero ints for cancelling powers of two
  89      * before dividing.
  90      * If the dividend and divisor don't share at least this many zero ints
  91      * at the end, {@code divideKnuth} does not eliminate common powers
  92      * of two from the dividend and divisor.
  93      */
  94     static final int KNUTH_POW2_THRESH_ZEROS = 3;
  95 
  96     // Constructors
  97 
  98     /**
  99      * The default constructor. An empty MutableBigInteger is created with
 100      * a one word capacity.
 101      */
 102     MutableBigInteger() {
 103         value = new int[1];
 104         intLen = 0;
 105     }
 106 
 107     /**
 108      * Construct a new MutableBigInteger with a magnitude specified by
 109      * the int val.
 110      */
 111     MutableBigInteger(int val) {
 112         value = new int[1];
 113         intLen = 1;
 114         value[0] = val;
 115     }
 116 
 117     /**
 118      * Construct a new MutableBigInteger with the specified value array
 119      * up to the length of the array supplied.
 120      */
 121     MutableBigInteger(int[] val) {
 122         value = val;
 123         intLen = val.length;
 124     }
 125 
 126     /**
 127      * Construct a new MutableBigInteger with a magnitude equal to the
 128      * specified BigInteger.
 129      */
 130     MutableBigInteger(BigInteger b) {
 131         intLen = b.mag.length;
 132         value = Arrays.copyOf(b.mag, intLen);
 133     }
 134 
 135     /**
 136      * Construct a new MutableBigInteger with a magnitude equal to the
 137      * specified MutableBigInteger.
 138      */
 139     MutableBigInteger(MutableBigInteger val) {
 140         intLen = val.intLen;
 141         value = Arrays.copyOfRange(val.value, val.offset, val.offset + intLen);
 142     }
 143 
 144     /**
 145      * Makes this number an {@code n}-int number all of whose bits are ones.
 146      * Used by Burnikel-Ziegler division.
 147      * @param n number of ints in the {@code value} array
 148      * @return a number equal to {@code ((1<<(32*n)))-1}
 149      */
 150     private void ones(int n) {
 151         if (n > value.length)
 152             value = new int[n];
 153         Arrays.fill(value, -1);
 154         offset = 0;
 155         intLen = n;
 156     }
 157 
 158     /**
 159      * Internal helper method to return the magnitude array. The caller is not
 160      * supposed to modify the returned array.
 161      */
 162     private int[] getMagnitudeArray() {
 163         if (offset > 0 || value.length != intLen)
 164             return Arrays.copyOfRange(value, offset, offset + intLen);
 165         return value;
 166     }
 167 
 168     /**
 169      * Convert this MutableBigInteger to a long value. The caller has to make
 170      * sure this MutableBigInteger can be fit into long.
 171      */
 172     private long toLong() {
 173         assert (intLen <= 2) : "this MutableBigInteger exceeds the range of long";
 174         if (intLen == 0)
 175             return 0;
 176         long d = value[offset] & LONG_MASK;
 177         return (intLen == 2) ? d << 32 | (value[offset + 1] & LONG_MASK) : d;
 178     }
 179 
 180     /**
 181      * Convert this MutableBigInteger to a BigInteger object.
 182      */
 183     BigInteger toBigInteger(int sign) {
 184         if (intLen == 0 || sign == 0)
 185             return BigInteger.ZERO;
 186         return new BigInteger(getMagnitudeArray(), sign);
 187     }
 188 
 189     /**
 190      * Converts this number to a nonnegative {@code BigInteger}.
 191      */
 192     BigInteger toBigInteger() {
 193         normalize();
 194         return toBigInteger(isZero() ? 0 : 1);
 195     }
 196 
 197     /**
 198      * Convert this MutableBigInteger to BigDecimal object with the specified sign
 199      * and scale.
 200      */
 201     BigDecimal toBigDecimal(int sign, int scale) {
 202         if (intLen == 0 || sign == 0)
 203             return BigDecimal.zeroValueOf(scale);
 204         int[] mag = getMagnitudeArray();
 205         int len = mag.length;
 206         int d = mag[0];
 207         // If this MutableBigInteger can't be fit into long, we need to
 208         // make a BigInteger object for the resultant BigDecimal object.
 209         if (len > 2 || (d < 0 && len == 2))
 210             return new BigDecimal(new BigInteger(mag, sign), INFLATED, scale, 0);
 211         long v = (len == 2) ?
 212             ((mag[1] & LONG_MASK) | (d & LONG_MASK) << 32) :
 213             d & LONG_MASK;
 214         return BigDecimal.valueOf(sign == -1 ? -v : v, scale);
 215     }
 216 
 217     /**
 218      * This is for internal use in converting from a MutableBigInteger
 219      * object into a long value given a specified sign.
 220      * returns INFLATED if value is not fit into long
 221      */
 222     long toCompactValue(int sign) {
 223         if (intLen == 0 || sign == 0)
 224             return 0L;
 225         int[] mag = getMagnitudeArray();
 226         int len = mag.length;
 227         int d = mag[0];
 228         // If this MutableBigInteger can not be fitted into long, we need to
 229         // make a BigInteger object for the resultant BigDecimal object.
 230         if (len > 2 || (d < 0 && len == 2))
 231             return INFLATED;
 232         long v = (len == 2) ?
 233             ((mag[1] & LONG_MASK) | (d & LONG_MASK) << 32) :
 234             d & LONG_MASK;
 235         return sign == -1 ? -v : v;
 236     }
 237 
 238     /**
 239      * Clear out a MutableBigInteger for reuse.
 240      */
 241     void clear() {
 242         offset = intLen = 0;
 243         for (int index=0, n=value.length; index < n; index++)
 244             value[index] = 0;
 245     }
 246 
 247     /**
 248      * Set a MutableBigInteger to zero, removing its offset.
 249      */
 250     void reset() {
 251         offset = intLen = 0;
 252     }
 253 
 254     /**
 255      * Compare the magnitude of two MutableBigIntegers. Returns -1, 0 or 1
 256      * as this MutableBigInteger is numerically less than, equal to, or
 257      * greater than <tt>b</tt>.
 258      */
 259     final int compare(MutableBigInteger b) {
 260         int blen = b.intLen;
 261         if (intLen < blen)
 262             return -1;
 263         if (intLen > blen)
 264            return 1;
 265 
 266         // Add Integer.MIN_VALUE to make the comparison act as unsigned integer
 267         // comparison.
 268         int[] bval = b.value;
 269         for (int i = offset, j = b.offset; i < intLen + offset; i++, j++) {
 270             int b1 = value[i] + 0x80000000;
 271             int b2 = bval[j]  + 0x80000000;
 272             if (b1 < b2)
 273                 return -1;
 274             if (b1 > b2)
 275                 return 1;
 276         }
 277         return 0;
 278     }
 279 
 280     /**
 281      * Returns a value equal to what {@code b.leftShift(32*ints); return compare(b);}
 282      * would return, but doesn't change the value of {@code b}.
 283      */
 284     private int compareShifted(MutableBigInteger b, int ints) {
 285         int blen = b.intLen;
 286         int alen = intLen - ints;
 287         if (alen < blen)
 288             return -1;
 289         if (alen > blen)
 290            return 1;
 291 
 292         // Add Integer.MIN_VALUE to make the comparison act as unsigned integer
 293         // comparison.
 294         int[] bval = b.value;
 295         for (int i = offset, j = b.offset; i < alen + offset; i++, j++) {
 296             int b1 = value[i] + 0x80000000;
 297             int b2 = bval[j]  + 0x80000000;
 298             if (b1 < b2)
 299                 return -1;
 300             if (b1 > b2)
 301                 return 1;
 302         }
 303         return 0;
 304     }
 305 
 306     /**
 307      * Compare this against half of a MutableBigInteger object (Needed for
 308      * remainder tests).
 309      * Assumes no leading unnecessary zeros, which holds for results
 310      * from divide().
 311      */
 312     final int compareHalf(MutableBigInteger b) {
 313         int blen = b.intLen;
 314         int len = intLen;
 315         if (len <= 0)
 316             return blen <=0 ? 0 : -1;
 317         if (len > blen)
 318             return 1;
 319         if (len < blen - 1)
 320             return -1;
 321         int[] bval = b.value;
 322         int bstart = 0;
 323         int carry = 0;
 324         // Only 2 cases left:len == blen or len == blen - 1
 325         if (len != blen) { // len == blen - 1
 326             if (bval[bstart] == 1) {
 327                 ++bstart;
 328                 carry = 0x80000000;
 329             } else
 330                 return -1;
 331         }
 332         // compare values with right-shifted values of b,
 333         // carrying shifted-out bits across words
 334         int[] val = value;
 335         for (int i = offset, j = bstart; i < len + offset;) {
 336             int bv = bval[j++];
 337             long hb = ((bv >>> 1) + carry) & LONG_MASK;
 338             long v = val[i++] & LONG_MASK;
 339             if (v != hb)
 340                 return v < hb ? -1 : 1;
 341             carry = (bv & 1) << 31; // carray will be either 0x80000000 or 0
 342         }
 343         return carry == 0? 0 : -1;
 344     }
 345 
 346     /**
 347      * Return the index of the lowest set bit in this MutableBigInteger. If the
 348      * magnitude of this MutableBigInteger is zero, -1 is returned.
 349      */
 350     private final int getLowestSetBit() {
 351         if (intLen == 0)
 352             return -1;
 353         int j, b;
 354         for (j=intLen-1; (j>0) && (value[j+offset]==0); j--)
 355             ;
 356         b = value[j+offset];
 357         if (b==0)
 358             return -1;
 359         return ((intLen-1-j)<<5) + Integer.numberOfTrailingZeros(b);
 360     }
 361 
 362     /**
 363      * Return the int in use in this MutableBigInteger at the specified
 364      * index. This method is not used because it is not inlined on all
 365      * platforms.
 366      */
 367     private final int getInt(int index) {
 368         return value[offset+index];
 369     }
 370 
 371     /**
 372      * Return a long which is equal to the unsigned value of the int in
 373      * use in this MutableBigInteger at the specified index. This method is
 374      * not used because it is not inlined on all platforms.
 375      */
 376     private final long getLong(int index) {
 377         return value[offset+index] & LONG_MASK;
 378     }
 379 
 380     /**
 381      * Ensure that the MutableBigInteger is in normal form, specifically
 382      * making sure that there are no leading zeros, and that if the
 383      * magnitude is zero, then intLen is zero.
 384      */
 385     final void normalize() {
 386         if (intLen == 0) {
 387             offset = 0;
 388             return;
 389         }
 390 
 391         int index = offset;
 392         if (value[index] != 0)
 393             return;
 394 
 395         int indexBound = index+intLen;
 396         do {
 397             index++;
 398         } while(index < indexBound && value[index]==0);
 399 
 400         int numZeros = index - offset;
 401         intLen -= numZeros;
 402         offset = (intLen==0 ?  0 : offset+numZeros);
 403     }
 404 
 405     /**
 406      * If this MutableBigInteger cannot hold len words, increase the size
 407      * of the value array to len words.
 408      */
 409     private final void ensureCapacity(int len) {
 410         if (value.length < len) {
 411             value = new int[len];
 412             offset = 0;
 413             intLen = len;
 414         }
 415     }
 416 
 417     /**
 418      * Convert this MutableBigInteger into an int array with no leading
 419      * zeros, of a length that is equal to this MutableBigInteger's intLen.
 420      */
 421     int[] toIntArray() {
 422         int[] result = new int[intLen];
 423         for(int i=0; i<intLen; i++)
 424             result[i] = value[offset+i];
 425         return result;
 426     }
 427 
 428     /**
 429      * Sets the int at index+offset in this MutableBigInteger to val.
 430      * This does not get inlined on all platforms so it is not used
 431      * as often as originally intended.
 432      */
 433     void setInt(int index, int val) {
 434         value[offset + index] = val;
 435     }
 436 
 437     /**
 438      * Sets this MutableBigInteger's value array to the specified array.
 439      * The intLen is set to the specified length.
 440      */
 441     void setValue(int[] val, int length) {
 442         value = val;
 443         intLen = length;
 444         offset = 0;
 445     }
 446 
 447     /**
 448      * Sets this MutableBigInteger's value array to a copy of the specified
 449      * array. The intLen is set to the length of the new array.
 450      */
 451     void copyValue(MutableBigInteger src) {
 452         int len = src.intLen;
 453         if (value.length < len)
 454             value = new int[len];
 455         System.arraycopy(src.value, src.offset, value, 0, len);
 456         intLen = len;
 457         offset = 0;
 458     }
 459 
 460     /**
 461      * Sets this MutableBigInteger's value array to a copy of the specified
 462      * array. The intLen is set to the length of the specified array.
 463      */
 464     void copyValue(int[] val) {
 465         int len = val.length;
 466         if (value.length < len)
 467             value = new int[len];
 468         System.arraycopy(val, 0, value, 0, len);
 469         intLen = len;
 470         offset = 0;
 471     }
 472 
 473     /**
 474      * Returns true iff this MutableBigInteger has a value of one.
 475      */
 476     boolean isOne() {
 477         return (intLen == 1) && (value[offset] == 1);
 478     }
 479 
 480     /**
 481      * Returns true iff this MutableBigInteger has a value of zero.
 482      */
 483     boolean isZero() {
 484         return (intLen == 0);
 485     }
 486 
 487     /**
 488      * Returns true iff this MutableBigInteger is even.
 489      */
 490     boolean isEven() {
 491         return (intLen == 0) || ((value[offset + intLen - 1] & 1) == 0);
 492     }
 493 
 494     /**
 495      * Returns true iff this MutableBigInteger is odd.
 496      */
 497     boolean isOdd() {
 498         return isZero() ? false : ((value[offset + intLen - 1] & 1) == 1);
 499     }
 500 
 501     /**
 502      * Returns true iff this MutableBigInteger is in normal form. A
 503      * MutableBigInteger is in normal form if it has no leading zeros
 504      * after the offset, and intLen + offset <= value.length.
 505      */
 506     boolean isNormal() {
 507         if (intLen + offset > value.length)
 508             return false;
 509         if (intLen ==0)
 510             return true;
 511         return (value[offset] != 0);
 512     }
 513 
 514     /**
 515      * Returns a String representation of this MutableBigInteger in radix 10.
 516      */
 517     public String toString() {
 518         BigInteger b = toBigInteger(1);
 519         return b.toString();
 520     }
 521 
 522     /**
 523      * Like {@link #rightShift(int)} but {@code n} can be greater than the length of the number.
 524      */
 525     void safeRightShift(int n) {
 526         if (n/32 >= intLen)
 527             reset();
 528         else
 529             rightShift(n);
 530     }
 531 
 532     /**
 533      * Right shift this MutableBigInteger n bits. The MutableBigInteger is left
 534      * in normal form.
 535      */
 536     void rightShift(int n) {
 537         if (intLen == 0)
 538             return;
 539         int nInts = n >>> 5;
 540         int nBits = n & 0x1F;
 541         this.intLen -= nInts;
 542         if (nBits == 0)
 543             return;
 544         int bitsInHighWord = BigInteger.bitLengthForInt(value[offset]);
 545         if (nBits >= bitsInHighWord) {
 546             this.primitiveLeftShift(32 - nBits);
 547             this.intLen--;
 548         } else {
 549             primitiveRightShift(nBits);
 550         }
 551     }
 552 
 553     /**
 554      * Like {@link #leftShift(int)} but {@code n} can be zero.
 555      */
 556     void safeLeftShift(int n) {
 557         if (n > 0)
 558             leftShift(n);
 559     }
 560 
 561     /**
 562      * Left shift this MutableBigInteger n bits.
 563      */
 564     void leftShift(int n) {
 565         /*
 566          * If there is enough storage space in this MutableBigInteger already
 567          * the available space will be used. Space to the right of the used
 568          * ints in the value array is faster to utilize, so the extra space
 569          * will be taken from the right if possible.
 570          */
 571         if (intLen == 0)
 572            return;
 573         int nInts = n >>> 5;
 574         int nBits = n&0x1F;
 575         int bitsInHighWord = BigInteger.bitLengthForInt(value[offset]);
 576 
 577         // If shift can be done without moving words, do so
 578         if (n <= (32-bitsInHighWord)) {
 579             primitiveLeftShift(nBits);
 580             return;
 581         }
 582 
 583         int newLen = intLen + nInts +1;
 584         if (nBits <= (32-bitsInHighWord))
 585             newLen--;
 586         if (value.length < newLen) {
 587             // The array must grow
 588             int[] result = new int[newLen];
 589             for (int i=0; i<intLen; i++)
 590                 result[i] = value[offset+i];
 591             setValue(result, newLen);
 592         } else if (value.length - offset >= newLen) {
 593             // Use space on right
 594             for(int i=0; i<newLen - intLen; i++)
 595                 value[offset+intLen+i] = 0;
 596         } else {
 597             // Must use space on left
 598             for (int i=0; i<intLen; i++)
 599                 value[i] = value[offset+i];
 600             for (int i=intLen; i<newLen; i++)
 601                 value[i] = 0;
 602             offset = 0;
 603         }
 604         intLen = newLen;
 605         if (nBits == 0)
 606             return;
 607         if (nBits <= (32-bitsInHighWord))
 608             primitiveLeftShift(nBits);
 609         else
 610             primitiveRightShift(32 -nBits);
 611     }
 612 
 613     /**
 614      * A primitive used for division. This method adds in one multiple of the
 615      * divisor a back to the dividend result at a specified offset. It is used
 616      * when qhat was estimated too large, and must be adjusted.
 617      */
 618     private int divadd(int[] a, int[] result, int offset) {
 619         long carry = 0;
 620 
 621         for (int j=a.length-1; j >= 0; j--) {
 622             long sum = (a[j] & LONG_MASK) +
 623                        (result[j+offset] & LONG_MASK) + carry;
 624             result[j+offset] = (int)sum;
 625             carry = sum >>> 32;
 626         }
 627         return (int)carry;
 628     }
 629 
 630     /**
 631      * This method is used for division. It multiplies an n word input a by one
 632      * word input x, and subtracts the n word product from q. This is needed
 633      * when subtracting qhat*divisor from dividend.
 634      */
 635     private int mulsub(int[] q, int[] a, int x, int len, int offset) {
 636         long xLong = x & LONG_MASK;
 637         long carry = 0;
 638         offset += len;
 639 
 640         for (int j=len-1; j >= 0; j--) {
 641             long product = (a[j] & LONG_MASK) * xLong + carry;
 642             long difference = q[offset] - product;
 643             q[offset--] = (int)difference;
 644             carry = (product >>> 32)
 645                      + (((difference & LONG_MASK) >
 646                          (((~(int)product) & LONG_MASK))) ? 1:0);
 647         }
 648         return (int)carry;
 649     }
 650 
 651     /**
 652      * The method is the same as mulsun, except the fact that q array is not
 653      * updated, the only result of the method is borrow flag.
 654      */
 655     private int mulsubBorrow(int[] q, int[] a, int x, int len, int offset) {
 656         long xLong = x & LONG_MASK;
 657         long carry = 0;
 658         offset += len;
 659         for (int j=len-1; j >= 0; j--) {
 660             long product = (a[j] & LONG_MASK) * xLong + carry;
 661             long difference = q[offset--] - product;
 662             carry = (product >>> 32)
 663                      + (((difference & LONG_MASK) >
 664                          (((~(int)product) & LONG_MASK))) ? 1:0);
 665         }
 666         return (int)carry;
 667     }
 668 
 669     /**
 670      * Right shift this MutableBigInteger n bits, where n is
 671      * less than 32.
 672      * Assumes that intLen > 0, n > 0 for speed
 673      */
 674     private final void primitiveRightShift(int n) {
 675         int[] val = value;
 676         int n2 = 32 - n;
 677         for (int i=offset+intLen-1, c=val[i]; i>offset; i--) {
 678             int b = c;
 679             c = val[i-1];
 680             val[i] = (c << n2) | (b >>> n);
 681         }
 682         val[offset] >>>= n;
 683     }
 684 
 685     /**
 686      * Left shift this MutableBigInteger n bits, where n is
 687      * less than 32.
 688      * Assumes that intLen > 0, n > 0 for speed
 689      */
 690     private final void primitiveLeftShift(int n) {
 691         int[] val = value;
 692         int n2 = 32 - n;
 693         for (int i=offset, c=val[i], m=i+intLen-1; i<m; i++) {
 694             int b = c;
 695             c = val[i+1];
 696             val[i] = (b << n) | (c >>> n2);
 697         }
 698         val[offset+intLen-1] <<= n;
 699     }
 700 
 701     /**
 702      * Returns a {@code BigInteger} equal to the {@code n}
 703      * low ints of this number.
 704      */
 705     private BigInteger getLower(int n) {
 706         if (isZero())
 707             return BigInteger.ZERO;
 708         else if (intLen < n)
 709             return toBigInteger(1);
 710         else {
 711             // strip zeros
 712             int len = n;
 713             while (len>0 && value[offset+intLen-len]==0)
 714                 len--;
 715             int sign = len>0 ? 1 : 0;
 716             return new BigInteger(Arrays.copyOfRange(value, offset+intLen-len, offset+intLen), sign);
 717         }
 718     }
 719 
 720     /**
 721      * Discards all ints whose index is greater than {@code n}.
 722      */
 723     private void keepLower(int n) {
 724         if (intLen >= n) {
 725             offset += intLen - n;
 726             intLen = n;
 727         }
 728     }
 729 
 730     /**
 731      * Adds the contents of two MutableBigInteger objects.The result
 732      * is placed within this MutableBigInteger.
 733      * The contents of the addend are not changed.
 734      */
 735     void add(MutableBigInteger addend) {
 736         int x = intLen;
 737         int y = addend.intLen;
 738         int resultLen = (intLen > addend.intLen ? intLen : addend.intLen);
 739         int[] result = (value.length < resultLen ? new int[resultLen] : value);
 740 
 741         int rstart = result.length-1;
 742         long sum;
 743         long carry = 0;
 744 
 745         // Add common parts of both numbers
 746         while(x>0 && y>0) {
 747             x--; y--;
 748             sum = (value[x+offset] & LONG_MASK) +
 749                 (addend.value[y+addend.offset] & LONG_MASK) + carry;
 750             result[rstart--] = (int)sum;
 751             carry = sum >>> 32;
 752         }
 753 
 754         // Add remainder of the longer number
 755         while(x>0) {
 756             x--;
 757             if (carry == 0 && result == value && rstart == (x + offset))
 758                 return;
 759             sum = (value[x+offset] & LONG_MASK) + carry;
 760             result[rstart--] = (int)sum;
 761             carry = sum >>> 32;
 762         }
 763         while(y>0) {
 764             y--;
 765             sum = (addend.value[y+addend.offset] & LONG_MASK) + carry;
 766             result[rstart--] = (int)sum;
 767             carry = sum >>> 32;
 768         }
 769 
 770         if (carry > 0) { // Result must grow in length
 771             resultLen++;
 772             if (result.length < resultLen) {
 773                 int temp[] = new int[resultLen];
 774                 // Result one word longer from carry-out; copy low-order
 775                 // bits into new result.
 776                 System.arraycopy(result, 0, temp, 1, result.length);
 777                 temp[0] = 1;
 778                 result = temp;
 779             } else {
 780                 result[rstart--] = 1;
 781             }
 782         }
 783 
 784         value = result;
 785         intLen = resultLen;
 786         offset = result.length - resultLen;
 787     }
 788 
 789     /**
 790      * Adds the value of {@code addend} shifted {@code n} ints to the left.
 791      * Has the same effect as {@code addend.leftShift(32*ints); add(b);}
 792      * but doesn't change the value of {@code b}.
 793      */
 794     void addShifted(MutableBigInteger addend, int n) {
 795         if (addend.isZero())
 796             return;
 797 
 798         int x = intLen;
 799         int y = addend.intLen + n;
 800         int resultLen = (intLen > y ? intLen : y);
 801         int[] result = (value.length < resultLen ? new int[resultLen] : value);
 802 
 803         int rstart = result.length-1;
 804         long sum;
 805         long carry = 0;
 806 
 807         // Add common parts of both numbers
 808         while(x>0 && y>0) {
 809             x--; y--;
 810             int bval = y+addend.offset<addend.value.length ? addend.value[y+addend.offset] : 0;
 811             sum = (value[x+offset] & LONG_MASK) +
 812                 (bval & LONG_MASK) + carry;
 813             result[rstart--] = (int)sum;
 814             carry = sum >>> 32;
 815         }
 816 
 817         // Add remainder of the longer number
 818         while(x>0) {
 819             x--;
 820             if (carry == 0 && result == value && rstart == (x + offset))
 821                 return;
 822             sum = (value[x+offset] & LONG_MASK) + carry;
 823             result[rstart--] = (int)sum;
 824             carry = sum >>> 32;
 825         }
 826         while(y>0) {
 827             y--;
 828             int bval = y+addend.offset<addend.value.length ? addend.value[y+addend.offset] : 0;
 829             sum = (bval & LONG_MASK) + carry;
 830             result[rstart--] = (int)sum;
 831             carry = sum >>> 32;
 832         }
 833 
 834         if (carry > 0) { // Result must grow in length
 835             resultLen++;
 836             if (result.length < resultLen) {
 837                 int temp[] = new int[resultLen];
 838                 // Result one word longer from carry-out; copy low-order
 839                 // bits into new result.
 840                 System.arraycopy(result, 0, temp, 1, result.length);
 841                 temp[0] = 1;
 842                 result = temp;
 843             } else {
 844                 result[rstart--] = 1;
 845             }
 846         }
 847 
 848         value = result;
 849         intLen = resultLen;
 850         offset = result.length - resultLen;
 851     }
 852 
 853     /**
 854      * Like {@link #addShifted(MutableBigInteger, int)} but {@code this.intLen} must
 855      * not be greater than {@code n}. In other words, concatenates {@code this}
 856      * and {@code addend}.
 857      */
 858     void addDisjoint(MutableBigInteger addend, int n) {
 859         if (addend.isZero())
 860             return;
 861 
 862         int x = intLen;
 863         int y = addend.intLen + n;
 864         int resultLen = (intLen > y ? intLen : y);
 865         int[] result;
 866         if (value.length < resultLen)
 867             result = new int[resultLen];
 868         else {
 869             result = value;
 870             Arrays.fill(value, offset+intLen, value.length, 0);
 871         }
 872 
 873         int rstart = result.length-1;
 874 
 875         // copy from this if needed
 876         System.arraycopy(value, offset, result, rstart+1-x, x);
 877         y -= x;
 878         rstart -= x;
 879 
 880         int len = Math.min(y, addend.value.length-addend.offset);
 881         System.arraycopy(addend.value, addend.offset, result, rstart+1-y, len);
 882 
 883         // zero the gap
 884         for (int i=rstart+1-y+len; i<rstart+1; i++)
 885             result[i] = 0;
 886 
 887         value = result;
 888         intLen = resultLen;
 889         offset = result.length - resultLen;
 890     }
 891 
 892     /**
 893      * Adds the low {@code n} ints of {@code addend}.
 894      */
 895     void addLower(MutableBigInteger addend, int n) {
 896         MutableBigInteger a = new MutableBigInteger(addend);
 897         if (a.offset + a.intLen >= n) {
 898             a.offset = a.offset + a.intLen - n;
 899             a.intLen = n;
 900         }
 901         a.normalize();
 902         add(a);
 903     }
 904 
 905     /**
 906      * Subtracts the smaller of this and b from the larger and places the
 907      * result into this MutableBigInteger.
 908      */
 909     int subtract(MutableBigInteger b) {
 910         MutableBigInteger a = this;
 911 
 912         int[] result = value;
 913         int sign = a.compare(b);
 914 
 915         if (sign == 0) {
 916             reset();
 917             return 0;
 918         }
 919         if (sign < 0) {
 920             MutableBigInteger tmp = a;
 921             a = b;
 922             b = tmp;
 923         }
 924 
 925         int resultLen = a.intLen;
 926         if (result.length < resultLen)
 927             result = new int[resultLen];
 928 
 929         long diff = 0;
 930         int x = a.intLen;
 931         int y = b.intLen;
 932         int rstart = result.length - 1;
 933 
 934         // Subtract common parts of both numbers
 935         while (y>0) {
 936             x--; y--;
 937 
 938             diff = (a.value[x+a.offset] & LONG_MASK) -
 939                    (b.value[y+b.offset] & LONG_MASK) - ((int)-(diff>>32));
 940             result[rstart--] = (int)diff;
 941         }
 942         // Subtract remainder of longer number
 943         while (x>0) {
 944             x--;
 945             diff = (a.value[x+a.offset] & LONG_MASK) - ((int)-(diff>>32));
 946             result[rstart--] = (int)diff;
 947         }
 948 
 949         value = result;
 950         intLen = resultLen;
 951         offset = value.length - resultLen;
 952         normalize();
 953         return sign;
 954     }
 955 
 956     /**
 957      * Subtracts the smaller of a and b from the larger and places the result
 958      * into the larger. Returns 1 if the answer is in a, -1 if in b, 0 if no
 959      * operation was performed.
 960      */
 961     private int difference(MutableBigInteger b) {
 962         MutableBigInteger a = this;
 963         int sign = a.compare(b);
 964         if (sign ==0)
 965             return 0;
 966         if (sign < 0) {
 967             MutableBigInteger tmp = a;
 968             a = b;
 969             b = tmp;
 970         }
 971 
 972         long diff = 0;
 973         int x = a.intLen;
 974         int y = b.intLen;
 975 
 976         // Subtract common parts of both numbers
 977         while (y>0) {
 978             x--; y--;
 979             diff = (a.value[a.offset+ x] & LONG_MASK) -
 980                 (b.value[b.offset+ y] & LONG_MASK) - ((int)-(diff>>32));
 981             a.value[a.offset+x] = (int)diff;
 982         }
 983         // Subtract remainder of longer number
 984         while (x>0) {
 985             x--;
 986             diff = (a.value[a.offset+ x] & LONG_MASK) - ((int)-(diff>>32));
 987             a.value[a.offset+x] = (int)diff;
 988         }
 989 
 990         a.normalize();
 991         return sign;
 992     }
 993 
 994     /**
 995      * Multiply the contents of two MutableBigInteger objects. The result is
 996      * placed into MutableBigInteger z. The contents of y are not changed.
 997      */
 998     void multiply(MutableBigInteger y, MutableBigInteger z) {
 999         int xLen = intLen;
1000         int yLen = y.intLen;
1001         int newLen = xLen + yLen;
1002 
1003         // Put z into an appropriate state to receive product
1004         if (z.value.length < newLen)
1005             z.value = new int[newLen];
1006         z.offset = 0;
1007         z.intLen = newLen;
1008 
1009         // The first iteration is hoisted out of the loop to avoid extra add
1010         long carry = 0;
1011         for (int j=yLen-1, k=yLen+xLen-1; j >= 0; j--, k--) {
1012                 long product = (y.value[j+y.offset] & LONG_MASK) *
1013                                (value[xLen-1+offset] & LONG_MASK) + carry;
1014                 z.value[k] = (int)product;
1015                 carry = product >>> 32;
1016         }
1017         z.value[xLen-1] = (int)carry;
1018 
1019         // Perform the multiplication word by word
1020         for (int i = xLen-2; i >= 0; i--) {
1021             carry = 0;
1022             for (int j=yLen-1, k=yLen+i; j >= 0; j--, k--) {
1023                 long product = (y.value[j+y.offset] & LONG_MASK) *
1024                                (value[i+offset] & LONG_MASK) +
1025                                (z.value[k] & LONG_MASK) + carry;
1026                 z.value[k] = (int)product;
1027                 carry = product >>> 32;
1028             }
1029             z.value[i] = (int)carry;
1030         }
1031 
1032         // Remove leading zeros from product
1033         z.normalize();
1034     }
1035 
1036     /**
1037      * Multiply the contents of this MutableBigInteger by the word y. The
1038      * result is placed into z.
1039      */
1040     void mul(int y, MutableBigInteger z) {
1041         if (y == 1) {
1042             z.copyValue(this);
1043             return;
1044         }
1045 
1046         if (y == 0) {
1047             z.clear();
1048             return;
1049         }
1050 
1051         // Perform the multiplication word by word
1052         long ylong = y & LONG_MASK;
1053         int[] zval = (z.value.length<intLen+1 ? new int[intLen + 1]
1054                                               : z.value);
1055         long carry = 0;
1056         for (int i = intLen-1; i >= 0; i--) {
1057             long product = ylong * (value[i+offset] & LONG_MASK) + carry;
1058             zval[i+1] = (int)product;
1059             carry = product >>> 32;
1060         }
1061 
1062         if (carry == 0) {
1063             z.offset = 1;
1064             z.intLen = intLen;
1065         } else {
1066             z.offset = 0;
1067             z.intLen = intLen + 1;
1068             zval[0] = (int)carry;
1069         }
1070         z.value = zval;
1071     }
1072 
1073      /**
1074      * This method is used for division of an n word dividend by a one word
1075      * divisor. The quotient is placed into quotient. The one word divisor is
1076      * specified by divisor.
1077      *
1078      * @return the remainder of the division is returned.
1079      *
1080      */
1081     int divideOneWord(int divisor, MutableBigInteger quotient) {
1082         long divisorLong = divisor & LONG_MASK;
1083 
1084         // Special case of one word dividend
1085         if (intLen == 1) {
1086             long dividendValue = value[offset] & LONG_MASK;
1087             int q = (int) (dividendValue / divisorLong);
1088             int r = (int) (dividendValue - q * divisorLong);
1089             quotient.value[0] = q;
1090             quotient.intLen = (q == 0) ? 0 : 1;
1091             quotient.offset = 0;
1092             return r;
1093         }
1094 
1095         if (quotient.value.length < intLen)
1096             quotient.value = new int[intLen];
1097         quotient.offset = 0;
1098         quotient.intLen = intLen;
1099 
1100         // Normalize the divisor
1101         int shift = Integer.numberOfLeadingZeros(divisor);
1102 
1103         int rem = value[offset];
1104         long remLong = rem & LONG_MASK;
1105         if (remLong < divisorLong) {
1106             quotient.value[0] = 0;
1107         } else {
1108             quotient.value[0] = (int)(remLong / divisorLong);
1109             rem = (int) (remLong - (quotient.value[0] * divisorLong));
1110             remLong = rem & LONG_MASK;
1111         }
1112         int xlen = intLen;
1113         while (--xlen > 0) {
1114             long dividendEstimate = (remLong << 32) |
1115                     (value[offset + intLen - xlen] & LONG_MASK);
1116             int q;
1117             if (dividendEstimate >= 0) {
1118                 q = (int) (dividendEstimate / divisorLong);
1119                 rem = (int) (dividendEstimate - q * divisorLong);
1120             } else {
1121                 long tmp = divWord(dividendEstimate, divisor);
1122                 q = (int) (tmp & LONG_MASK);
1123                 rem = (int) (tmp >>> 32);
1124             }
1125             quotient.value[intLen - xlen] = q;
1126             remLong = rem & LONG_MASK;
1127         }
1128 
1129         quotient.normalize();
1130         // Unnormalize
1131         if (shift > 0)
1132             return rem % divisor;
1133         else
1134             return rem;
1135     }
1136 
1137     /**
1138      * Calculates the quotient of this div b and places the quotient in the
1139      * provided MutableBigInteger objects and the remainder object is returned.
1140      *
1141      */
1142     MutableBigInteger divide(MutableBigInteger b, MutableBigInteger quotient) {
1143         return divide(b,quotient,true);
1144     }
1145 
1146     MutableBigInteger divide(MutableBigInteger b, MutableBigInteger quotient, boolean needRemainder) {
1147         if (intLen<BigInteger.BURNIKEL_ZIEGLER_THRESHOLD || b.intLen<BigInteger.BURNIKEL_ZIEGLER_THRESHOLD)
1148             return divideKnuth(b, quotient, needRemainder);
1149         else
1150             return divideAndRemainderBurnikelZiegler(b, quotient);
1151     }
1152 
1153     /**
1154      * @see #divideKnuth(MutableBigInteger, MutableBigInteger, boolean)
1155      */
1156     MutableBigInteger divideKnuth(MutableBigInteger b, MutableBigInteger quotient) {
1157         return divideKnuth(b,quotient,true);
1158     }
1159 
1160     /**
1161      * Calculates the quotient of this div b and places the quotient in the
1162      * provided MutableBigInteger objects and the remainder object is returned.
1163      *
1164      * Uses Algorithm D in Knuth section 4.3.1.
1165      * Many optimizations to that algorithm have been adapted from the Colin
1166      * Plumb C library.
1167      * It special cases one word divisors for speed. The content of b is not
1168      * changed.
1169      *
1170      */
1171     MutableBigInteger divideKnuth(MutableBigInteger b, MutableBigInteger quotient, boolean needRemainder) {
1172         if (b.intLen == 0)
1173             throw new ArithmeticException("BigInteger divide by zero");
1174 
1175         // Dividend is zero
1176         if (intLen == 0) {
1177             quotient.intLen = quotient.offset = 0;
1178             return needRemainder ? new MutableBigInteger() : null;
1179         }
1180 
1181         int cmp = compare(b);
1182         // Dividend less than divisor
1183         if (cmp < 0) {
1184             quotient.intLen = quotient.offset = 0;
1185             return needRemainder ? new MutableBigInteger(this) : null;
1186         }
1187         // Dividend equal to divisor
1188         if (cmp == 0) {
1189             quotient.value[0] = quotient.intLen = 1;
1190             quotient.offset = 0;
1191             return needRemainder ? new MutableBigInteger() : null;
1192         }
1193 
1194         quotient.clear();
1195         // Special case one word divisor
1196         if (b.intLen == 1) {
1197             int r = divideOneWord(b.value[b.offset], quotient);
1198             if(needRemainder) {
1199                 if (r == 0)
1200                     return new MutableBigInteger();
1201                 return new MutableBigInteger(r);
1202             } else {
1203                 return null;
1204             }
1205         }
1206 
1207         // Cancel common powers of two if we're above the KNUTH_POW2_* thresholds
1208         if (intLen >= KNUTH_POW2_THRESH_LEN) {
1209             int trailingZeroBits = Math.min(getLowestSetBit(), b.getLowestSetBit());
1210             if (trailingZeroBits >= KNUTH_POW2_THRESH_ZEROS*32) {
1211                 MutableBigInteger a = new MutableBigInteger(this);
1212                 b = new MutableBigInteger(b);
1213                 a.rightShift(trailingZeroBits);
1214                 b.rightShift(trailingZeroBits);
1215                 MutableBigInteger r = a.divideKnuth(b, quotient);
1216                 r.leftShift(trailingZeroBits);
1217                 return r;
1218             }
1219         }
1220 
1221         return divideMagnitude(b, quotient, needRemainder);
1222     }
1223 
1224     /**
1225      * Computes {@code this/b} and {@code this%b} using the
1226      * <a href="http://cr.yp.to/bib/1998/burnikel.ps"> Burnikel-Ziegler algorithm</a>.
1227      * This method implements algorithm 3 from pg. 9 of the Burnikel-Ziegler paper.
1228      * The parameter beta was chosen to b 2<sup>32</sup> so almost all shifts are
1229      * multiples of 32 bits.<br/>
1230      * {@code this} and {@code b} must be nonnegative.
1231      * @param b the divisor
1232      * @param quotient output parameter for {@code this/b}
1233      * @return the remainder
1234      */
1235     MutableBigInteger divideAndRemainderBurnikelZiegler(MutableBigInteger b, MutableBigInteger quotient) {
1236         int r = intLen;
1237         int s = b.intLen;
1238 
1239         if (r < s)
1240             return this;
1241         else {
1242             // Unlike Knuth division, we don't check for common powers of two here because
1243             // BZ already runs faster if both numbers contain powers of two and cancelling them has no
1244             // additional benefit.
1245 
1246             // step 1: let m = min{2^k | (2^k)*BURNIKEL_ZIEGLER_THRESHOLD > s}
1247             int m = 1 << (32-Integer.numberOfLeadingZeros(s/BigInteger.BURNIKEL_ZIEGLER_THRESHOLD));
1248 
1249             int j = (s+m-1) / m;      // step 2a: j = ceil(s/m)
1250             int n = j * m;            // step 2b: block length in 32-bit units
1251             int n32 = 32 * n;         // block length in bits
1252             int sigma = Math.max(0, n32 - b.bitLength());   // step 3: sigma = max{T | (2^T)*B < beta^n}
1253             MutableBigInteger bShifted = new MutableBigInteger(b);
1254             bShifted.safeLeftShift(sigma);   // step 4a: shift b so its length is a multiple of n
1255             safeLeftShift(sigma);     // step 4b: shift this by the same amount
1256 
1257             // step 5: t is the number of blocks needed to accommodate this plus one additional bit
1258             int t = (bitLength()+n32) / n32;
1259             if (t < 2)
1260                 t = 2;
1261 
1262             // step 6: conceptually split this into blocks a[t-1], ..., a[0]
1263             MutableBigInteger a1 = getBlock(t-1, t, n);   // the most significant block of this
1264 
1265             // step 7: z[t-2] = [a[t-1], a[t-2]]
1266             MutableBigInteger z = getBlock(t-2, t, n);    // the second to most significant block
1267             z.addDisjoint(a1, n);   // z[t-2]
1268 
1269             // do schoolbook division on blocks, dividing 2-block numbers by 1-block numbers
1270             MutableBigInteger qi = new MutableBigInteger();
1271             MutableBigInteger ri;
1272             quotient.offset = quotient.intLen = 0;
1273             for (int i=t-2; i>0; i--) {
1274                 // step 8a: compute (qi,ri) such that z=b*qi+ri
1275                 ri = z.divide2n1n(bShifted, qi);
1276 
1277                 // step 8b: z = [ri, a[i-1]]
1278                 z = getBlock(i-1, t, n);   // a[i-1]
1279                 z.addDisjoint(ri, n);
1280                 quotient.addShifted(qi, i*n);   // update q (part of step 9)
1281             }
1282             // final iteration of step 8: do the loop one more time for i=0 but leave z unchanged
1283             ri = z.divide2n1n(bShifted, qi);
1284             quotient.add(qi);
1285 
1286             ri.rightShift(sigma);   // step 9: this and b were shifted, so shift back
1287             return ri;
1288         }
1289     }
1290 
1291     /**
1292      * This method implements algorithm 1 from pg. 4 of the Burnikel-Ziegler paper.
1293      * It divides a 2n-digit number by a n-digit number.<br/>
1294      * The parameter beta is 2<sup>32</sup> so all shifts are multiples of 32 bits.
1295      * <br/>
1296      * {@code this} must be a nonnegative number such that {@code this.bitLength() <= 2*b.bitLength()}
1297      * @param b a positive number such that {@code b.bitLength()} is even
1298      * @param quotient output parameter for {@code this/b}
1299      * @return {@code this%b}
1300      */
1301     private MutableBigInteger divide2n1n(MutableBigInteger b, MutableBigInteger quotient) {
1302         int n = b.intLen;
1303 
1304         // step 1: base case
1305         if (n%2!=0 || n<BigInteger.BURNIKEL_ZIEGLER_THRESHOLD)
1306             return divideKnuth(b, quotient);
1307 
1308         // step 2: view this as [a1,a2,a3,a4] where each ai is n/2 ints or less
1309         MutableBigInteger aUpper = new MutableBigInteger(this);
1310         aUpper.safeRightShift(32*(n/2));   // aUpper = [a1,a2,a3]
1311         keepLower(n/2);   // this = a4
1312 
1313         // step 3: q1=aUpper/b, r1=aUpper%b
1314         MutableBigInteger q1 = new MutableBigInteger();
1315         MutableBigInteger r1 = aUpper.divide3n2n(b, q1);
1316 
1317         // step 4: quotient=[r1,this]/b, r2=[r1,this]%b
1318         addDisjoint(r1, n/2);   // this = [r1,this]
1319         MutableBigInteger r2 = divide3n2n(b, quotient);
1320 
1321         // step 5: let quotient=[q1,quotient] and return r2
1322         quotient.addDisjoint(q1, n/2);
1323         return r2;
1324     }
1325 
1326     /**
1327      * This method implements algorithm 2 from pg. 5 of the Burnikel-Ziegler paper.
1328      * It divides a 3n-digit number by a 2n-digit number.<br/>
1329      * The parameter beta is 2<sup>32</sup> so all shifts are multiples of 32 bits.<br/>
1330      * <br/>
1331      * {@code this} must be a nonnegative number such that {@code 2*this.bitLength() <= 3*b.bitLength()}
1332      * @param quotient output parameter for {@code this/b}
1333      * @return {@code this%b}
1334      */
1335     private MutableBigInteger divide3n2n(MutableBigInteger b, MutableBigInteger quotient) {
1336         int n = b.intLen / 2;   // half the length of b in ints
1337 
1338         // step 1: view this as [a1,a2,a3] where each ai is n ints or less; let a12=[a1,a2]
1339         MutableBigInteger a12 = new MutableBigInteger(this);
1340         a12.safeRightShift(32*n);
1341 
1342         // step 2: view b as [b1,b2] where each bi is n ints or less
1343         MutableBigInteger b1 = new MutableBigInteger(b);
1344         b1.safeRightShift(n * 32);
1345         BigInteger b2 = b.getLower(n);
1346 
1347         MutableBigInteger r;
1348         MutableBigInteger d;
1349         if (compareShifted(b, n) < 0) {
1350             // step 3a: if a1<b1, let quotient=a12/b1 and r=a12%b1
1351             r = a12.divide2n1n(b1, quotient);
1352 
1353             // step 4: d=quotient*b2
1354             d = new MutableBigInteger(quotient.toBigInteger().multiply(b2));
1355         }
1356         else {
1357             // step 3b: if a1>=b1, let quotient=beta^n-1 and r=a12-b1*2^n+b1
1358             quotient.ones(n);
1359             a12.add(b1);
1360             b1.leftShift(32*n);
1361             a12.subtract(b1);
1362             r = a12;
1363 
1364             // step 4: d=quotient*b2=(b2 << 32*n) - b2
1365             d = new MutableBigInteger(b2);
1366             d.leftShift(32 * n);
1367             d.subtract(new MutableBigInteger(b2));
1368         }
1369 
1370         // step 5: r = r*beta^n + a3 - d (paper says a4)
1371         // However, don't subtract d until after the while loop so r doesn't become negative
1372         r.leftShift(32 * n);
1373         r.addLower(this, n);
1374 
1375         // step 6: add b until r>=d
1376         while (r.compare(d) < 0) {
1377             r.add(b);
1378             quotient.subtract(MutableBigInteger.ONE);
1379         }
1380         r.subtract(d);
1381 
1382         return r;
1383     }
1384 
1385     /**
1386      * Returns a {@code MutableBigInteger} containing {@code blockLength} ints from
1387      * {@code this} number, starting at {@code index*blockLength}.<br/>
1388      * Used by Burnikel-Ziegler division.
1389      * @param index the block index
1390      * @param numBlocks the total number of blocks in {@code this} number
1391      * @param blockLength length of one block in units of 32 bits
1392      * @return
1393      */
1394     private MutableBigInteger getBlock(int index, int numBlocks, int blockLength) {
1395         int blockStart = index * blockLength;
1396         if (blockStart >= intLen)
1397             return new MutableBigInteger();
1398 
1399         int blockEnd;
1400         if (index == numBlocks-1)
1401             blockEnd = intLen;
1402         else
1403             blockEnd = (index+1) * blockLength;
1404         if (blockEnd > intLen)
1405             return new MutableBigInteger();
1406 
1407         int[] newVal = Arrays.copyOfRange(value, offset+intLen-blockEnd, offset+intLen-blockStart);
1408         return new MutableBigInteger(newVal);
1409     }
1410 
1411     /** @see BigInteger#bitLength() */
1412     int bitLength() {
1413         if (intLen == 0)
1414             return 0;
1415         return intLen*32 - Integer.numberOfLeadingZeros(value[offset]);
1416     }
1417 
1418     /**
1419      * Internally used  to calculate the quotient of this div v and places the
1420      * quotient in the provided MutableBigInteger object and the remainder is
1421      * returned.
1422      *
1423      * @return the remainder of the division will be returned.
1424      */
1425     long divide(long v, MutableBigInteger quotient) {
1426         if (v == 0)
1427             throw new ArithmeticException("BigInteger divide by zero");
1428 
1429         // Dividend is zero
1430         if (intLen == 0) {
1431             quotient.intLen = quotient.offset = 0;
1432             return 0;
1433         }
1434         if (v < 0)
1435             v = -v;
1436 
1437         int d = (int)(v >>> 32);
1438         quotient.clear();
1439         // Special case on word divisor
1440         if (d == 0)
1441             return divideOneWord((int)v, quotient) & LONG_MASK;
1442         else {
1443             return divideLongMagnitude(v, quotient).toLong();
1444         }
1445     }
1446 
1447     private static void copyAndShift(int[] src, int srcFrom, int srcLen, int[] dst, int dstFrom, int shift) {
1448         int n2 = 32 - shift;
1449         int c=src[srcFrom];
1450         for (int i=0; i < srcLen-1; i++) {
1451             int b = c;
1452             c = src[++srcFrom];
1453             dst[dstFrom+i] = (b << shift) | (c >>> n2);
1454         }
1455         dst[dstFrom+srcLen-1] = c << shift;
1456     }
1457 
1458     /**
1459      * Divide this MutableBigInteger by the divisor.
1460      * The quotient will be placed into the provided quotient object &
1461      * the remainder object is returned.
1462      */
1463     private MutableBigInteger divideMagnitude(MutableBigInteger div,
1464                                               MutableBigInteger quotient,
1465                                               boolean needRemainder ) {
1466         // assert div.intLen > 1
1467         // D1 normalize the divisor
1468         int shift = Integer.numberOfLeadingZeros(div.value[div.offset]);
1469         // Copy divisor value to protect divisor
1470         final int dlen = div.intLen;
1471         int[] divisor;
1472         MutableBigInteger rem; // Remainder starts as dividend with space for a leading zero
1473         if (shift > 0) {
1474             divisor = new int[dlen];
1475             copyAndShift(div.value,div.offset,dlen,divisor,0,shift);
1476             if(Integer.numberOfLeadingZeros(value[offset])>=shift) {
1477                 int[] remarr = new int[intLen + 1];
1478                 rem = new MutableBigInteger(remarr);
1479                 rem.intLen = intLen;
1480                 rem.offset = 1;
1481                 copyAndShift(value,offset,intLen,remarr,1,shift);
1482             } else {
1483                 int[] remarr = new int[intLen + 2];
1484                 rem = new MutableBigInteger(remarr);
1485                 rem.intLen = intLen+1;
1486                 rem.offset = 1;
1487                 int rFrom = offset;
1488                 int c=0;
1489                 int n2 = 32 - shift;
1490                 for (int i=1; i < intLen+1; i++,rFrom++) {
1491                     int b = c;
1492                     c = value[rFrom];
1493                     remarr[i] = (b << shift) | (c >>> n2);
1494                 }
1495                 remarr[intLen+1] = c << shift;
1496             }
1497         } else {
1498             divisor = Arrays.copyOfRange(div.value, div.offset, div.offset + div.intLen);
1499             rem = new MutableBigInteger(new int[intLen + 1]);
1500             System.arraycopy(value, offset, rem.value, 1, intLen);
1501             rem.intLen = intLen;
1502             rem.offset = 1;
1503         }
1504 
1505         int nlen = rem.intLen;
1506 
1507         // Set the quotient size
1508         final int limit = nlen - dlen + 1;
1509         if (quotient.value.length < limit) {
1510             quotient.value = new int[limit];
1511             quotient.offset = 0;
1512         }
1513         quotient.intLen = limit;
1514         int[] q = quotient.value;
1515 
1516 
1517         // Must insert leading 0 in rem if its length did not change
1518         if (rem.intLen == nlen) {
1519             rem.offset = 0;
1520             rem.value[0] = 0;
1521             rem.intLen++;
1522         }
1523 
1524         int dh = divisor[0];
1525         long dhLong = dh & LONG_MASK;
1526         int dl = divisor[1];
1527 
1528         // D2 Initialize j
1529         for(int j=0; j<limit-1; j++) {
1530             // D3 Calculate qhat
1531             // estimate qhat
1532             int qhat = 0;
1533             int qrem = 0;
1534             boolean skipCorrection = false;
1535             int nh = rem.value[j+rem.offset];
1536             int nh2 = nh + 0x80000000;
1537             int nm = rem.value[j+1+rem.offset];
1538 
1539             if (nh == dh) {
1540                 qhat = ~0;
1541                 qrem = nh + nm;
1542                 skipCorrection = qrem + 0x80000000 < nh2;
1543             } else {
1544                 long nChunk = (((long)nh) << 32) | (nm & LONG_MASK);
1545                 if (nChunk >= 0) {
1546                     qhat = (int) (nChunk / dhLong);
1547                     qrem = (int) (nChunk - (qhat * dhLong));
1548                 } else {
1549                     long tmp = divWord(nChunk, dh);
1550                     qhat = (int) (tmp & LONG_MASK);
1551                     qrem = (int) (tmp >>> 32);
1552                 }
1553             }
1554 
1555             if (qhat == 0)
1556                 continue;
1557 
1558             if (!skipCorrection) { // Correct qhat
1559                 long nl = rem.value[j+2+rem.offset] & LONG_MASK;
1560                 long rs = ((qrem & LONG_MASK) << 32) | nl;
1561                 long estProduct = (dl & LONG_MASK) * (qhat & LONG_MASK);
1562 
1563                 if (unsignedLongCompare(estProduct, rs)) {
1564                     qhat--;
1565                     qrem = (int)((qrem & LONG_MASK) + dhLong);
1566                     if ((qrem & LONG_MASK) >=  dhLong) {
1567                         estProduct -= (dl & LONG_MASK);
1568                         rs = ((qrem & LONG_MASK) << 32) | nl;
1569                         if (unsignedLongCompare(estProduct, rs))
1570                             qhat--;
1571                     }
1572                 }
1573             }
1574 
1575             // D4 Multiply and subtract
1576             rem.value[j+rem.offset] = 0;
1577             int borrow = mulsub(rem.value, divisor, qhat, dlen, j+rem.offset);
1578 
1579             // D5 Test remainder
1580             if (borrow + 0x80000000 > nh2) {
1581                 // D6 Add back
1582                 divadd(divisor, rem.value, j+1+rem.offset);
1583                 qhat--;
1584             }
1585 
1586             // Store the quotient digit
1587             q[j] = qhat;
1588         } // D7 loop on j
1589         // D3 Calculate qhat
1590         // estimate qhat
1591         int qhat = 0;
1592         int qrem = 0;
1593         boolean skipCorrection = false;
1594         int nh = rem.value[limit - 1 + rem.offset];
1595         int nh2 = nh + 0x80000000;
1596         int nm = rem.value[limit + rem.offset];
1597 
1598         if (nh == dh) {
1599             qhat = ~0;
1600             qrem = nh + nm;
1601             skipCorrection = qrem + 0x80000000 < nh2;
1602         } else {
1603             long nChunk = (((long) nh) << 32) | (nm & LONG_MASK);
1604             if (nChunk >= 0) {
1605                 qhat = (int) (nChunk / dhLong);
1606                 qrem = (int) (nChunk - (qhat * dhLong));
1607             } else {
1608                 long tmp = divWord(nChunk, dh);
1609                 qhat = (int) (tmp & LONG_MASK);
1610                 qrem = (int) (tmp >>> 32);
1611             }
1612         }
1613         if (qhat != 0) {
1614             if (!skipCorrection) { // Correct qhat
1615                 long nl = rem.value[limit + 1 + rem.offset] & LONG_MASK;
1616                 long rs = ((qrem & LONG_MASK) << 32) | nl;
1617                 long estProduct = (dl & LONG_MASK) * (qhat & LONG_MASK);
1618 
1619                 if (unsignedLongCompare(estProduct, rs)) {
1620                     qhat--;
1621                     qrem = (int) ((qrem & LONG_MASK) + dhLong);
1622                     if ((qrem & LONG_MASK) >= dhLong) {
1623                         estProduct -= (dl & LONG_MASK);
1624                         rs = ((qrem & LONG_MASK) << 32) | nl;
1625                         if (unsignedLongCompare(estProduct, rs))
1626                             qhat--;
1627                     }
1628                 }
1629             }
1630 
1631 
1632             // D4 Multiply and subtract
1633             int borrow;
1634             rem.value[limit - 1 + rem.offset] = 0;
1635             if(needRemainder)
1636                 borrow = mulsub(rem.value, divisor, qhat, dlen, limit - 1 + rem.offset);
1637             else
1638                 borrow = mulsubBorrow(rem.value, divisor, qhat, dlen, limit - 1 + rem.offset);
1639 
1640             // D5 Test remainder
1641             if (borrow + 0x80000000 > nh2) {
1642                 // D6 Add back
1643                 if(needRemainder)
1644                     divadd(divisor, rem.value, limit - 1 + 1 + rem.offset);
1645                 qhat--;
1646             }
1647 
1648             // Store the quotient digit
1649             q[(limit - 1)] = qhat;
1650         }
1651 
1652 
1653         if(needRemainder) {
1654             // D8 Unnormalize
1655             if (shift > 0)
1656                 rem.rightShift(shift);
1657             rem.normalize();
1658         }
1659         quotient.normalize();
1660         return needRemainder ? rem : null;
1661     }
1662 
1663     /**
1664      * Divide this MutableBigInteger by the divisor represented by positive long
1665      * value. The quotient will be placed into the provided quotient object &
1666      * the remainder object is returned.
1667      */
1668     private MutableBigInteger divideLongMagnitude(long ldivisor, MutableBigInteger quotient) {
1669         // Remainder starts as dividend with space for a leading zero
1670         MutableBigInteger rem = new MutableBigInteger(new int[intLen + 1]);
1671         System.arraycopy(value, offset, rem.value, 1, intLen);
1672         rem.intLen = intLen;
1673         rem.offset = 1;
1674 
1675         int nlen = rem.intLen;
1676 
1677         int limit = nlen - 2 + 1;
1678         if (quotient.value.length < limit) {
1679             quotient.value = new int[limit];
1680             quotient.offset = 0;
1681         }
1682         quotient.intLen = limit;
1683         int[] q = quotient.value;
1684 
1685         // D1 normalize the divisor
1686         int shift = Long.numberOfLeadingZeros(ldivisor);
1687         if (shift > 0) {
1688             ldivisor<<=shift;
1689             rem.leftShift(shift);
1690         }
1691 
1692         // Must insert leading 0 in rem if its length did not change
1693         if (rem.intLen == nlen) {
1694             rem.offset = 0;
1695             rem.value[0] = 0;
1696             rem.intLen++;
1697         }
1698 
1699         int dh = (int)(ldivisor >>> 32);
1700         long dhLong = dh & LONG_MASK;
1701         int dl = (int)(ldivisor & LONG_MASK);
1702 
1703         // D2 Initialize j
1704         for (int j = 0; j < limit; j++) {
1705             // D3 Calculate qhat
1706             // estimate qhat
1707             int qhat = 0;
1708             int qrem = 0;
1709             boolean skipCorrection = false;
1710             int nh = rem.value[j + rem.offset];
1711             int nh2 = nh + 0x80000000;
1712             int nm = rem.value[j + 1 + rem.offset];
1713 
1714             if (nh == dh) {
1715                 qhat = ~0;
1716                 qrem = nh + nm;
1717                 skipCorrection = qrem + 0x80000000 < nh2;
1718             } else {
1719                 long nChunk = (((long) nh) << 32) | (nm & LONG_MASK);
1720                 if (nChunk >= 0) {
1721                     qhat = (int) (nChunk / dhLong);
1722                     qrem = (int) (nChunk - (qhat * dhLong));
1723                 } else {
1724                     long tmp = divWord(nChunk, dh);
1725                     qhat =(int)(tmp & LONG_MASK);
1726                     qrem = (int)(tmp>>>32);
1727                 }
1728             }
1729 
1730             if (qhat == 0)
1731                 continue;
1732 
1733             if (!skipCorrection) { // Correct qhat
1734                 long nl = rem.value[j + 2 + rem.offset] & LONG_MASK;
1735                 long rs = ((qrem & LONG_MASK) << 32) | nl;
1736                 long estProduct = (dl & LONG_MASK) * (qhat & LONG_MASK);
1737 
1738                 if (unsignedLongCompare(estProduct, rs)) {
1739                     qhat--;
1740                     qrem = (int) ((qrem & LONG_MASK) + dhLong);
1741                     if ((qrem & LONG_MASK) >= dhLong) {
1742                         estProduct -= (dl & LONG_MASK);
1743                         rs = ((qrem & LONG_MASK) << 32) | nl;
1744                         if (unsignedLongCompare(estProduct, rs))
1745                             qhat--;
1746                     }
1747                 }
1748             }
1749 
1750             // D4 Multiply and subtract
1751             rem.value[j + rem.offset] = 0;
1752             int borrow = mulsubLong(rem.value, dh, dl, qhat,  j + rem.offset);
1753 
1754             // D5 Test remainder
1755             if (borrow + 0x80000000 > nh2) {
1756                 // D6 Add back
1757                 divaddLong(dh,dl, rem.value, j + 1 + rem.offset);
1758                 qhat--;
1759             }
1760 
1761             // Store the quotient digit
1762             q[j] = qhat;
1763         } // D7 loop on j
1764 
1765         // D8 Unnormalize
1766         if (shift > 0)
1767             rem.rightShift(shift);
1768 
1769         quotient.normalize();
1770         rem.normalize();
1771         return rem;
1772     }
1773 
1774     /**
1775      * A primitive used for division by long.
1776      * Specialized version of the method divadd.
1777      * dh is a high part of the divisor, dl is a low part
1778      */
1779     private int divaddLong(int dh, int dl, int[] result, int offset) {
1780         long carry = 0;
1781 
1782         long sum = (dl & LONG_MASK) + (result[1+offset] & LONG_MASK);
1783         result[1+offset] = (int)sum;
1784 
1785         sum = (dh & LONG_MASK) + (result[offset] & LONG_MASK) + carry;
1786         result[offset] = (int)sum;
1787         carry = sum >>> 32;
1788         return (int)carry;
1789     }
1790 
1791     /**
1792      * This method is used for division by long.
1793      * Specialized version of the method sulsub.
1794      * dh is a high part of the divisor, dl is a low part
1795      */
1796     private int mulsubLong(int[] q, int dh, int dl, int x, int offset) {
1797         long xLong = x & LONG_MASK;
1798         offset += 2;
1799         long product = (dl & LONG_MASK) * xLong;
1800         long difference = q[offset] - product;
1801         q[offset--] = (int)difference;
1802         long carry = (product >>> 32)
1803                  + (((difference & LONG_MASK) >
1804                      (((~(int)product) & LONG_MASK))) ? 1:0);
1805         product = (dh & LONG_MASK) * xLong + carry;
1806         difference = q[offset] - product;
1807         q[offset--] = (int)difference;
1808         carry = (product >>> 32)
1809                  + (((difference & LONG_MASK) >
1810                      (((~(int)product) & LONG_MASK))) ? 1:0);
1811         return (int)carry;
1812     }
1813 
1814     /**
1815      * Compare two longs as if they were unsigned.
1816      * Returns true iff one is bigger than two.
1817      */
1818     private boolean unsignedLongCompare(long one, long two) {
1819         return (one+Long.MIN_VALUE) > (two+Long.MIN_VALUE);
1820     }
1821 
1822     /**
1823      * This method divides a long quantity by an int to estimate
1824      * qhat for two multi precision numbers. It is used when
1825      * the signed value of n is less than zero.
1826      * Returns long value where high 32 bits contain remainder value and
1827      * low 32 bits contain quotient value.
1828      */
1829     static long divWord(long n, int d) {
1830         long dLong = d & LONG_MASK;
1831         long r;
1832         long q;
1833         if (dLong == 1) {
1834             q = (int)n;
1835             r = 0;
1836             return (r << 32) | (q & LONG_MASK);
1837         }
1838 
1839         // Approximate the quotient and remainder
1840         q = (n >>> 1) / (dLong >>> 1);
1841         r = n - q*dLong;
1842 
1843         // Correct the approximation
1844         while (r < 0) {
1845             r += dLong;
1846             q--;
1847         }
1848         while (r >= dLong) {
1849             r -= dLong;
1850             q++;
1851         }
1852         // n - q*dlong == r && 0 <= r <dLong, hence we're done.
1853         return (r << 32) | (q & LONG_MASK);
1854     }
1855 
1856     /**
1857      * Calculate GCD of this and b. This and b are changed by the computation.
1858      */
1859     MutableBigInteger hybridGCD(MutableBigInteger b) {
1860         // Use Euclid's algorithm until the numbers are approximately the
1861         // same length, then use the binary GCD algorithm to find the GCD.
1862         MutableBigInteger a = this;
1863         MutableBigInteger q = new MutableBigInteger();
1864 
1865         while (b.intLen != 0) {
1866             if (Math.abs(a.intLen - b.intLen) < 2)
1867                 return a.binaryGCD(b);
1868 
1869             MutableBigInteger r = a.divide(b, q);
1870             a = b;
1871             b = r;
1872         }
1873         return a;
1874     }
1875 
1876     /**
1877      * Calculate GCD of this and v.
1878      * Assumes that this and v are not zero.
1879      */
1880     private MutableBigInteger binaryGCD(MutableBigInteger v) {
1881         // Algorithm B from Knuth section 4.5.2
1882         MutableBigInteger u = this;
1883         MutableBigInteger r = new MutableBigInteger();
1884 
1885         // step B1
1886         int s1 = u.getLowestSetBit();
1887         int s2 = v.getLowestSetBit();
1888         int k = (s1 < s2) ? s1 : s2;
1889         if (k != 0) {
1890             u.rightShift(k);
1891             v.rightShift(k);
1892         }
1893 
1894         // step B2
1895         boolean uOdd = (k==s1);
1896         MutableBigInteger t = uOdd ? v: u;
1897         int tsign = uOdd ? -1 : 1;
1898 
1899         int lb;
1900         while ((lb = t.getLowestSetBit()) >= 0) {
1901             // steps B3 and B4
1902             t.rightShift(lb);
1903             // step B5
1904             if (tsign > 0)
1905                 u = t;
1906             else
1907                 v = t;
1908 
1909             // Special case one word numbers
1910             if (u.intLen < 2 && v.intLen < 2) {
1911                 int x = u.value[u.offset];
1912                 int y = v.value[v.offset];
1913                 x  = binaryGcd(x, y);
1914                 r.value[0] = x;
1915                 r.intLen = 1;
1916                 r.offset = 0;
1917                 if (k > 0)
1918                     r.leftShift(k);
1919                 return r;
1920             }
1921 
1922             // step B6
1923             if ((tsign = u.difference(v)) == 0)
1924                 break;
1925             t = (tsign >= 0) ? u : v;
1926         }
1927 
1928         if (k > 0)
1929             u.leftShift(k);
1930         return u;
1931     }
1932 
1933     /**
1934      * Calculate GCD of a and b interpreted as unsigned integers.
1935      */
1936     static int binaryGcd(int a, int b) {
1937         if (b==0)
1938             return a;
1939         if (a==0)
1940             return b;
1941 
1942         // Right shift a & b till their last bits equal to 1.
1943         int aZeros = Integer.numberOfTrailingZeros(a);
1944         int bZeros = Integer.numberOfTrailingZeros(b);
1945         a >>>= aZeros;
1946         b >>>= bZeros;
1947 
1948         int t = (aZeros < bZeros ? aZeros : bZeros);
1949 
1950         while (a != b) {
1951             if ((a+0x80000000) > (b+0x80000000)) {  // a > b as unsigned
1952                 a -= b;
1953                 a >>>= Integer.numberOfTrailingZeros(a);
1954             } else {
1955                 b -= a;
1956                 b >>>= Integer.numberOfTrailingZeros(b);
1957             }
1958         }
1959         return a<<t;
1960     }
1961 
1962     /**
1963      * Returns the modInverse of this mod p. This and p are not affected by
1964      * the operation.
1965      */
1966     MutableBigInteger mutableModInverse(MutableBigInteger p) {
1967         // Modulus is odd, use Schroeppel's algorithm
1968         if (p.isOdd())
1969             return modInverse(p);
1970 
1971         // Base and modulus are even, throw exception
1972         if (isEven())
1973             throw new ArithmeticException("BigInteger not invertible.");
1974 
1975         // Get even part of modulus expressed as a power of 2
1976         int powersOf2 = p.getLowestSetBit();
1977 
1978         // Construct odd part of modulus
1979         MutableBigInteger oddMod = new MutableBigInteger(p);
1980         oddMod.rightShift(powersOf2);
1981 
1982         if (oddMod.isOne())
1983             return modInverseMP2(powersOf2);
1984 
1985         // Calculate 1/a mod oddMod
1986         MutableBigInteger oddPart = modInverse(oddMod);
1987 
1988         // Calculate 1/a mod evenMod
1989         MutableBigInteger evenPart = modInverseMP2(powersOf2);
1990 
1991         // Combine the results using Chinese Remainder Theorem
1992         MutableBigInteger y1 = modInverseBP2(oddMod, powersOf2);
1993         MutableBigInteger y2 = oddMod.modInverseMP2(powersOf2);
1994 
1995         MutableBigInteger temp1 = new MutableBigInteger();
1996         MutableBigInteger temp2 = new MutableBigInteger();
1997         MutableBigInteger result = new MutableBigInteger();
1998 
1999         oddPart.leftShift(powersOf2);
2000         oddPart.multiply(y1, result);
2001 
2002         evenPart.multiply(oddMod, temp1);
2003         temp1.multiply(y2, temp2);
2004 
2005         result.add(temp2);
2006         return result.divide(p, temp1);
2007     }
2008 
2009     /*
2010      * Calculate the multiplicative inverse of this mod 2^k.
2011      */
2012     MutableBigInteger modInverseMP2(int k) {
2013         if (isEven())
2014             throw new ArithmeticException("Non-invertible. (GCD != 1)");
2015 
2016         if (k > 64)
2017             return euclidModInverse(k);
2018 
2019         int t = inverseMod32(value[offset+intLen-1]);
2020 
2021         if (k < 33) {
2022             t = (k == 32 ? t : t & ((1 << k) - 1));
2023             return new MutableBigInteger(t);
2024         }
2025 
2026         long pLong = (value[offset+intLen-1] & LONG_MASK);
2027         if (intLen > 1)
2028             pLong |=  ((long)value[offset+intLen-2] << 32);
2029         long tLong = t & LONG_MASK;
2030         tLong = tLong * (2 - pLong * tLong);  // 1 more Newton iter step
2031         tLong = (k == 64 ? tLong : tLong & ((1L << k) - 1));
2032 
2033         MutableBigInteger result = new MutableBigInteger(new int[2]);
2034         result.value[0] = (int)(tLong >>> 32);
2035         result.value[1] = (int)tLong;
2036         result.intLen = 2;
2037         result.normalize();
2038         return result;
2039     }
2040 
2041     /**
2042      * Returns the multiplicative inverse of val mod 2^32.  Assumes val is odd.
2043      */
2044     static int inverseMod32(int val) {
2045         // Newton's iteration!
2046         int t = val;
2047         t *= 2 - val*t;
2048         t *= 2 - val*t;
2049         t *= 2 - val*t;
2050         t *= 2 - val*t;
2051         return t;
2052     }
2053 
2054     /**
2055      * Calculate the multiplicative inverse of 2^k mod mod, where mod is odd.
2056      */
2057     static MutableBigInteger modInverseBP2(MutableBigInteger mod, int k) {
2058         // Copy the mod to protect original
2059         return fixup(new MutableBigInteger(1), new MutableBigInteger(mod), k);
2060     }
2061 
2062     /**
2063      * Calculate the multiplicative inverse of this mod mod, where mod is odd.
2064      * This and mod are not changed by the calculation.
2065      *
2066      * This method implements an algorithm due to Richard Schroeppel, that uses
2067      * the same intermediate representation as Montgomery Reduction
2068      * ("Montgomery Form").  The algorithm is described in an unpublished
2069      * manuscript entitled "Fast Modular Reciprocals."
2070      */
2071     private MutableBigInteger modInverse(MutableBigInteger mod) {
2072         MutableBigInteger p = new MutableBigInteger(mod);
2073         MutableBigInteger f = new MutableBigInteger(this);
2074         MutableBigInteger g = new MutableBigInteger(p);
2075         SignedMutableBigInteger c = new SignedMutableBigInteger(1);
2076         SignedMutableBigInteger d = new SignedMutableBigInteger();
2077         MutableBigInteger temp = null;
2078         SignedMutableBigInteger sTemp = null;
2079 
2080         int k = 0;
2081         // Right shift f k times until odd, left shift d k times
2082         if (f.isEven()) {
2083             int trailingZeros = f.getLowestSetBit();
2084             f.rightShift(trailingZeros);
2085             d.leftShift(trailingZeros);
2086             k = trailingZeros;
2087         }
2088 
2089         // The Almost Inverse Algorithm
2090         while(!f.isOne()) {
2091             // If gcd(f, g) != 1, number is not invertible modulo mod
2092             if (f.isZero())
2093                 throw new ArithmeticException("BigInteger not invertible.");
2094 
2095             // If f < g exchange f, g and c, d
2096             if (f.compare(g) < 0) {
2097                 temp = f; f = g; g = temp;
2098                 sTemp = d; d = c; c = sTemp;
2099             }
2100 
2101             // If f == g (mod 4)
2102             if (((f.value[f.offset + f.intLen - 1] ^
2103                  g.value[g.offset + g.intLen - 1]) & 3) == 0) {
2104                 f.subtract(g);
2105                 c.signedSubtract(d);
2106             } else { // If f != g (mod 4)
2107                 f.add(g);
2108                 c.signedAdd(d);
2109             }
2110 
2111             // Right shift f k times until odd, left shift d k times
2112             int trailingZeros = f.getLowestSetBit();
2113             f.rightShift(trailingZeros);
2114             d.leftShift(trailingZeros);
2115             k += trailingZeros;
2116         }
2117 
2118         while (c.sign < 0)
2119            c.signedAdd(p);
2120 
2121         return fixup(c, p, k);
2122     }
2123 
2124     /**
2125      * The Fixup Algorithm
2126      * Calculates X such that X = C * 2^(-k) (mod P)
2127      * Assumes C<P and P is odd.
2128      */
2129     static MutableBigInteger fixup(MutableBigInteger c, MutableBigInteger p,
2130                                                                       int k) {
2131         MutableBigInteger temp = new MutableBigInteger();
2132         // Set r to the multiplicative inverse of p mod 2^32
2133         int r = -inverseMod32(p.value[p.offset+p.intLen-1]);
2134 
2135         for(int i=0, numWords = k >> 5; i<numWords; i++) {
2136             // V = R * c (mod 2^j)
2137             int  v = r * c.value[c.offset + c.intLen-1];
2138             // c = c + (v * p)
2139             p.mul(v, temp);
2140             c.add(temp);
2141             // c = c / 2^j
2142             c.intLen--;
2143         }
2144         int numBits = k & 0x1f;
2145         if (numBits != 0) {
2146             // V = R * c (mod 2^j)
2147             int v = r * c.value[c.offset + c.intLen-1];
2148             v &= ((1<<numBits) - 1);
2149             // c = c + (v * p)
2150             p.mul(v, temp);
2151             c.add(temp);
2152             // c = c / 2^j
2153             c.rightShift(numBits);
2154         }
2155 
2156         // In theory, c may be greater than p at this point (Very rare!)
2157         while (c.compare(p) >= 0)
2158             c.subtract(p);
2159 
2160         return c;
2161     }
2162 
2163     /**
2164      * Uses the extended Euclidean algorithm to compute the modInverse of base
2165      * mod a modulus that is a power of 2. The modulus is 2^k.
2166      */
2167     MutableBigInteger euclidModInverse(int k) {
2168         MutableBigInteger b = new MutableBigInteger(1);
2169         b.leftShift(k);
2170         MutableBigInteger mod = new MutableBigInteger(b);
2171 
2172         MutableBigInteger a = new MutableBigInteger(this);
2173         MutableBigInteger q = new MutableBigInteger();
2174         MutableBigInteger r = b.divide(a, q);
2175 
2176         MutableBigInteger swapper = b;
2177         // swap b & r
2178         b = r;
2179         r = swapper;
2180 
2181         MutableBigInteger t1 = new MutableBigInteger(q);
2182         MutableBigInteger t0 = new MutableBigInteger(1);
2183         MutableBigInteger temp = new MutableBigInteger();
2184 
2185         while (!b.isOne()) {
2186             r = a.divide(b, q);
2187 
2188             if (r.intLen == 0)
2189                 throw new ArithmeticException("BigInteger not invertible.");
2190 
2191             swapper = r;
2192             a = swapper;
2193 
2194             if (q.intLen == 1)
2195                 t1.mul(q.value[q.offset], temp);
2196             else
2197                 q.multiply(t1, temp);
2198             swapper = q;
2199             q = temp;
2200             temp = swapper;
2201             t0.add(q);
2202 
2203             if (a.isOne())
2204                 return t0;
2205 
2206             r = b.divide(a, q);
2207 
2208             if (r.intLen == 0)
2209                 throw new ArithmeticException("BigInteger not invertible.");
2210 
2211             swapper = b;
2212             b =  r;
2213 
2214             if (q.intLen == 1)
2215                 t0.mul(q.value[q.offset], temp);
2216             else
2217                 q.multiply(t0, temp);
2218             swapper = q; q = temp; temp = swapper;
2219 
2220             t1.add(q);
2221         }
2222         mod.subtract(t1);
2223         return mod;
2224     }
2225 }