Source file src/math/big/int.go

Documentation: math/big

     1  // Copyright 2009 The Go Authors. All rights reserved.
     2  // Use of this source code is governed by a BSD-style
     3  // license that can be found in the LICENSE file.
     4  
     5  // This file implements signed multi-precision integers.
     6  
     7  package big
     8  
     9  import (
    10  	"fmt"
    11  	"io"
    12  	"math/rand"
    13  	"strings"
    14  )
    15  
    16  // An Int represents a signed multi-precision integer.
    17  // The zero value for an Int represents the value 0.
    18  //
    19  // Operations always take pointer arguments (*Int) rather
    20  // than Int values, and each unique Int value requires
    21  // its own unique *Int pointer. To "copy" an Int value,
    22  // an existing (or newly allocated) Int must be set to
    23  // a new value using the Int.Set method; shallow copies
    24  // of Ints are not supported and may lead to errors.
    25  type Int struct {
    26  	neg bool // sign
    27  	abs nat  // absolute value of the integer
    28  }
    29  
    30  var intOne = &Int{false, natOne}
    31  
    32  // Sign returns:
    33  //
    34  //	-1 if x <  0
    35  //	 0 if x == 0
    36  //	+1 if x >  0
    37  //
    38  func (x *Int) Sign() int {
    39  	if len(x.abs) == 0 {
    40  		return 0
    41  	}
    42  	if x.neg {
    43  		return -1
    44  	}
    45  	return 1
    46  }
    47  
    48  // SetInt64 sets z to x and returns z.
    49  func (z *Int) SetInt64(x int64) *Int {
    50  	neg := false
    51  	if x < 0 {
    52  		neg = true
    53  		x = -x
    54  	}
    55  	z.abs = z.abs.setUint64(uint64(x))
    56  	z.neg = neg
    57  	return z
    58  }
    59  
    60  // SetUint64 sets z to x and returns z.
    61  func (z *Int) SetUint64(x uint64) *Int {
    62  	z.abs = z.abs.setUint64(x)
    63  	z.neg = false
    64  	return z
    65  }
    66  
    67  // NewInt allocates and returns a new Int set to x.
    68  func NewInt(x int64) *Int {
    69  	return new(Int).SetInt64(x)
    70  }
    71  
    72  // Set sets z to x and returns z.
    73  func (z *Int) Set(x *Int) *Int {
    74  	if z != x {
    75  		z.abs = z.abs.set(x.abs)
    76  		z.neg = x.neg
    77  	}
    78  	return z
    79  }
    80  
    81  // Bits provides raw (unchecked but fast) access to x by returning its
    82  // absolute value as a little-endian Word slice. The result and x share
    83  // the same underlying array.
    84  // Bits is intended to support implementation of missing low-level Int
    85  // functionality outside this package; it should be avoided otherwise.
    86  func (x *Int) Bits() []Word {
    87  	return x.abs
    88  }
    89  
    90  // SetBits provides raw (unchecked but fast) access to z by setting its
    91  // value to abs, interpreted as a little-endian Word slice, and returning
    92  // z. The result and abs share the same underlying array.
    93  // SetBits is intended to support implementation of missing low-level Int
    94  // functionality outside this package; it should be avoided otherwise.
    95  func (z *Int) SetBits(abs []Word) *Int {
    96  	z.abs = nat(abs).norm()
    97  	z.neg = false
    98  	return z
    99  }
   100  
   101  // Abs sets z to |x| (the absolute value of x) and returns z.
   102  func (z *Int) Abs(x *Int) *Int {
   103  	z.Set(x)
   104  	z.neg = false
   105  	return z
   106  }
   107  
   108  // Neg sets z to -x and returns z.
   109  func (z *Int) Neg(x *Int) *Int {
   110  	z.Set(x)
   111  	z.neg = len(z.abs) > 0 && !z.neg // 0 has no sign
   112  	return z
   113  }
   114  
   115  // Add sets z to the sum x+y and returns z.
   116  func (z *Int) Add(x, y *Int) *Int {
   117  	neg := x.neg
   118  	if x.neg == y.neg {
   119  		// x + y == x + y
   120  		// (-x) + (-y) == -(x + y)
   121  		z.abs = z.abs.add(x.abs, y.abs)
   122  	} else {
   123  		// x + (-y) == x - y == -(y - x)
   124  		// (-x) + y == y - x == -(x - y)
   125  		if x.abs.cmp(y.abs) >= 0 {
   126  			z.abs = z.abs.sub(x.abs, y.abs)
   127  		} else {
   128  			neg = !neg
   129  			z.abs = z.abs.sub(y.abs, x.abs)
   130  		}
   131  	}
   132  	z.neg = len(z.abs) > 0 && neg // 0 has no sign
   133  	return z
   134  }
   135  
   136  // Sub sets z to the difference x-y and returns z.
   137  func (z *Int) Sub(x, y *Int) *Int {
   138  	neg := x.neg
   139  	if x.neg != y.neg {
   140  		// x - (-y) == x + y
   141  		// (-x) - y == -(x + y)
   142  		z.abs = z.abs.add(x.abs, y.abs)
   143  	} else {
   144  		// x - y == x - y == -(y - x)
   145  		// (-x) - (-y) == y - x == -(x - y)
   146  		if x.abs.cmp(y.abs) >= 0 {
   147  			z.abs = z.abs.sub(x.abs, y.abs)
   148  		} else {
   149  			neg = !neg
   150  			z.abs = z.abs.sub(y.abs, x.abs)
   151  		}
   152  	}
   153  	z.neg = len(z.abs) > 0 && neg // 0 has no sign
   154  	return z
   155  }
   156  
   157  // Mul sets z to the product x*y and returns z.
   158  func (z *Int) Mul(x, y *Int) *Int {
   159  	// x * y == x * y
   160  	// x * (-y) == -(x * y)
   161  	// (-x) * y == -(x * y)
   162  	// (-x) * (-y) == x * y
   163  	if x == y {
   164  		z.abs = z.abs.sqr(x.abs)
   165  		z.neg = false
   166  		return z
   167  	}
   168  	z.abs = z.abs.mul(x.abs, y.abs)
   169  	z.neg = len(z.abs) > 0 && x.neg != y.neg // 0 has no sign
   170  	return z
   171  }
   172  
   173  // MulRange sets z to the product of all integers
   174  // in the range [a, b] inclusively and returns z.
   175  // If a > b (empty range), the result is 1.
   176  func (z *Int) MulRange(a, b int64) *Int {
   177  	switch {
   178  	case a > b:
   179  		return z.SetInt64(1) // empty range
   180  	case a <= 0 && b >= 0:
   181  		return z.SetInt64(0) // range includes 0
   182  	}
   183  	// a <= b && (b < 0 || a > 0)
   184  
   185  	neg := false
   186  	if a < 0 {
   187  		neg = (b-a)&1 == 0
   188  		a, b = -b, -a
   189  	}
   190  
   191  	z.abs = z.abs.mulRange(uint64(a), uint64(b))
   192  	z.neg = neg
   193  	return z
   194  }
   195  
   196  // Binomial sets z to the binomial coefficient of (n, k) and returns z.
   197  func (z *Int) Binomial(n, k int64) *Int {
   198  	// reduce the number of multiplications by reducing k
   199  	if n/2 < k && k <= n {
   200  		k = n - k // Binomial(n, k) == Binomial(n, n-k)
   201  	}
   202  	var a, b Int
   203  	a.MulRange(n-k+1, n)
   204  	b.MulRange(1, k)
   205  	return z.Quo(&a, &b)
   206  }
   207  
   208  // Quo sets z to the quotient x/y for y != 0 and returns z.
   209  // If y == 0, a division-by-zero run-time panic occurs.
   210  // Quo implements truncated division (like Go); see QuoRem for more details.
   211  func (z *Int) Quo(x, y *Int) *Int {
   212  	z.abs, _ = z.abs.div(nil, x.abs, y.abs)
   213  	z.neg = len(z.abs) > 0 && x.neg != y.neg // 0 has no sign
   214  	return z
   215  }
   216  
   217  // Rem sets z to the remainder x%y for y != 0 and returns z.
   218  // If y == 0, a division-by-zero run-time panic occurs.
   219  // Rem implements truncated modulus (like Go); see QuoRem for more details.
   220  func (z *Int) Rem(x, y *Int) *Int {
   221  	_, z.abs = nat(nil).div(z.abs, x.abs, y.abs)
   222  	z.neg = len(z.abs) > 0 && x.neg // 0 has no sign
   223  	return z
   224  }
   225  
   226  // QuoRem sets z to the quotient x/y and r to the remainder x%y
   227  // and returns the pair (z, r) for y != 0.
   228  // If y == 0, a division-by-zero run-time panic occurs.
   229  //
   230  // QuoRem implements T-division and modulus (like Go):
   231  //
   232  //	q = x/y      with the result truncated to zero
   233  //	r = x - y*q
   234  //
   235  // (See Daan Leijen, ``Division and Modulus for Computer Scientists''.)
   236  // See DivMod for Euclidean division and modulus (unlike Go).
   237  //
   238  func (z *Int) QuoRem(x, y, r *Int) (*Int, *Int) {
   239  	z.abs, r.abs = z.abs.div(r.abs, x.abs, y.abs)
   240  	z.neg, r.neg = len(z.abs) > 0 && x.neg != y.neg, len(r.abs) > 0 && x.neg // 0 has no sign
   241  	return z, r
   242  }
   243  
   244  // Div sets z to the quotient x/y for y != 0 and returns z.
   245  // If y == 0, a division-by-zero run-time panic occurs.
   246  // Div implements Euclidean division (unlike Go); see DivMod for more details.
   247  func (z *Int) Div(x, y *Int) *Int {
   248  	y_neg := y.neg // z may be an alias for y
   249  	var r Int
   250  	z.QuoRem(x, y, &r)
   251  	if r.neg {
   252  		if y_neg {
   253  			z.Add(z, intOne)
   254  		} else {
   255  			z.Sub(z, intOne)
   256  		}
   257  	}
   258  	return z
   259  }
   260  
   261  // Mod sets z to the modulus x%y for y != 0 and returns z.
   262  // If y == 0, a division-by-zero run-time panic occurs.
   263  // Mod implements Euclidean modulus (unlike Go); see DivMod for more details.
   264  func (z *Int) Mod(x, y *Int) *Int {
   265  	y0 := y // save y
   266  	if z == y || alias(z.abs, y.abs) {
   267  		y0 = new(Int).Set(y)
   268  	}
   269  	var q Int
   270  	q.QuoRem(x, y, z)
   271  	if z.neg {
   272  		if y0.neg {
   273  			z.Sub(z, y0)
   274  		} else {
   275  			z.Add(z, y0)
   276  		}
   277  	}
   278  	return z
   279  }
   280  
   281  // DivMod sets z to the quotient x div y and m to the modulus x mod y
   282  // and returns the pair (z, m) for y != 0.
   283  // If y == 0, a division-by-zero run-time panic occurs.
   284  //
   285  // DivMod implements Euclidean division and modulus (unlike Go):
   286  //
   287  //	q = x div y  such that
   288  //	m = x - y*q  with 0 <= m < |y|
   289  //
   290  // (See Raymond T. Boute, ``The Euclidean definition of the functions
   291  // div and mod''. ACM Transactions on Programming Languages and
   292  // Systems (TOPLAS), 14(2):127-144, New York, NY, USA, 4/1992.
   293  // ACM press.)
   294  // See QuoRem for T-division and modulus (like Go).
   295  //
   296  func (z *Int) DivMod(x, y, m *Int) (*Int, *Int) {
   297  	y0 := y // save y
   298  	if z == y || alias(z.abs, y.abs) {
   299  		y0 = new(Int).Set(y)
   300  	}
   301  	z.QuoRem(x, y, m)
   302  	if m.neg {
   303  		if y0.neg {
   304  			z.Add(z, intOne)
   305  			m.Sub(m, y0)
   306  		} else {
   307  			z.Sub(z, intOne)
   308  			m.Add(m, y0)
   309  		}
   310  	}
   311  	return z, m
   312  }
   313  
   314  // Cmp compares x and y and returns:
   315  //
   316  //   -1 if x <  y
   317  //    0 if x == y
   318  //   +1 if x >  y
   319  //
   320  func (x *Int) Cmp(y *Int) (r int) {
   321  	// x cmp y == x cmp y
   322  	// x cmp (-y) == x
   323  	// (-x) cmp y == y
   324  	// (-x) cmp (-y) == -(x cmp y)
   325  	switch {
   326  	case x.neg == y.neg:
   327  		r = x.abs.cmp(y.abs)
   328  		if x.neg {
   329  			r = -r
   330  		}
   331  	case x.neg:
   332  		r = -1
   333  	default:
   334  		r = 1
   335  	}
   336  	return
   337  }
   338  
   339  // CmpAbs compares the absolute values of x and y and returns:
   340  //
   341  //   -1 if |x| <  |y|
   342  //    0 if |x| == |y|
   343  //   +1 if |x| >  |y|
   344  //
   345  func (x *Int) CmpAbs(y *Int) int {
   346  	return x.abs.cmp(y.abs)
   347  }
   348  
   349  // low32 returns the least significant 32 bits of x.
   350  func low32(x nat) uint32 {
   351  	if len(x) == 0 {
   352  		return 0
   353  	}
   354  	return uint32(x[0])
   355  }
   356  
   357  // low64 returns the least significant 64 bits of x.
   358  func low64(x nat) uint64 {
   359  	if len(x) == 0 {
   360  		return 0
   361  	}
   362  	v := uint64(x[0])
   363  	if _W == 32 && len(x) > 1 {
   364  		return uint64(x[1])<<32 | v
   365  	}
   366  	return v
   367  }
   368  
   369  // Int64 returns the int64 representation of x.
   370  // If x cannot be represented in an int64, the result is undefined.
   371  func (x *Int) Int64() int64 {
   372  	v := int64(low64(x.abs))
   373  	if x.neg {
   374  		v = -v
   375  	}
   376  	return v
   377  }
   378  
   379  // Uint64 returns the uint64 representation of x.
   380  // If x cannot be represented in a uint64, the result is undefined.
   381  func (x *Int) Uint64() uint64 {
   382  	return low64(x.abs)
   383  }
   384  
   385  // IsInt64 reports whether x can be represented as an int64.
   386  func (x *Int) IsInt64() bool {
   387  	if len(x.abs) <= 64/_W {
   388  		w := int64(low64(x.abs))
   389  		return w >= 0 || x.neg && w == -w
   390  	}
   391  	return false
   392  }
   393  
   394  // IsUint64 reports whether x can be represented as a uint64.
   395  func (x *Int) IsUint64() bool {
   396  	return !x.neg && len(x.abs) <= 64/_W
   397  }
   398  
   399  // SetString sets z to the value of s, interpreted in the given base,
   400  // and returns z and a boolean indicating success. The entire string
   401  // (not just a prefix) must be valid for success. If SetString fails,
   402  // the value of z is undefined but the returned value is nil.
   403  //
   404  // The base argument must be 0 or a value between 2 and MaxBase.
   405  // For base 0, the number prefix determines the actual base: A prefix of
   406  // ``0b'' or ``0B'' selects base 2, ``0'', ``0o'' or ``0O'' selects base 8,
   407  // and ``0x'' or ``0X'' selects base 16. Otherwise, the selected base is 10
   408  // and no prefix is accepted.
   409  //
   410  // For bases <= 36, lower and upper case letters are considered the same:
   411  // The letters 'a' to 'z' and 'A' to 'Z' represent digit values 10 to 35.
   412  // For bases > 36, the upper case letters 'A' to 'Z' represent the digit
   413  // values 36 to 61.
   414  //
   415  // For base 0, an underscore character ``_'' may appear between a base
   416  // prefix and an adjacent digit, and between successive digits; such
   417  // underscores do not change the value of the number.
   418  // Incorrect placement of underscores is reported as an error if there
   419  // are no other errors. If base != 0, underscores are not recognized
   420  // and act like any other character that is not a valid digit.
   421  //
   422  func (z *Int) SetString(s string, base int) (*Int, bool) {
   423  	return z.setFromScanner(strings.NewReader(s), base)
   424  }
   425  
   426  // setFromScanner implements SetString given an io.BytesScanner.
   427  // For documentation see comments of SetString.
   428  func (z *Int) setFromScanner(r io.ByteScanner, base int) (*Int, bool) {
   429  	if _, _, err := z.scan(r, base); err != nil {
   430  		return nil, false
   431  	}
   432  	// entire content must have been consumed
   433  	if _, err := r.ReadByte(); err != io.EOF {
   434  		return nil, false
   435  	}
   436  	return z, true // err == io.EOF => scan consumed all content of r
   437  }
   438  
   439  // SetBytes interprets buf as the bytes of a big-endian unsigned
   440  // integer, sets z to that value, and returns z.
   441  func (z *Int) SetBytes(buf []byte) *Int {
   442  	z.abs = z.abs.setBytes(buf)
   443  	z.neg = false
   444  	return z
   445  }
   446  
   447  // Bytes returns the absolute value of x as a big-endian byte slice.
   448  func (x *Int) Bytes() []byte {
   449  	buf := make([]byte, len(x.abs)*_S)
   450  	return buf[x.abs.bytes(buf):]
   451  }
   452  
   453  // BitLen returns the length of the absolute value of x in bits.
   454  // The bit length of 0 is 0.
   455  func (x *Int) BitLen() int {
   456  	return x.abs.bitLen()
   457  }
   458  
   459  // TrailingZeroBits returns the number of consecutive least significant zero
   460  // bits of |x|.
   461  func (x *Int) TrailingZeroBits() uint {
   462  	return x.abs.trailingZeroBits()
   463  }
   464  
   465  // Exp sets z = x**y mod |m| (i.e. the sign of m is ignored), and returns z.
   466  // If m == nil or m == 0, z = x**y unless y <= 0 then z = 1. If m > 0, y < 0,
   467  // and x and n are not relatively prime, z is unchanged and nil is returned.
   468  //
   469  // Modular exponentation of inputs of a particular size is not a
   470  // cryptographically constant-time operation.
   471  func (z *Int) Exp(x, y, m *Int) *Int {
   472  	// See Knuth, volume 2, section 4.6.3.
   473  	xWords := x.abs
   474  	if y.neg {
   475  		if m == nil || len(m.abs) == 0 {
   476  			return z.SetInt64(1)
   477  		}
   478  		// for y < 0: x**y mod m == (x**(-1))**|y| mod m
   479  		inverse := new(Int).ModInverse(x, m)
   480  		if inverse == nil {
   481  			return nil
   482  		}
   483  		xWords = inverse.abs
   484  	}
   485  	yWords := y.abs
   486  
   487  	var mWords nat
   488  	if m != nil {
   489  		mWords = m.abs // m.abs may be nil for m == 0
   490  	}
   491  
   492  	z.abs = z.abs.expNN(xWords, yWords, mWords)
   493  	z.neg = len(z.abs) > 0 && x.neg && len(yWords) > 0 && yWords[0]&1 == 1 // 0 has no sign
   494  	if z.neg && len(mWords) > 0 {
   495  		// make modulus result positive
   496  		z.abs = z.abs.sub(mWords, z.abs) // z == x**y mod |m| && 0 <= z < |m|
   497  		z.neg = false
   498  	}
   499  
   500  	return z
   501  }
   502  
   503  // GCD sets z to the greatest common divisor of a and b, which both must
   504  // be > 0, and returns z.
   505  // If x or y are not nil, GCD sets their value such that z = a*x + b*y.
   506  // If either a or b is <= 0, GCD sets z = x = y = 0.
   507  func (z *Int) GCD(x, y, a, b *Int) *Int {
   508  	if a.Sign() <= 0 || b.Sign() <= 0 {
   509  		z.SetInt64(0)
   510  		if x != nil {
   511  			x.SetInt64(0)
   512  		}
   513  		if y != nil {
   514  			y.SetInt64(0)
   515  		}
   516  		return z
   517  	}
   518  
   519  	return z.lehmerGCD(x, y, a, b)
   520  }
   521  
   522  // lehmerSimulate attempts to simulate several Euclidean update steps
   523  // using the leading digits of A and B.  It returns u0, u1, v0, v1
   524  // such that A and B can be updated as:
   525  //		A = u0*A + v0*B
   526  //		B = u1*A + v1*B
   527  // Requirements: A >= B and len(B.abs) >= 2
   528  // Since we are calculating with full words to avoid overflow,
   529  // we use 'even' to track the sign of the cosequences.
   530  // For even iterations: u0, v1 >= 0 && u1, v0 <= 0
   531  // For odd  iterations: u0, v1 <= 0 && u1, v0 >= 0
   532  func lehmerSimulate(A, B *Int) (u0, u1, v0, v1 Word, even bool) {
   533  	// initialize the digits
   534  	var a1, a2, u2, v2 Word
   535  
   536  	m := len(B.abs) // m >= 2
   537  	n := len(A.abs) // n >= m >= 2
   538  
   539  	// extract the top Word of bits from A and B
   540  	h := nlz(A.abs[n-1])
   541  	a1 = A.abs[n-1]<<h | A.abs[n-2]>>(_W-h)
   542  	// B may have implicit zero words in the high bits if the lengths differ
   543  	switch {
   544  	case n == m:
   545  		a2 = B.abs[n-1]<<h | B.abs[n-2]>>(_W-h)
   546  	case n == m+1:
   547  		a2 = B.abs[n-2] >> (_W - h)
   548  	default:
   549  		a2 = 0
   550  	}
   551  
   552  	// Since we are calculating with full words to avoid overflow,
   553  	// we use 'even' to track the sign of the cosequences.
   554  	// For even iterations: u0, v1 >= 0 && u1, v0 <= 0
   555  	// For odd  iterations: u0, v1 <= 0 && u1, v0 >= 0
   556  	// The first iteration starts with k=1 (odd).
   557  	even = false
   558  	// variables to track the cosequences
   559  	u0, u1, u2 = 0, 1, 0
   560  	v0, v1, v2 = 0, 0, 1
   561  
   562  	// Calculate the quotient and cosequences using Collins' stopping condition.
   563  	// Note that overflow of a Word is not possible when computing the remainder
   564  	// sequence and cosequences since the cosequence size is bounded by the input size.
   565  	// See section 4.2 of Jebelean for details.
   566  	for a2 >= v2 && a1-a2 >= v1+v2 {
   567  		q, r := a1/a2, a1%a2
   568  		a1, a2 = a2, r
   569  		u0, u1, u2 = u1, u2, u1+q*u2
   570  		v0, v1, v2 = v1, v2, v1+q*v2
   571  		even = !even
   572  	}
   573  	return
   574  }
   575  
   576  // lehmerUpdate updates the inputs A and B such that:
   577  //		A = u0*A + v0*B
   578  //		B = u1*A + v1*B
   579  // where the signs of u0, u1, v0, v1 are given by even
   580  // For even == true: u0, v1 >= 0 && u1, v0 <= 0
   581  // For even == false: u0, v1 <= 0 && u1, v0 >= 0
   582  // q, r, s, t are temporary variables to avoid allocations in the multiplication
   583  func lehmerUpdate(A, B, q, r, s, t *Int, u0, u1, v0, v1 Word, even bool) {
   584  
   585  	t.abs = t.abs.setWord(u0)
   586  	s.abs = s.abs.setWord(v0)
   587  	t.neg = !even
   588  	s.neg = even
   589  
   590  	t.Mul(A, t)
   591  	s.Mul(B, s)
   592  
   593  	r.abs = r.abs.setWord(u1)
   594  	q.abs = q.abs.setWord(v1)
   595  	r.neg = even
   596  	q.neg = !even
   597  
   598  	r.Mul(A, r)
   599  	q.Mul(B, q)
   600  
   601  	A.Add(t, s)
   602  	B.Add(r, q)
   603  }
   604  
   605  // euclidUpdate performs a single step of the Euclidean GCD algorithm
   606  // if extended is true, it also updates the cosequence Ua, Ub
   607  func euclidUpdate(A, B, Ua, Ub, q, r, s, t *Int, extended bool) {
   608  	q, r = q.QuoRem(A, B, r)
   609  
   610  	*A, *B, *r = *B, *r, *A
   611  
   612  	if extended {
   613  		// Ua, Ub = Ub, Ua - q*Ub
   614  		t.Set(Ub)
   615  		s.Mul(Ub, q)
   616  		Ub.Sub(Ua, s)
   617  		Ua.Set(t)
   618  	}
   619  }
   620  
   621  // lehmerGCD sets z to the greatest common divisor of a and b,
   622  // which both must be > 0, and returns z.
   623  // If x or y are not nil, their values are set such that z = a*x + b*y.
   624  // See Knuth, The Art of Computer Programming, Vol. 2, Section 4.5.2, Algorithm L.
   625  // This implementation uses the improved condition by Collins requiring only one
   626  // quotient and avoiding the possibility of single Word overflow.
   627  // See Jebelean, "Improving the multiprecision Euclidean algorithm",
   628  // Design and Implementation of Symbolic Computation Systems, pp 45-58.
   629  // The cosequences are updated according to Algorithm 10.45 from
   630  // Cohen et al. "Handbook of Elliptic and Hyperelliptic Curve Cryptography" pp 192.
   631  func (z *Int) lehmerGCD(x, y, a, b *Int) *Int {
   632  	var A, B, Ua, Ub *Int
   633  
   634  	A = new(Int).Set(a)
   635  	B = new(Int).Set(b)
   636  
   637  	extended := x != nil || y != nil
   638  
   639  	if extended {
   640  		// Ua (Ub) tracks how many times input a has been accumulated into A (B).
   641  		Ua = new(Int).SetInt64(1)
   642  		Ub = new(Int)
   643  	}
   644  
   645  	// temp variables for multiprecision update
   646  	q := new(Int)
   647  	r := new(Int)
   648  	s := new(Int)
   649  	t := new(Int)
   650  
   651  	// ensure A >= B
   652  	if A.abs.cmp(B.abs) < 0 {
   653  		A, B = B, A
   654  		Ub, Ua = Ua, Ub
   655  	}
   656  
   657  	// loop invariant A >= B
   658  	for len(B.abs) > 1 {
   659  		// Attempt to calculate in single-precision using leading words of A and B.
   660  		u0, u1, v0, v1, even := lehmerSimulate(A, B)
   661  
   662  		// multiprecision Step
   663  		if v0 != 0 {
   664  			// Simulate the effect of the single-precision steps using the cosequences.
   665  			// A = u0*A + v0*B
   666  			// B = u1*A + v1*B
   667  			lehmerUpdate(A, B, q, r, s, t, u0, u1, v0, v1, even)
   668  
   669  			if extended {
   670  				// Ua = u0*Ua + v0*Ub
   671  				// Ub = u1*Ua + v1*Ub
   672  				lehmerUpdate(Ua, Ub, q, r, s, t, u0, u1, v0, v1, even)
   673  			}
   674  
   675  		} else {
   676  			// Single-digit calculations failed to simulate any quotients.
   677  			// Do a standard Euclidean step.
   678  			euclidUpdate(A, B, Ua, Ub, q, r, s, t, extended)
   679  		}
   680  	}
   681  
   682  	if len(B.abs) > 0 {
   683  		// extended Euclidean algorithm base case if B is a single Word
   684  		if len(A.abs) > 1 {
   685  			// A is longer than a single Word, so one update is needed.
   686  			euclidUpdate(A, B, Ua, Ub, q, r, s, t, extended)
   687  		}
   688  		if len(B.abs) > 0 {
   689  			// A and B are both a single Word.
   690  			aWord, bWord := A.abs[0], B.abs[0]
   691  			if extended {
   692  				var ua, ub, va, vb Word
   693  				ua, ub = 1, 0
   694  				va, vb = 0, 1
   695  				even := true
   696  				for bWord != 0 {
   697  					q, r := aWord/bWord, aWord%bWord
   698  					aWord, bWord = bWord, r
   699  					ua, ub = ub, ua+q*ub
   700  					va, vb = vb, va+q*vb
   701  					even = !even
   702  				}
   703  
   704  				t.abs = t.abs.setWord(ua)
   705  				s.abs = s.abs.setWord(va)
   706  				t.neg = !even
   707  				s.neg = even
   708  
   709  				t.Mul(Ua, t)
   710  				s.Mul(Ub, s)
   711  
   712  				Ua.Add(t, s)
   713  			} else {
   714  				for bWord != 0 {
   715  					aWord, bWord = bWord, aWord%bWord
   716  				}
   717  			}
   718  			A.abs[0] = aWord
   719  		}
   720  	}
   721  
   722  	if y != nil {
   723  		// avoid aliasing b needed in the division below
   724  		if y == b {
   725  			B.Set(b)
   726  		} else {
   727  			B = b
   728  		}
   729  		// y = (z - a*x)/b
   730  		y.Mul(a, Ua) // y can safely alias a
   731  		y.Sub(A, y)
   732  		y.Div(y, B)
   733  	}
   734  
   735  	if x != nil {
   736  		*x = *Ua
   737  	}
   738  
   739  	*z = *A
   740  
   741  	return z
   742  }
   743  
   744  // Rand sets z to a pseudo-random number in [0, n) and returns z.
   745  //
   746  // As this uses the math/rand package, it must not be used for
   747  // security-sensitive work. Use crypto/rand.Int instead.
   748  func (z *Int) Rand(rnd *rand.Rand, n *Int) *Int {
   749  	z.neg = false
   750  	if n.neg || len(n.abs) == 0 {
   751  		z.abs = nil
   752  		return z
   753  	}
   754  	z.abs = z.abs.random(rnd, n.abs, n.abs.bitLen())
   755  	return z
   756  }
   757  
   758  // ModInverse sets z to the multiplicative inverse of g in the ring ℤ/nℤ
   759  // and returns z. If g and n are not relatively prime, g has no multiplicative
   760  // inverse in the ring ℤ/nℤ.  In this case, z is unchanged and the return value
   761  // is nil.
   762  func (z *Int) ModInverse(g, n *Int) *Int {
   763  	// GCD expects parameters a and b to be > 0.
   764  	if n.neg {
   765  		var n2 Int
   766  		n = n2.Neg(n)
   767  	}
   768  	if g.neg {
   769  		var g2 Int
   770  		g = g2.Mod(g, n)
   771  	}
   772  	var d, x Int
   773  	d.GCD(&x, nil, g, n)
   774  
   775  	// if and only if d==1, g and n are relatively prime
   776  	if d.Cmp(intOne) != 0 {
   777  		return nil
   778  	}
   779  
   780  	// x and y are such that g*x + n*y = 1, therefore x is the inverse element,
   781  	// but it may be negative, so convert to the range 0 <= z < |n|
   782  	if x.neg {
   783  		z.Add(&x, n)
   784  	} else {
   785  		z.Set(&x)
   786  	}
   787  	return z
   788  }
   789  
   790  // Jacobi returns the Jacobi symbol (x/y), either +1, -1, or 0.
   791  // The y argument must be an odd integer.
   792  func Jacobi(x, y *Int) int {
   793  	if len(y.abs) == 0 || y.abs[0]&1 == 0 {
   794  		panic(fmt.Sprintf("big: invalid 2nd argument to Int.Jacobi: need odd integer but got %s", y))
   795  	}
   796  
   797  	// We use the formulation described in chapter 2, section 2.4,
   798  	// "The Yacas Book of Algorithms":
   799  	// http://yacas.sourceforge.net/Algo.book.pdf
   800  
   801  	var a, b, c Int
   802  	a.Set(x)
   803  	b.Set(y)
   804  	j := 1
   805  
   806  	if b.neg {
   807  		if a.neg {
   808  			j = -1
   809  		}
   810  		b.neg = false
   811  	}
   812  
   813  	for {
   814  		if b.Cmp(intOne) == 0 {
   815  			return j
   816  		}
   817  		if len(a.abs) == 0 {
   818  			return 0
   819  		}
   820  		a.Mod(&a, &b)
   821  		if len(a.abs) == 0 {
   822  			return 0
   823  		}
   824  		// a > 0
   825  
   826  		// handle factors of 2 in 'a'
   827  		s := a.abs.trailingZeroBits()
   828  		if s&1 != 0 {
   829  			bmod8 := b.abs[0] & 7
   830  			if bmod8 == 3 || bmod8 == 5 {
   831  				j = -j
   832  			}
   833  		}
   834  		c.Rsh(&a, s) // a = 2^s*c
   835  
   836  		// swap numerator and denominator
   837  		if b.abs[0]&3 == 3 && c.abs[0]&3 == 3 {
   838  			j = -j
   839  		}
   840  		a.Set(&b)
   841  		b.Set(&c)
   842  	}
   843  }
   844  
   845  // modSqrt3Mod4 uses the identity
   846  //      (a^((p+1)/4))^2  mod p
   847  //   == u^(p+1)          mod p
   848  //   == u^2              mod p
   849  // to calculate the square root of any quadratic residue mod p quickly for 3
   850  // mod 4 primes.
   851  func (z *Int) modSqrt3Mod4Prime(x, p *Int) *Int {
   852  	e := new(Int).Add(p, intOne) // e = p + 1
   853  	e.Rsh(e, 2)                  // e = (p + 1) / 4
   854  	z.Exp(x, e, p)               // z = x^e mod p
   855  	return z
   856  }
   857  
   858  // modSqrt5Mod8 uses Atkin's observation that 2 is not a square mod p
   859  //   alpha ==  (2*a)^((p-5)/8)    mod p
   860  //   beta  ==  2*a*alpha^2        mod p  is a square root of -1
   861  //   b     ==  a*alpha*(beta-1)   mod p  is a square root of a
   862  // to calculate the square root of any quadratic residue mod p quickly for 5
   863  // mod 8 primes.
   864  func (z *Int) modSqrt5Mod8Prime(x, p *Int) *Int {
   865  	// p == 5 mod 8 implies p = e*8 + 5
   866  	// e is the quotient and 5 the remainder on division by 8
   867  	e := new(Int).Rsh(p, 3)  // e = (p - 5) / 8
   868  	tx := new(Int).Lsh(x, 1) // tx = 2*x
   869  	alpha := new(Int).Exp(tx, e, p)
   870  	beta := new(Int).Mul(alpha, alpha)
   871  	beta.Mod(beta, p)
   872  	beta.Mul(beta, tx)
   873  	beta.Mod(beta, p)
   874  	beta.Sub(beta, intOne)
   875  	beta.Mul(beta, x)
   876  	beta.Mod(beta, p)
   877  	beta.Mul(beta, alpha)
   878  	z.Mod(beta, p)
   879  	return z
   880  }
   881  
   882  // modSqrtTonelliShanks uses the Tonelli-Shanks algorithm to find the square
   883  // root of a quadratic residue modulo any prime.
   884  func (z *Int) modSqrtTonelliShanks(x, p *Int) *Int {
   885  	// Break p-1 into s*2^e such that s is odd.
   886  	var s Int
   887  	s.Sub(p, intOne)
   888  	e := s.abs.trailingZeroBits()
   889  	s.Rsh(&s, e)
   890  
   891  	// find some non-square n
   892  	var n Int
   893  	n.SetInt64(2)
   894  	for Jacobi(&n, p) != -1 {
   895  		n.Add(&n, intOne)
   896  	}
   897  
   898  	// Core of the Tonelli-Shanks algorithm. Follows the description in
   899  	// section 6 of "Square roots from 1; 24, 51, 10 to Dan Shanks" by Ezra
   900  	// Brown:
   901  	// https://www.maa.org/sites/default/files/pdf/upload_library/22/Polya/07468342.di020786.02p0470a.pdf
   902  	var y, b, g, t Int
   903  	y.Add(&s, intOne)
   904  	y.Rsh(&y, 1)
   905  	y.Exp(x, &y, p)  // y = x^((s+1)/2)
   906  	b.Exp(x, &s, p)  // b = x^s
   907  	g.Exp(&n, &s, p) // g = n^s
   908  	r := e
   909  	for {
   910  		// find the least m such that ord_p(b) = 2^m
   911  		var m uint
   912  		t.Set(&b)
   913  		for t.Cmp(intOne) != 0 {
   914  			t.Mul(&t, &t).Mod(&t, p)
   915  			m++
   916  		}
   917  
   918  		if m == 0 {
   919  			return z.Set(&y)
   920  		}
   921  
   922  		t.SetInt64(0).SetBit(&t, int(r-m-1), 1).Exp(&g, &t, p)
   923  		// t = g^(2^(r-m-1)) mod p
   924  		g.Mul(&t, &t).Mod(&g, p) // g = g^(2^(r-m)) mod p
   925  		y.Mul(&y, &t).Mod(&y, p)
   926  		b.Mul(&b, &g).Mod(&b, p)
   927  		r = m
   928  	}
   929  }
   930  
   931  // ModSqrt sets z to a square root of x mod p if such a square root exists, and
   932  // returns z. The modulus p must be an odd prime. If x is not a square mod p,
   933  // ModSqrt leaves z unchanged and returns nil. This function panics if p is
   934  // not an odd integer.
   935  func (z *Int) ModSqrt(x, p *Int) *Int {
   936  	switch Jacobi(x, p) {
   937  	case -1:
   938  		return nil // x is not a square mod p
   939  	case 0:
   940  		return z.SetInt64(0) // sqrt(0) mod p = 0
   941  	case 1:
   942  		break
   943  	}
   944  	if x.neg || x.Cmp(p) >= 0 { // ensure 0 <= x < p
   945  		x = new(Int).Mod(x, p)
   946  	}
   947  
   948  	switch {
   949  	case p.abs[0]%4 == 3:
   950  		// Check whether p is 3 mod 4, and if so, use the faster algorithm.
   951  		return z.modSqrt3Mod4Prime(x, p)
   952  	case p.abs[0]%8 == 5:
   953  		// Check whether p is 5 mod 8, use Atkin's algorithm.
   954  		return z.modSqrt5Mod8Prime(x, p)
   955  	default:
   956  		// Otherwise, use Tonelli-Shanks.
   957  		return z.modSqrtTonelliShanks(x, p)
   958  	}
   959  }
   960  
   961  // Lsh sets z = x << n and returns z.
   962  func (z *Int) Lsh(x *Int, n uint) *Int {
   963  	z.abs = z.abs.shl(x.abs, n)
   964  	z.neg = x.neg
   965  	return z
   966  }
   967  
   968  // Rsh sets z = x >> n and returns z.
   969  func (z *Int) Rsh(x *Int, n uint) *Int {
   970  	if x.neg {
   971  		// (-x) >> s == ^(x-1) >> s == ^((x-1) >> s) == -(((x-1) >> s) + 1)
   972  		t := z.abs.sub(x.abs, natOne) // no underflow because |x| > 0
   973  		t = t.shr(t, n)
   974  		z.abs = t.add(t, natOne)
   975  		z.neg = true // z cannot be zero if x is negative
   976  		return z
   977  	}
   978  
   979  	z.abs = z.abs.shr(x.abs, n)
   980  	z.neg = false
   981  	return z
   982  }
   983  
   984  // Bit returns the value of the i'th bit of x. That is, it
   985  // returns (x>>i)&1. The bit index i must be >= 0.
   986  func (x *Int) Bit(i int) uint {
   987  	if i == 0 {
   988  		// optimization for common case: odd/even test of x
   989  		if len(x.abs) > 0 {
   990  			return uint(x.abs[0] & 1) // bit 0 is same for -x
   991  		}
   992  		return 0
   993  	}
   994  	if i < 0 {
   995  		panic("negative bit index")
   996  	}
   997  	if x.neg {
   998  		t := nat(nil).sub(x.abs, natOne)
   999  		return t.bit(uint(i)) ^ 1
  1000  	}
  1001  
  1002  	return x.abs.bit(uint(i))
  1003  }
  1004  
  1005  // SetBit sets z to x, with x's i'th bit set to b (0 or 1).
  1006  // That is, if b is 1 SetBit sets z = x | (1 << i);
  1007  // if b is 0 SetBit sets z = x &^ (1 << i). If b is not 0 or 1,
  1008  // SetBit will panic.
  1009  func (z *Int) SetBit(x *Int, i int, b uint) *Int {
  1010  	if i < 0 {
  1011  		panic("negative bit index")
  1012  	}
  1013  	if x.neg {
  1014  		t := z.abs.sub(x.abs, natOne)
  1015  		t = t.setBit(t, uint(i), b^1)
  1016  		z.abs = t.add(t, natOne)
  1017  		z.neg = len(z.abs) > 0
  1018  		return z
  1019  	}
  1020  	z.abs = z.abs.setBit(x.abs, uint(i), b)
  1021  	z.neg = false
  1022  	return z
  1023  }
  1024  
  1025  // And sets z = x & y and returns z.
  1026  func (z *Int) And(x, y *Int) *Int {
  1027  	if x.neg == y.neg {
  1028  		if x.neg {
  1029  			// (-x) & (-y) == ^(x-1) & ^(y-1) == ^((x-1) | (y-1)) == -(((x-1) | (y-1)) + 1)
  1030  			x1 := nat(nil).sub(x.abs, natOne)
  1031  			y1 := nat(nil).sub(y.abs, natOne)
  1032  			z.abs = z.abs.add(z.abs.or(x1, y1), natOne)
  1033  			z.neg = true // z cannot be zero if x and y are negative
  1034  			return z
  1035  		}
  1036  
  1037  		// x & y == x & y
  1038  		z.abs = z.abs.and(x.abs, y.abs)
  1039  		z.neg = false
  1040  		return z
  1041  	}
  1042  
  1043  	// x.neg != y.neg
  1044  	if x.neg {
  1045  		x, y = y, x // & is symmetric
  1046  	}
  1047  
  1048  	// x & (-y) == x & ^(y-1) == x &^ (y-1)
  1049  	y1 := nat(nil).sub(y.abs, natOne)
  1050  	z.abs = z.abs.andNot(x.abs, y1)
  1051  	z.neg = false
  1052  	return z
  1053  }
  1054  
  1055  // AndNot sets z = x &^ y and returns z.
  1056  func (z *Int) AndNot(x, y *Int) *Int {
  1057  	if x.neg == y.neg {
  1058  		if x.neg {
  1059  			// (-x) &^ (-y) == ^(x-1) &^ ^(y-1) == ^(x-1) & (y-1) == (y-1) &^ (x-1)
  1060  			x1 := nat(nil).sub(x.abs, natOne)
  1061  			y1 := nat(nil).sub(y.abs, natOne)
  1062  			z.abs = z.abs.andNot(y1, x1)
  1063  			z.neg = false
  1064  			return z
  1065  		}
  1066  
  1067  		// x &^ y == x &^ y
  1068  		z.abs = z.abs.andNot(x.abs, y.abs)
  1069  		z.neg = false
  1070  		return z
  1071  	}
  1072  
  1073  	if x.neg {
  1074  		// (-x) &^ y == ^(x-1) &^ y == ^(x-1) & ^y == ^((x-1) | y) == -(((x-1) | y) + 1)
  1075  		x1 := nat(nil).sub(x.abs, natOne)
  1076  		z.abs = z.abs.add(z.abs.or(x1, y.abs), natOne)
  1077  		z.neg = true // z cannot be zero if x is negative and y is positive
  1078  		return z
  1079  	}
  1080  
  1081  	// x &^ (-y) == x &^ ^(y-1) == x & (y-1)
  1082  	y1 := nat(nil).sub(y.abs, natOne)
  1083  	z.abs = z.abs.and(x.abs, y1)
  1084  	z.neg = false
  1085  	return z
  1086  }
  1087  
  1088  // Or sets z = x | y and returns z.
  1089  func (z *Int) Or(x, y *Int) *Int {
  1090  	if x.neg == y.neg {
  1091  		if x.neg {
  1092  			// (-x) | (-y) == ^(x-1) | ^(y-1) == ^((x-1) & (y-1)) == -(((x-1) & (y-1)) + 1)
  1093  			x1 := nat(nil).sub(x.abs, natOne)
  1094  			y1 := nat(nil).sub(y.abs, natOne)
  1095  			z.abs = z.abs.add(z.abs.and(x1, y1), natOne)
  1096  			z.neg = true // z cannot be zero if x and y are negative
  1097  			return z
  1098  		}
  1099  
  1100  		// x | y == x | y
  1101  		z.abs = z.abs.or(x.abs, y.abs)
  1102  		z.neg = false
  1103  		return z
  1104  	}
  1105  
  1106  	// x.neg != y.neg
  1107  	if x.neg {
  1108  		x, y = y, x // | is symmetric
  1109  	}
  1110  
  1111  	// x | (-y) == x | ^(y-1) == ^((y-1) &^ x) == -(^((y-1) &^ x) + 1)
  1112  	y1 := nat(nil).sub(y.abs, natOne)
  1113  	z.abs = z.abs.add(z.abs.andNot(y1, x.abs), natOne)
  1114  	z.neg = true // z cannot be zero if one of x or y is negative
  1115  	return z
  1116  }
  1117  
  1118  // Xor sets z = x ^ y and returns z.
  1119  func (z *Int) Xor(x, y *Int) *Int {
  1120  	if x.neg == y.neg {
  1121  		if x.neg {
  1122  			// (-x) ^ (-y) == ^(x-1) ^ ^(y-1) == (x-1) ^ (y-1)
  1123  			x1 := nat(nil).sub(x.abs, natOne)
  1124  			y1 := nat(nil).sub(y.abs, natOne)
  1125  			z.abs = z.abs.xor(x1, y1)
  1126  			z.neg = false
  1127  			return z
  1128  		}
  1129  
  1130  		// x ^ y == x ^ y
  1131  		z.abs = z.abs.xor(x.abs, y.abs)
  1132  		z.neg = false
  1133  		return z
  1134  	}
  1135  
  1136  	// x.neg != y.neg
  1137  	if x.neg {
  1138  		x, y = y, x // ^ is symmetric
  1139  	}
  1140  
  1141  	// x ^ (-y) == x ^ ^(y-1) == ^(x ^ (y-1)) == -((x ^ (y-1)) + 1)
  1142  	y1 := nat(nil).sub(y.abs, natOne)
  1143  	z.abs = z.abs.add(z.abs.xor(x.abs, y1), natOne)
  1144  	z.neg = true // z cannot be zero if only one of x or y is negative
  1145  	return z
  1146  }
  1147  
  1148  // Not sets z = ^x and returns z.
  1149  func (z *Int) Not(x *Int) *Int {
  1150  	if x.neg {
  1151  		// ^(-x) == ^(^(x-1)) == x-1
  1152  		z.abs = z.abs.sub(x.abs, natOne)
  1153  		z.neg = false
  1154  		return z
  1155  	}
  1156  
  1157  	// ^x == -x-1 == -(x+1)
  1158  	z.abs = z.abs.add(x.abs, natOne)
  1159  	z.neg = true // z cannot be zero if x is positive
  1160  	return z
  1161  }
  1162  
  1163  // Sqrt sets z to ⌊√x⌋, the largest integer such that z² ≤ x, and returns z.
  1164  // It panics if x is negative.
  1165  func (z *Int) Sqrt(x *Int) *Int {
  1166  	if x.neg {
  1167  		panic("square root of negative number")
  1168  	}
  1169  	z.neg = false
  1170  	z.abs = z.abs.sqrt(x.abs)
  1171  	return z
  1172  }
  1173  

View as plain text