Botan 1.10.17
bigint.h
Go to the documentation of this file.
1/*
2* BigInt
3* (C) 1999-2008 Jack Lloyd
4* 2007 FlexSecure
5*
6* Distributed under the terms of the Botan license
7*/
8
9#ifndef BOTAN_BIGINT_H__
10#define BOTAN_BIGINT_H__
11
12#include <botan/rng.h>
13#include <botan/secmem.h>
14#include <botan/mp_types.h>
15#include <iosfwd>
16
17namespace Botan {
18
19/**
20* Arbitrary precision integer
21*/
22class BOTAN_DLL BigInt
23 {
24 public:
25 /**
26 * Base enumerator for encoding and decoding
27 */
28 enum Base { Octal = 8, Decimal = 10, Hexadecimal = 16, Binary = 256 };
29
30 /**
31 * Sign symbol definitions for positive and negative numbers
32 */
33 enum Sign { Negative = 0, Positive = 1 };
34
35 /**
36 * Number types (currently only power-of-2 supported)
37 */
39
40 /**
41 * DivideByZero Exception
42 */
43 struct BOTAN_DLL DivideByZero : public Exception
44 { DivideByZero() : Exception("BigInt divide by zero") {} };
45
46 /**
47 * += operator
48 * @param y the BigInt to add to this
49 */
50 BigInt& operator+=(const BigInt& y);
51
52 /**
53 * -= operator
54 * @param y the BigInt to subtract from this
55 */
56 BigInt& operator-=(const BigInt& y);
57
58 /**
59 * *= operator
60 * @param y the BigInt to multiply with this
61 */
62 BigInt& operator*=(const BigInt& y);
63
64 /**
65 * /= operator
66 * @param y the BigInt to divide this by
67 */
68 BigInt& operator/=(const BigInt& y);
69
70 /**
71 * Modulo operator
72 * @param y the modulus to reduce this by
73 */
74 BigInt& operator%=(const BigInt& y);
75
76 /**
77 * Modulo operator
78 * @param y the modulus (word) to reduce this by
79 */
80 word operator%=(word y);
81
82 /**
83 * Left shift operator
84 * @param shift the number of bits to shift this left by
85 */
86 BigInt& operator<<=(size_t shift);
87
88 /**
89 * Right shift operator
90 * @param shift the number of bits to shift this right by
91 */
92 BigInt& operator>>=(size_t shift);
93
94 /**
95 * Increment operator
96 */
97 BigInt& operator++() { return (*this += 1); }
98
99 /**
100 * Decrement operator
101 */
102 BigInt& operator--() { return (*this -= 1); }
103
104 /**
105 * Postfix increment operator
106 */
107 BigInt operator++(int) { BigInt x = (*this); ++(*this); return x; }
108
109 /**
110 * Postfix decrement operator
111 */
112 BigInt operator--(int) { BigInt x = (*this); --(*this); return x; }
113
114 /**
115 * Unary negation operator
116 * @return negative this
117 */
118 BigInt operator-() const;
119
120 /**
121 * ! operator
122 * @return true iff this is zero, otherwise false
123 */
124 bool operator !() const { return (!is_nonzero()); }
125
126 /**
127 * [] operator (array access)
128 * @param i a word index
129 * @return the word at index i
130 */
131 word& operator[](size_t i) { return reg[i]; }
132
133 /**
134 * [] operator (array access)
135 * @param i a word index
136 * @return the word at index i
137 */
138 const word& operator[](size_t i) const { return reg[i]; }
139
140 /**
141 * Zeroize the BigInt
142 */
143 void clear() { zeroise(reg); }
144
145 /**
146 * Compare this to another BigInt
147 * @param n the BigInt value to compare with
148 * @param check_signs include sign in comparison?
149 * @result if (this<n) return -1, if (this>n) return 1, if both
150 * values are identical return 0 [like Perl's <=> operator]
151 */
152 s32bit cmp(const BigInt& n, bool check_signs = true) const;
153
154 /**
155 * Test if the integer has an even value
156 * @result true if the integer is even, false otherwise
157 */
158 bool is_even() const { return (get_bit(0) == 0); }
159
160 /**
161 * Test if the integer has an odd value
162 * @result true if the integer is odd, false otherwise
163 */
164 bool is_odd() const { return (get_bit(0) == 1); }
165
166 /**
167 * Test if the integer is not zero
168 * @result true if the integer is non-zero, false otherwise
169 */
170 bool is_nonzero() const { return (!is_zero()); }
171
172 /**
173 * Test if the integer is zero
174 * @result true if the integer is zero, false otherwise
175 */
176 bool is_zero() const
177 {
178 const size_t sw = sig_words();
179
180 for(size_t i = 0; i != sw; ++i)
181 if(reg[i])
182 return false;
183 return true;
184 }
185
186 /**
187 * Set bit at specified position
188 * @param n bit position to set
189 */
190 void set_bit(size_t n);
191
192 /**
193 * Clear bit at specified position
194 * @param n bit position to clear
195 */
196 void clear_bit(size_t n);
197
198 /**
199 * Clear all but the lowest n bits
200 * @param n amount of bits to keep
201 */
202 void mask_bits(size_t n);
203
204 /**
205 * Return bit value at specified position
206 * @param n the bit offset to test
207 * @result true, if the bit at position n is set, false otherwise
208 */
209 bool get_bit(size_t n) const;
210
211 /**
212 * Return (a maximum of) 32 bits of the complete value
213 * @param offset the offset to start extracting
214 * @param length amount of bits to extract (starting at offset)
215 * @result the integer extracted from the register starting at
216 * offset with specified length
217 */
218 u32bit get_substring(size_t offset, size_t length) const;
219
220 /**
221 * Convert this value into a u32bit, if it is in the range
222 * [0 ... 2**32-1], or otherwise throw an exception.
223 * @result the value as a u32bit if conversion is possible
224 */
225 u32bit to_u32bit() const;
226
227 /**
228 * @param n the offset to get a byte from
229 * @result byte at offset n
230 */
231 byte byte_at(size_t n) const;
232
233 /**
234 * Return the word at a specified position of the internal register
235 * @param n position in the register
236 * @return value at position n
237 */
238 word word_at(size_t n) const
239 { return ((n < size()) ? reg[n] : 0); }
240
241 /**
242 * Tests if the sign of the integer is negative
243 * @result true, iff the integer has a negative sign
244 */
245 bool is_negative() const { return (sign() == Negative); }
246
247 /**
248 * Tests if the sign of the integer is positive
249 * @result true, iff the integer has a positive sign
250 */
251 bool is_positive() const { return (sign() == Positive); }
252
253 /**
254 * Return the sign of the integer
255 * @result the sign of the integer
256 */
257 Sign sign() const { return (signedness); }
258
259 /**
260 * @result the opposite sign of the represented integer value
261 */
262 Sign reverse_sign() const;
263
264 /**
265 * Flip the sign of this BigInt
266 */
267 void flip_sign();
268
269 /**
270 * Set sign of the integer
271 * @param sign new Sign to set
272 */
273 void set_sign(Sign sign);
274
275 /**
276 * @result absolute (positive) value of this
277 */
278 BigInt abs() const;
279
280 /**
281 * Give size of internal register
282 * @result size of internal register in words
283 */
284 size_t size() const { return get_reg().size(); }
285
286 /**
287 * Return how many words we need to hold this value
288 * @result significant words of the represented integer value
289 */
290 size_t sig_words() const
291 {
292 const word* x = &reg[0];
293 size_t sig = reg.size();
294
295 while(sig && (x[sig-1] == 0))
296 sig--;
297 return sig;
298 }
299
300 /**
301 * Give byte length of the integer
302 * @result byte length of the represented integer value
303 */
304 size_t bytes() const;
305
306 /**
307 * Get the bit length of the integer
308 * @result bit length of the represented integer value
309 */
310 size_t bits() const;
311
312 /**
313 * Return a pointer to the big integer word register
314 * @result a pointer to the start of the internal register of
315 * the integer value
316 */
317 const word* data() const { return &reg[0]; }
318
319 /**
320 * return a reference to the internal register containing the value
321 * @result a reference to the word-array (SecureVector<word>)
322 * with the internal register value (containing the integer
323 * value)
324 */
325 SecureVector<word>& get_reg() { return reg; }
326
327 /**
328 * return a const reference to the internal register containing the value
329 * @result a const reference to the word-array (SecureVector<word>)
330 * with the internal register value (containing the integer value)
331 */
332 const SecureVector<word>& get_reg() const { return reg; }
333
334 /**
335 * Assign using a plain word array
336 */
337 void assign(const word x[], size_t length)
338 {
339 reg.resize(length);
340 copy_mem(&reg[0], x, length);
341 }
342
343 /**
344 * Increase internal register buffer by n words
345 * @param n increase by n words
346 */
347 void grow_reg(size_t n);
348
349 void grow_to(size_t n);
350
351 /**
352 * Fill BigInt with a random number with size of bitsize
353 * @param rng the random number generator to use
354 * @param bitsize number of bits the created random value should have
355 */
356 void randomize(RandomNumberGenerator& rng, size_t bitsize = 0);
357
358 /**
359 * Store BigInt-value in a given byte array
360 * @param buf destination byte array for the integer value
361 */
362 void binary_encode(byte buf[]) const;
363
364 /**
365 * Read integer value from a byte array with given size
366 * @param buf byte array buffer containing the integer
367 * @param length size of buf
368 */
369 void binary_decode(const byte buf[], size_t length);
370
371 /**
372 * Read integer value from a byte array (MemoryRegion<byte>)
373 * @param buf the array to load from
374 */
375 void binary_decode(const MemoryRegion<byte>& buf);
376
377 /**
378 * @param base the base to measure the size for
379 * @return size of this integer in base base
380 */
381 size_t encoded_size(Base base = Binary) const;
382
383 /**
384 * @param rng a random number generator
385 * @param min the minimum value
386 * @param max the maximum value
387 * @return random integer between min and max
388 */
389 static BigInt random_integer(RandomNumberGenerator& rng,
390 const BigInt& min,
391 const BigInt& max);
392
393 /**
394 * Encode the integer value from a BigInt to a SecureVector of bytes
395 * @param n the BigInt to use as integer source
396 * @param base number-base of resulting byte array representation
397 * @result SecureVector of bytes containing the integer with given base
398 */
399 static SecureVector<byte> encode(const BigInt& n, Base base = Binary);
400
401 /**
402 * Encode the integer value from a BigInt to a byte array
403 * @param buf destination byte array for the encoded integer
404 * value with given base
405 * @param n the BigInt to use as integer source
406 * @param base number-base of resulting byte array representation
407 */
408 static void encode(byte buf[], const BigInt& n, Base base = Binary);
409
410 /**
411 * Create a BigInt from an integer in a byte array
412 * @param buf the binary value to load
413 * @param length size of buf
414 * @param base number-base of the integer in buf
415 * @result BigInt representing the integer in the byte array
416 */
417 static BigInt decode(const byte buf[], size_t length,
418 Base base = Binary);
419
420 /**
421 * Create a BigInt from an integer in a byte array
422 * @param buf the binary value to load
423 * @param base number-base of the integer in buf
424 * @result BigInt representing the integer in the byte array
425 */
426 static BigInt decode(const MemoryRegion<byte>& buf,
427 Base base = Binary);
428
429 /**
430 * Encode a BigInt to a byte array according to IEEE 1363
431 * @param n the BigInt to encode
432 * @param bytes the length of the resulting SecureVector<byte>
433 * @result a SecureVector<byte> containing the encoded BigInt
434 */
435 static SecureVector<byte> encode_1363(const BigInt& n, size_t bytes);
436
437 /**
438 * Swap this value with another
439 * @param other BigInt to swap values with
440 */
441 void swap(BigInt& other);
442
443 /**
444 * Create empty BigInt
445 */
446 BigInt() { signedness = Positive; }
447
448 /**
449 * Create BigInt from 64 bit integer
450 * @param n initial value of this BigInt
451 */
452 BigInt(u64bit n);
453
454 /**
455 * Copy Constructor
456 * @param other the BigInt to copy
457 */
458 BigInt(const BigInt& other);
459
460 /**
461 * Create BigInt from a string. If the string starts with 0x the
462 * rest of the string will be interpreted as hexadecimal digits.
463 * If the string starts with 0 and the second character is NOT an
464 * 'x' the string will be interpreted as octal digits. If the
465 * string starts with non-zero digit, it will be interpreted as a
466 * decimal number.
467 *
468 * @param str the string to parse for an integer value
469 */
470 BigInt(const std::string& str);
471
472 /**
473 * Create a BigInt from an integer in a byte array
474 * @param buf the byte array holding the value
475 * @param length size of buf
476 * @param base is the number base of the integer in buf
477 */
478 BigInt(const byte buf[], size_t length, Base base = Binary);
479
480 /**
481 * Create a random BigInt of the specified size
482 * @param rng random number generator
483 * @param bits size in bits
484 */
485 BigInt(RandomNumberGenerator& rng, size_t bits);
486
487 /**
488 * Create BigInt of specified size, all zeros
489 * @param sign the sign
490 * @param n size of the internal register in words
491 */
492 BigInt(Sign sign, size_t n);
493
494 /**
495 * Create a number of the specified type and size
496 * @param type the type of number to create. For Power2,
497 * will create the integer 2^n
498 * @param n a size/length parameter, interpretation depends upon
499 * the value of type
500 */
501 BigInt(NumberType type, size_t n);
502
503 void shrink_to_fit();
504
505 static void const_time_lookup(SecureVector<word>& output,
506 const std::vector<BigInt>& vec,
507 size_t idx);
508
509 private:
511 Sign signedness;
512 };
513
514/*
515* Arithmetic Operators
516*/
517BigInt BOTAN_DLL operator+(const BigInt& x, const BigInt& y);
518BigInt BOTAN_DLL operator-(const BigInt& x, const BigInt& y);
519BigInt BOTAN_DLL operator*(const BigInt& x, const BigInt& y);
520BigInt BOTAN_DLL operator/(const BigInt& x, const BigInt& d);
521BigInt BOTAN_DLL operator%(const BigInt& x, const BigInt& m);
522word BOTAN_DLL operator%(const BigInt& x, word m);
523BigInt BOTAN_DLL operator<<(const BigInt& x, size_t n);
524BigInt BOTAN_DLL operator>>(const BigInt& x, size_t n);
525
526/*
527* Comparison Operators
528*/
529inline bool operator==(const BigInt& a, const BigInt& b)
530 { return (a.cmp(b) == 0); }
531inline bool operator!=(const BigInt& a, const BigInt& b)
532 { return (a.cmp(b) != 0); }
533inline bool operator<=(const BigInt& a, const BigInt& b)
534 { return (a.cmp(b) <= 0); }
535inline bool operator>=(const BigInt& a, const BigInt& b)
536 { return (a.cmp(b) >= 0); }
537inline bool operator<(const BigInt& a, const BigInt& b)
538 { return (a.cmp(b) < 0); }
539inline bool operator>(const BigInt& a, const BigInt& b)
540 { return (a.cmp(b) > 0); }
541
542/*
543* I/O Operators
544*/
545BOTAN_DLL std::ostream& operator<<(std::ostream&, const BigInt&);
546BOTAN_DLL std::istream& operator>>(std::istream&, BigInt&);
547
548}
549
550namespace std {
551
552template<>
554 {
555 x.swap(y);
556 }
557
558}
559
560#endif
BigInt & operator++()
Definition bigint.h:97
BigInt & operator>>=(size_t shift)
Definition big_ops2.cpp:204
void swap(BigInt &other)
Definition bigint.cpp:107
size_t sig_words() const
Definition bigint.h:290
BigInt & operator/=(const BigInt &y)
Definition big_ops2.cpp:133
SecureVector< word > & get_reg()
Definition bigint.h:325
bool is_odd() const
Definition bigint.h:164
size_t size() const
Definition bigint.h:284
BigInt & operator*=(const BigInt &y)
Definition big_ops2.cpp:95
void assign(const word x[], size_t length)
Definition bigint.h:337
const word * data() const
Definition bigint.h:317
word word_at(size_t n) const
Definition bigint.h:238
BigInt & operator-=(const BigInt &y)
Definition big_ops2.cpp:53
s32bit cmp(const BigInt &n, bool check_signs=true) const
Definition bigint.cpp:133
BigInt operator--(int)
Definition bigint.h:112
bool is_even() const
Definition bigint.h:158
BigInt & operator+=(const BigInt &y)
Definition big_ops2.cpp:18
BigInt & operator%=(const BigInt &y)
Definition big_ops2.cpp:145
void clear()
Definition bigint.h:143
Sign sign() const
Definition bigint.h:257
BigInt & operator<<=(size_t shift)
Definition big_ops2.cpp:186
BigInt & operator--()
Definition bigint.h:102
word & operator[](size_t i)
Definition bigint.h:131
const word & operator[](size_t i) const
Definition bigint.h:138
bool is_zero() const
Definition bigint.h:176
BigInt operator++(int)
Definition bigint.h:107
bool is_negative() const
Definition bigint.h:245
bool is_nonzero() const
Definition bigint.h:170
bool is_positive() const
Definition bigint.h:251
bool get_bit(size_t n) const
Definition bigint.cpp:161
const SecureVector< word > & get_reg() const
Definition bigint.h:332
bool BOTAN_DLL operator>(const X509_Time &, const X509_Time &)
Definition asn1_tm.cpp:288
void zeroise(MemoryRegion< T > &vec)
Definition secmem.h:428
int operator>>(int fd, Pipe &pipe)
Definition fd_unix.cpp:39
bool BOTAN_DLL operator<(const X509_Time &, const X509_Time &)
Definition asn1_tm.cpp:286
signed int s32bit
Definition types.h:37
bool BOTAN_DLL operator>=(const X509_Time &, const X509_Time &)
Definition asn1_tm.cpp:283
BigInt operator/(const BigInt &x, const BigInt &y)
Definition big_ops3.cpp:109
OctetString operator+(const OctetString &k1, const OctetString &k2)
Definition symkey.cpp:114
BigInt abs(const BigInt &n)
Definition numthry.h:44
unsigned long long u64bit
Definition types.h:49
std::runtime_error Exception
Definition exceptn.h:19
void copy_mem(T *out, const T *in, size_t n)
Definition mem_ops.h:22
u32bit to_u32bit(const std::string &number)
Definition parsing.cpp:18
BigInt operator*(const BigInt &x, const BigInt &y)
Definition big_ops3.cpp:83
bool operator!=(const OctetString &s1, const OctetString &s2)
Definition symkey.cpp:106
unsigned int u32bit
Definition types.h:32
bool BOTAN_DLL operator<=(const X509_Time &, const X509_Time &)
Definition asn1_tm.cpp:281
BigInt operator-(const BigInt &x, const BigInt &y)
Definition big_ops3.cpp:48
int operator<<(int fd, Pipe &pipe)
Definition fd_unix.cpp:17
BigInt operator%(const BigInt &n, const BigInt &mod)
Definition big_ops3.cpp:119
bool operator==(const OctetString &s1, const OctetString &s2)
Definition symkey.cpp:98
Definition secmem.h:435
void swap(Botan::MemoryRegion< T > &x, Botan::MemoryRegion< T > &y)
Definition secmem.h:438