aboutsummaryrefslogtreecommitdiff
path: root/src/arith_uint256.h
diff options
context:
space:
mode:
authorWladimir J. van der Laan <laanwj@gmail.com>2014-12-15 10:22:19 +0100
committerWladimir J. van der Laan <laanwj@gmail.com>2015-01-05 15:45:35 +0100
commitbfc6070342b9f43bcf125526e6a3c8ed34e29a71 (patch)
tree8e6665341afc799f922f8b011e2659f02682cdbb /src/arith_uint256.h
parent734f85c4f0b40efd3f6c0367683c1bab1a2a7b19 (diff)
uint256->arith_uint256 blob256->uint256
Introduce new opaque implementation of `uint256`, move old "arithmetic" implementation to `arith_uint256.
Diffstat (limited to 'src/arith_uint256.h')
-rw-r--r--src/arith_uint256.h351
1 files changed, 341 insertions, 10 deletions
diff --git a/src/arith_uint256.h b/src/arith_uint256.h
index 3bb384ca8a..9e32b124c2 100644
--- a/src/arith_uint256.h
+++ b/src/arith_uint256.h
@@ -1,19 +1,350 @@
+// Copyright (c) 2009-2010 Satoshi Nakamoto
+// Copyright (c) 2009-2014 The Bitcoin developers
+// Distributed under the MIT software license, see the accompanying
+// file COPYING or http://www.opensource.org/licenses/mit-license.php.
+
#ifndef BITCOIN_ARITH_UINT256_H
#define BITCOIN_ARITH_UINT256_H
-// Temporary for migration to opaque uint160/256
-#include "uint256.h"
+#include <assert.h>
+#include <cstring>
+#include <stdexcept>
+#include <stdint.h>
+#include <string>
+#include <vector>
-class arith_uint256 : public uint256 {
+class uint_error : public std::runtime_error {
public:
- arith_uint256() {}
- arith_uint256(const base_uint<256>& b) : uint256(b) {}
- arith_uint256(uint64_t b) : uint256(b) {}
- explicit arith_uint256(const std::string& str) : uint256(str) {}
- explicit arith_uint256(const std::vector<unsigned char>& vch) : uint256(vch) {}
+ explicit uint_error(const std::string& str) : std::runtime_error(str) {}
+};
+
+/** Template base class for unsigned big integers. */
+template<unsigned int BITS>
+class base_uint
+{
+protected:
+ enum { WIDTH=BITS/32 };
+ uint32_t pn[WIDTH];
+public:
+
+ base_uint()
+ {
+ for (int i = 0; i < WIDTH; i++)
+ pn[i] = 0;
+ }
+
+ base_uint(const base_uint& b)
+ {
+ for (int i = 0; i < WIDTH; i++)
+ pn[i] = b.pn[i];
+ }
+
+ base_uint& operator=(const base_uint& b)
+ {
+ for (int i = 0; i < WIDTH; i++)
+ pn[i] = b.pn[i];
+ return *this;
+ }
+
+ base_uint(uint64_t b)
+ {
+ pn[0] = (unsigned int)b;
+ pn[1] = (unsigned int)(b >> 32);
+ for (int i = 2; i < WIDTH; i++)
+ pn[i] = 0;
+ }
+
+ explicit base_uint(const std::string& str);
+ explicit base_uint(const std::vector<unsigned char>& vch);
+
+ bool operator!() const
+ {
+ for (int i = 0; i < WIDTH; i++)
+ if (pn[i] != 0)
+ return false;
+ return true;
+ }
+
+ const base_uint operator~() const
+ {
+ base_uint ret;
+ for (int i = 0; i < WIDTH; i++)
+ ret.pn[i] = ~pn[i];
+ return ret;
+ }
+
+ const base_uint operator-() const
+ {
+ base_uint ret;
+ for (int i = 0; i < WIDTH; i++)
+ ret.pn[i] = ~pn[i];
+ ret++;
+ return ret;
+ }
+
+ double getdouble() const;
+
+ base_uint& operator=(uint64_t b)
+ {
+ pn[0] = (unsigned int)b;
+ pn[1] = (unsigned int)(b >> 32);
+ for (int i = 2; i < WIDTH; i++)
+ pn[i] = 0;
+ return *this;
+ }
+
+ base_uint& operator^=(const base_uint& b)
+ {
+ for (int i = 0; i < WIDTH; i++)
+ pn[i] ^= b.pn[i];
+ return *this;
+ }
+
+ base_uint& operator&=(const base_uint& b)
+ {
+ for (int i = 0; i < WIDTH; i++)
+ pn[i] &= b.pn[i];
+ return *this;
+ }
+
+ base_uint& operator|=(const base_uint& b)
+ {
+ for (int i = 0; i < WIDTH; i++)
+ pn[i] |= b.pn[i];
+ return *this;
+ }
+
+ base_uint& operator^=(uint64_t b)
+ {
+ pn[0] ^= (unsigned int)b;
+ pn[1] ^= (unsigned int)(b >> 32);
+ return *this;
+ }
+
+ base_uint& operator|=(uint64_t b)
+ {
+ pn[0] |= (unsigned int)b;
+ pn[1] |= (unsigned int)(b >> 32);
+ return *this;
+ }
+
+ base_uint& operator<<=(unsigned int shift);
+ base_uint& operator>>=(unsigned int shift);
+
+ base_uint& operator+=(const base_uint& b)
+ {
+ uint64_t carry = 0;
+ for (int i = 0; i < WIDTH; i++)
+ {
+ uint64_t n = carry + pn[i] + b.pn[i];
+ pn[i] = n & 0xffffffff;
+ carry = n >> 32;
+ }
+ return *this;
+ }
+
+ base_uint& operator-=(const base_uint& b)
+ {
+ *this += -b;
+ return *this;
+ }
+
+ base_uint& operator+=(uint64_t b64)
+ {
+ base_uint b;
+ b = b64;
+ *this += b;
+ return *this;
+ }
+
+ base_uint& operator-=(uint64_t b64)
+ {
+ base_uint b;
+ b = b64;
+ *this += -b;
+ return *this;
+ }
+
+ base_uint& operator*=(uint32_t b32);
+ base_uint& operator*=(const base_uint& b);
+ base_uint& operator/=(const base_uint& b);
+
+ base_uint& operator++()
+ {
+ // prefix operator
+ int i = 0;
+ while (++pn[i] == 0 && i < WIDTH-1)
+ i++;
+ return *this;
+ }
+
+ const base_uint operator++(int)
+ {
+ // postfix operator
+ const base_uint ret = *this;
+ ++(*this);
+ return ret;
+ }
+
+ base_uint& operator--()
+ {
+ // prefix operator
+ int i = 0;
+ while (--pn[i] == (uint32_t)-1 && i < WIDTH-1)
+ i++;
+ return *this;
+ }
+
+ const base_uint operator--(int)
+ {
+ // postfix operator
+ const base_uint ret = *this;
+ --(*this);
+ return ret;
+ }
+
+ int CompareTo(const base_uint& b) const;
+ bool EqualTo(uint64_t b) const;
+
+ friend inline const base_uint operator+(const base_uint& a, const base_uint& b) { return base_uint(a) += b; }
+ friend inline const base_uint operator-(const base_uint& a, const base_uint& b) { return base_uint(a) -= b; }
+ friend inline const base_uint operator*(const base_uint& a, const base_uint& b) { return base_uint(a) *= b; }
+ friend inline const base_uint operator/(const base_uint& a, const base_uint& b) { return base_uint(a) /= b; }
+ friend inline const base_uint operator|(const base_uint& a, const base_uint& b) { return base_uint(a) |= b; }
+ friend inline const base_uint operator&(const base_uint& a, const base_uint& b) { return base_uint(a) &= b; }
+ friend inline const base_uint operator^(const base_uint& a, const base_uint& b) { return base_uint(a) ^= b; }
+ friend inline const base_uint operator>>(const base_uint& a, int shift) { return base_uint(a) >>= shift; }
+ friend inline const base_uint operator<<(const base_uint& a, int shift) { return base_uint(a) <<= shift; }
+ friend inline const base_uint operator*(const base_uint& a, uint32_t b) { return base_uint(a) *= b; }
+ friend inline bool operator==(const base_uint& a, const base_uint& b) { return memcmp(a.pn, b.pn, sizeof(a.pn)) == 0; }
+ friend inline bool operator!=(const base_uint& a, const base_uint& b) { return memcmp(a.pn, b.pn, sizeof(a.pn)) != 0; }
+ friend inline bool operator>(const base_uint& a, const base_uint& b) { return a.CompareTo(b) > 0; }
+ friend inline bool operator<(const base_uint& a, const base_uint& b) { return a.CompareTo(b) < 0; }
+ friend inline bool operator>=(const base_uint& a, const base_uint& b) { return a.CompareTo(b) >= 0; }
+ friend inline bool operator<=(const base_uint& a, const base_uint& b) { return a.CompareTo(b) <= 0; }
+ friend inline bool operator==(const base_uint& a, uint64_t b) { return a.EqualTo(b); }
+ friend inline bool operator!=(const base_uint& a, uint64_t b) { return !a.EqualTo(b); }
+
+ std::string GetHex() const;
+ void SetHex(const char* psz);
+ void SetHex(const std::string& str);
+ std::string ToString() const;
+
+ unsigned char* begin()
+ {
+ return (unsigned char*)&pn[0];
+ }
+
+ unsigned char* end()
+ {
+ return (unsigned char*)&pn[WIDTH];
+ }
+
+ const unsigned char* begin() const
+ {
+ return (unsigned char*)&pn[0];
+ }
+
+ const unsigned char* end() const
+ {
+ return (unsigned char*)&pn[WIDTH];
+ }
+
+ unsigned int size() const
+ {
+ return sizeof(pn);
+ }
+
+ /**
+ * Returns the position of the highest bit set plus one, or zero if the
+ * value is zero.
+ */
+ unsigned int bits() const;
+
+ uint64_t GetLow64() const
+ {
+ assert(WIDTH >= 2);
+ return pn[0] | (uint64_t)pn[1] << 32;
+ }
+
+ unsigned int GetSerializeSize(int nType, int nVersion) const
+ {
+ return sizeof(pn);
+ }
+
+ template<typename Stream>
+ void Serialize(Stream& s, int nType, int nVersion) const
+ {
+ s.write((char*)pn, sizeof(pn));
+ }
+
+ template<typename Stream>
+ void Unserialize(Stream& s, int nType, int nVersion)
+ {
+ s.read((char*)pn, sizeof(pn));
+ }
+
+ // Temporary for migration to blob160/256
+ uint64_t GetCheapHash() const
+ {
+ return GetLow64();
+ }
+ void SetNull()
+ {
+ memset(pn, 0, sizeof(pn));
+ }
+ bool IsNull() const
+ {
+ for (int i = 0; i < WIDTH; i++)
+ if (pn[i] != 0)
+ return false;
+ return true;
+ }
};
-#define ArithToUint256(x) (x)
-#define UintToArith256(x) (x)
+/** 160-bit unsigned big integer. */
+class arith_uint160 : public base_uint<160> {
+public:
+ arith_uint160() {}
+ arith_uint160(const base_uint<160>& b) : base_uint<160>(b) {}
+ arith_uint160(uint64_t b) : base_uint<160>(b) {}
+ explicit arith_uint160(const std::string& str) : base_uint<160>(str) {}
+ explicit arith_uint160(const std::vector<unsigned char>& vch) : base_uint<160>(vch) {}
+};
+
+/** 256-bit unsigned big integer. */
+class arith_uint256 : public base_uint<256> {
+public:
+ arith_uint256() {}
+ arith_uint256(const base_uint<256>& b) : base_uint<256>(b) {}
+ arith_uint256(uint64_t b) : base_uint<256>(b) {}
+ explicit arith_uint256(const std::string& str) : base_uint<256>(str) {}
+ explicit arith_uint256(const std::vector<unsigned char>& vch) : base_uint<256>(vch) {}
+
+ /**
+ * The "compact" format is a representation of a whole
+ * number N using an unsigned 32bit number similar to a
+ * floating point format.
+ * The most significant 8 bits are the unsigned exponent of base 256.
+ * This exponent can be thought of as "number of bytes of N".
+ * The lower 23 bits are the mantissa.
+ * Bit number 24 (0x800000) represents the sign of N.
+ * N = (-1^sign) * mantissa * 256^(exponent-3)
+ *
+ * Satoshi's original implementation used BN_bn2mpi() and BN_mpi2bn().
+ * MPI uses the most significant bit of the first byte as sign.
+ * Thus 0x1234560000 is compact (0x05123456)
+ * and 0xc0de000000 is compact (0x0600c0de)
+ *
+ * Bitcoin only uses this "compact" format for encoding difficulty
+ * targets, which are unsigned 256bit quantities. Thus, all the
+ * complexities of the sign bit and using base 256 are probably an
+ * implementation accident.
+ */
+ arith_uint256& SetCompact(uint32_t nCompact, bool *pfNegative = NULL, bool *pfOverflow = NULL);
+ uint32_t GetCompact(bool fNegative = false) const;
+
+ uint64_t GetHash(const arith_uint256& salt) const;
+};
#endif // BITCOIN_UINT256_H