// Copyright (c) 2012-present The Bitcoin Core developers

// Distributed under the MIT software license, see the accompanying

// file COPYING or http://www.opensource.org/licenses/mit-license.php.

#include <common/bloom.h>

#include <hash.h>

#include <primitives/transaction.h>

#include <random.h>

#include <script/script.h>

#include <script/solver.h>

#include <span.h>

#include <streams.h>

#include <util/fastrange.h>

#include <util/overflow.h>

#include <algorithm>

#include <cmath>

#include <compare>

#include <vector>

static constexpr double LN2SQUARED = 0.4804530139182014246671025263266649717305529515945455;

static constexpr double LN2 = 0.6931471805599453094172321214581765680755001343602552;

CBloomFilter::CBloomFilter(const unsigned int nElements, const double nFPRate, const unsigned int nTweakIn, unsigned char nFlagsIn) :

    /**

     * The ideal size for a bloom filter with a given number of elements and false positive rate is:

     * - nElements * log(fp rate) / ln(2)^2

     * We ignore filter parameters which will create a bloom filter larger than the protocol limits

     */

    vData(std::min((unsigned int)(-1  / LN2SQUARED * nElements * log(nFPRate)), MAX_BLOOM_FILTER_SIZE * 8) / 8),

    /**

     * The ideal number of hash functions is filter size * ln(2) / number of elements

     * Again, we ignore filter parameters which will create a bloom filter with more hash functions than the protocol limits

     * See https://en.wikipedia.org/wiki/Bloom_filter for an explanation of these formulas

     */

    nHashFuncs(std::min((unsigned int)(vData.size() * 8 / nElements * LN2), MAX_HASH_FUNCS)),

    nTweak(nTweakIn),

    nFlags(nFlagsIn)

{

}

inline unsigned int CBloomFilter::Hash(unsigned int nHashNum, std::span<const unsigned char> vDataToHash) const

{

    // 0xFBA4C795 chosen as it guarantees a reasonable bit difference between nHashNum values.

    return MurmurHash3(nHashNum * 0xFBA4C795 + nTweak, vDataToHash) % (vData.size() * 8);

}

void CBloomFilter::insert(std::span<const unsigned char> vKey)

{

    if (vData.empty()) // Avoid divide-by-zero (CVE-2013-5700)

  Branch (51:9): [True: 21.7k, False: 67.3k]

        return;

    for (unsigned int i = 0; i < nHashFuncs; i++)

  Branch (53:30): [True: 303k, False: 67.3k]

    {

        unsigned int nIndex = Hash(i, vKey);

        // Sets bit nIndex of vData

        vData[nIndex >> 3] |= (1 << (7 & nIndex));

    }

}

void CBloomFilter::insert(const COutPoint& outpoint)

{

    DataStream stream{};

    stream << outpoint;

    insert(MakeUCharSpan(stream));

}

bool CBloomFilter::contains(std::span<const unsigned char> vKey) const

{

    if (vData.empty()) // Avoid divide-by-zero (CVE-2013-5700)

  Branch (70:9): [True: 43.3k, False: 544k]

        return true;

    for (unsigned int i = 0; i < nHashFuncs; i++)

  Branch (72:30): [True: 1.18M, False: 334k]

    {

        unsigned int nIndex = Hash(i, vKey);

        // Checks bit nIndex of vData

        if (!(vData[nIndex >> 3] & (1 << (7 & nIndex))))

  Branch (76:13): [True: 209k, False: 977k]

            return false;

    }

    return true;

}

bool CBloomFilter::contains(const COutPoint& outpoint) const

{

    DataStream stream{};

    stream << outpoint;

    return contains(MakeUCharSpan(stream));

}

bool CBloomFilter::IsWithinSizeConstraints() const

{

    return vData.size() <= MAX_BLOOM_FILTER_SIZE && nHashFuncs <= MAX_HASH_FUNCS;

  Branch (91:12): [True: 171k, False: 4]
  Branch (91:53): [True: 171k, False: 308]

}

bool CBloomFilter::IsRelevantAndUpdate(const CTransaction& tx)

{

    bool fFound = false;

    // Match if the filter contains the hash of tx

    //  for finding tx when they appear in a block

    if (vData.empty()) // zero-size = "match-all" filter

  Branch (99:9): [True: 283k, False: 90.7k]

        return true;

    const Txid& hash = tx.GetHash();

    if (contains(hash.ToUint256()))

  Branch (102:9): [True: 77.4k, False: 13.3k]

        fFound = true;

    for (unsigned int i = 0; i < tx.vout.size(); i++)

  Branch (105:30): [True: 293k, False: 90.7k]

    {

        const CTxOut& txout = tx.vout[i];

        // Match if the filter contains any arbitrary script data element in any scriptPubKey in tx

        // If this matches, also add the specific output that was matched.

        // This means clients don't have to update the filter themselves when a new relevant tx

        // is discovered in order to find spending transactions, which avoids round-tripping and race conditions.

        CScript::const_iterator pc = txout.scriptPubKey.begin();

        std::vector<unsigned char> data;

        while (pc < txout.scriptPubKey.end())

  Branch (114:16): [True: 1.50M, False: 72.2k]

        {

            opcodetype opcode;

            if (!txout.scriptPubKey.GetOp(pc, opcode, data))

  Branch (117:17): [True: 76.6k, False: 1.42M]

                break;

            if (data.size() != 0 && contains(data))

  Branch (119:17): [True: 249k, False: 1.17M]
  Branch (119:37): [True: 144k, False: 105k]

            {

                fFound = true;

                if ((nFlags & BLOOM_UPDATE_MASK) == BLOOM_UPDATE_ALL)

  Branch (122:21): [True: 1.46k, False: 142k]

                    insert(COutPoint(hash, i));

                else if ((nFlags & BLOOM_UPDATE_MASK) == BLOOM_UPDATE_P2PUBKEY_ONLY)

  Branch (124:26): [True: 132k, False: 10.8k]

                {

                    std::vector<std::vector<unsigned char> > vSolutions;

                    TxoutType type = Solver(txout.scriptPubKey, vSolutions);

                    if (type == TxoutType::PUBKEY || type == TxoutType::MULTISIG) {

  Branch (128:25): [True: 7.12k, False: 124k]
  Branch (128:54): [True: 1.98k, False: 122k]

                        insert(COutPoint(hash, i));

                    }

                }

                break;

            }

        }

    }

    if (fFound)

  Branch (137:9): [True: 78.0k, False: 12.6k]

        return true;

    for (const CTxIn& txin : tx.vin)

  Branch (140:28): [True: 45.9k, False: 9.82k]

    {

        // Match if the filter contains an outpoint tx spends

        if (contains(txin.prevout))

  Branch (143:13): [True: 1.56k, False: 44.4k]

            return true;

        // Match if the filter contains any arbitrary script data element in any scriptSig in tx

        CScript::const_iterator pc = txin.scriptSig.begin();

        std::vector<unsigned char> data;

        while (pc < txin.scriptSig.end())

  Branch (149:16): [True: 416k, False: 29.4k]

        {

            opcodetype opcode;

            if (!txin.scriptSig.GetOp(pc, opcode, data))

  Branch (152:17): [True: 13.6k, False: 403k]

                break;

            if (data.size() != 0 && contains(data))

  Branch (154:17): [True: 45.1k, False: 357k]
  Branch (154:37): [True: 1.31k, False: 43.8k]

                return true;

        }

    }

    return false;

}

CRollingBloomFilter::CRollingBloomFilter(const unsigned int nElements, const double fpRate)

{

    double logFpRate = log(fpRate);

    /* The optimal number of hash functions is log(fpRate) / log(0.5), but

     * restrict it to the range 1-50. */

    nHashFuncs = std::max(1, std::min((int)round(logFpRate / log(0.5)), 50));

    /* In this rolling bloom filter, we'll store between 2 and 3 generations of nElements / 2 entries. */

    nEntriesPerGeneration = CeilDiv(nElements, 2u);

    uint32_t nMaxElements = nEntriesPerGeneration * 3;

    /* The maximum fpRate = pow(1.0 - exp(-nHashFuncs * nMaxElements / nFilterBits), nHashFuncs)

     * =>          pow(fpRate, 1.0 / nHashFuncs) = 1.0 - exp(-nHashFuncs * nMaxElements / nFilterBits)

     * =>          1.0 - pow(fpRate, 1.0 / nHashFuncs) = exp(-nHashFuncs * nMaxElements / nFilterBits)

     * =>          log(1.0 - pow(fpRate, 1.0 / nHashFuncs)) = -nHashFuncs * nMaxElements / nFilterBits

     * =>          nFilterBits = -nHashFuncs * nMaxElements / log(1.0 - pow(fpRate, 1.0 / nHashFuncs))

     * =>          nFilterBits = -nHashFuncs * nMaxElements / log(1.0 - exp(logFpRate / nHashFuncs))

     */

    uint32_t nFilterBits = (uint32_t)ceil(-1.0 * nHashFuncs * nMaxElements / log(1.0 - exp(logFpRate / nHashFuncs)));

    data.clear();

    /* For each data element we need to store 2 bits. If both bits are 0, the

     * bit is treated as unset. If the bits are (01), (10), or (11), the bit is

     * treated as set in generation 1, 2, or 3 respectively.

     * These bits are stored in separate integers: position P corresponds to bit

     * (P & 63) of the integers data[(P >> 6) * 2] and data[(P >> 6) * 2 + 1]. */

    data.resize(CeilDiv(nFilterBits, 64u) << 1);

    reset();

}

/* Similar to CBloomFilter::Hash */

static inline uint32_t RollingBloomHash(unsigned int nHashNum, uint32_t nTweak, std::span<const unsigned char> vDataToHash)

{

    return MurmurHash3(nHashNum * 0xFBA4C795 + nTweak, vDataToHash);

}

void CRollingBloomFilter::insert(std::span<const unsigned char> vKey)

{

    if (nEntriesThisGeneration == nEntriesPerGeneration) {

  Branch (197:9): [True: 21.4k, False: 841k]

        nEntriesThisGeneration = 0;

        nGeneration++;

        if (nGeneration == 4) {

  Branch (200:13): [True: 6.28k, False: 15.1k]

            nGeneration = 1;

        }

        uint64_t nGenerationMask1 = 0 - (uint64_t)(nGeneration & 1);

        uint64_t nGenerationMask2 = 0 - (uint64_t)(nGeneration >> 1);

        /* Wipe old entries that used this generation number. */

        for (uint32_t p = 0; p < data.size(); p += 2) {

  Branch (206:30): [True: 136k, False: 21.4k]

            uint64_t p1 = data[p], p2 = data[p + 1];

            uint64_t mask = (p1 ^ nGenerationMask1) | (p2 ^ nGenerationMask2);

            data[p] = p1 & mask;

            data[p + 1] = p2 & mask;

        }

    }

    nEntriesThisGeneration++;

    for (int n = 0; n < nHashFuncs; n++) {

  Branch (215:21): [True: 13.1M, False: 863k]

        uint32_t h = RollingBloomHash(n, nTweak, vKey);

        int bit = h & 0x3F;

        /* FastMod works with the upper bits of h, so it is safe to ignore that the lower bits of h are already used for bit. */

        uint32_t pos = FastRange32(h, data.size());

        /* The lowest bit of pos is ignored, and set to zero for the first bit, and to one for the second. */

        data[pos & ~1U] = (data[pos & ~1U] & ~(uint64_t{1} << bit)) | (uint64_t(nGeneration & 1)) << bit;

        data[pos | 1] = (data[pos | 1] & ~(uint64_t{1} << bit)) | (uint64_t(nGeneration >> 1)) << bit;

    }

}

bool CRollingBloomFilter::contains(std::span<const unsigned char> vKey) const

{

    for (int n = 0; n < nHashFuncs; n++) {

  Branch (228:21): [True: 15.5M, False: 379k]

        uint32_t h = RollingBloomHash(n, nTweak, vKey);

        int bit = h & 0x3F;

        uint32_t pos = FastRange32(h, data.size());

        /* If the relevant bit is not set in either data[pos & ~1] or data[pos | 1], the filter does not contain vKey */

        if (!(((data[pos & ~1U] | data[pos | 1]) >> bit) & 1)) {

  Branch (233:13): [True: 7.62M, False: 7.89M]

            return false;

        }

    }

    return true;

}

void CRollingBloomFilter::reset()

{

    nTweak = FastRandomContext().rand<unsigned int>();

    nEntriesThisGeneration = 0;

    nGeneration = 1;

    std::fill(data.begin(), data.end(), 0);

}