Add a CBloomFilter class for use as a transaction filter.

7 files changed, 209 insertions, 0 deletions

diff --git a/bitcoin-qt.pro b/bitcoin-qt.pro
index 852879f8cc..50ca80cb0f 100644
--- a/bitcoin-qt.pro
+++ b/bitcoin-qt.pro
@@ -155,6 +155,7 @@ HEADERS += src/qt/bitcoingui.h \
     src/script.h \
     src/init.h \
     src/irc.h \
+    src/bloom.h \
     src/mruset.h \
     src/json/json_spirit_writer_template.h \
     src/json/json_spirit_writer.h \
@@ -223,6 +224,7 @@ SOURCES += src/qt/bitcoin.cpp src/qt/bitcoingui.cpp \
     src/init.cpp \
     src/net.cpp \
     src/irc.cpp \
+    src/bloom.cpp \
     src/checkpoints.cpp \
     src/addrman.cpp \
     src/db.cpp \
diff --git a/src/bloom.cpp b/src/bloom.cpp
new file mode 100644
index 0000000000..5fac1d06ab
--- /dev/null
+++ b/src/bloom.cpp
@@ -0,0 +1,133 @@
+// Copyright (c) 2012 The Bitcoin developers
+// Distributed under the MIT/X11 software license, see the accompanying
+// file COPYING or http://www.opensource.org/licenses/mit-license.php.
+#include <math.h>
+#include <stdlib.h>
+
+#include "bloom.h"
+#include "main.h"
+#include "script.h"
+
+#define LN2SQUARED 0.4804530139182014246671025263266649717305529515945455
+#define LN2 0.6931471805599453094172321214581765680755001343602552
+
+using namespace std;
+
+static const unsigned char bit_mask[8] = {0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80};
+
+CBloomFilter::CBloomFilter(unsigned int nElements, double nFPRate) :
+// 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(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 http://en.wikipedia.org/wiki/Bloom_filter for an explanation of these formulas
+nHashFuncs(min((unsigned int)(vData.size() * 8 / nElements * LN2), MAX_HASH_FUNCS))
+{
+}
+
+inline unsigned int CBloomFilter::Hash(unsigned int nHashNum, const std::vector<unsigned char>& vDataToHash) const
+{
+    // 0xFBA4C795 chosen as it guarantees a reasonable bit difference between nHashNum values.
+    return MurmurHash3(nHashNum * 0xFBA4C795, vDataToHash) % (vData.size() * 8);
+}
+
+void CBloomFilter::insert(const vector<unsigned char>& vKey)
+{
+    for (unsigned int i = 0; i < nHashFuncs; i++)
+    {
+        unsigned int nIndex = Hash(i, vKey);
+        // Sets bit nIndex of vData
+        vData[nIndex >> 3] |= bit_mask[7 & nIndex];
+    }
+}
+
+void CBloomFilter::insert(const COutPoint& outpoint)
+{
+    CDataStream stream(SER_NETWORK, PROTOCOL_VERSION);
+    stream << outpoint;
+    vector<unsigned char> data(stream.begin(), stream.end());
+    insert(data);
+}
+
+void CBloomFilter::insert(const uint256& hash)
+{
+    vector<unsigned char> data(hash.begin(), hash.end());
+    insert(data);
+}
+
+bool CBloomFilter::contains(const vector<unsigned char>& vKey) const
+{
+    for (unsigned int i = 0; i < nHashFuncs; i++)
+    {
+        unsigned int nIndex = Hash(i, vKey);
+        // Checks bit nIndex of vData
+        if (!(vData[nIndex >> 3] & bit_mask[7 & nIndex]))
+            return false;
+    }
+    return true;
+}
+
+bool CBloomFilter::contains(const COutPoint& outpoint) const
+{
+    CDataStream stream(SER_NETWORK, PROTOCOL_VERSION);
+    stream << outpoint;
+    vector<unsigned char> data(stream.begin(), stream.end());
+    return contains(data);
+}
+
+bool CBloomFilter::contains(const uint256& hash) const
+{
+    vector<unsigned char> data(hash.begin(), hash.end());
+    return contains(data);
+}
+
+bool CBloomFilter::IsWithinSizeConstraints() const
+{
+    return vData.size() <= MAX_BLOOM_FILTER_SIZE && nHashFuncs <= MAX_HASH_FUNCS;
+}
+
+bool CBloomFilter::IsTransactionRelevantToFilter(const CTransaction& tx) const
+{
+    // Match if the filter contains the hash of tx
+    //  for finding tx when they appear in a block
+    if (contains(tx.GetHash()))
+        return true;
+
+    BOOST_FOREACH(const CTxOut& txout, tx.vout)
+    {
+        // Match if the filter contains any arbitrary script data element in any scriptPubKey in tx
+        CScript::const_iterator pc = txout.scriptPubKey.begin();
+        vector<unsigned char> data;
+        while (pc < txout.scriptPubKey.end())
+        {
+            opcodetype opcode;
+            if (!txout.scriptPubKey.GetOp(pc, opcode, data))
+                break;
+            if (data.size() != 0 && contains(data))
+                return true;
+        }
+    }
+
+    BOOST_FOREACH(const CTxIn& txin, tx.vin)
+    {
+        // Match if the filter contains an outpoint tx spends
+        if (contains(txin.prevout))
+            return true;
+
+        // Match if the filter contains any arbitrary script data element in any scriptSig in tx
+        CScript::const_iterator pc = txin.scriptSig.begin();
+        vector<unsigned char> data;
+        while (pc < txin.scriptSig.end())
+        {
+            opcodetype opcode;
+            if (!txin.scriptSig.GetOp(pc, opcode, data))
+                break;
+            if (data.size() != 0 && contains(data))
+                return true;
+        }
+    }
+
+    return false;
+}
diff --git a/src/bloom.h b/src/bloom.h
new file mode 100644
index 0000000000..ce84e66552
--- /dev/null
+++ b/src/bloom.h
@@ -0,0 +1,70 @@
+// Copyright (c) 2012 The Bitcoin developers
+// Distributed under the MIT/X11 software license, see the accompanying
+// file COPYING or http://www.opensource.org/licenses/mit-license.php.
+#ifndef BITCOIN_BLOOM_H
+#define BITCOIN_BLOOM_H
+
+#include <vector>
+
+#include "uint256.h"
+#include "serialize.h"
+
+class COutPoint;
+class CTransaction;
+
+// 20,000 items with fp rate < 0.1% or 10,000 items and <0.0001%
+static const unsigned int MAX_BLOOM_FILTER_SIZE = 36000; // bytes
+static const unsigned int MAX_HASH_FUNCS = 50;
+
+
+/**
+ * BloomFilter is a probabilistic filter which SPV clients provide
+ * so that we can filter the transactions we sends them.
+ * 
+ * This allows for significantly more efficient transaction and block downloads.
+ * 
+ * Because bloom filters are probabilistic, an SPV node can increase the false-
+ * positive rate, making us send them transactions which aren't actually theirs, 
+ * allowing clients to trade more bandwidth for more privacy by obfuscating which
+ * keys are owned by them.
+ */
+class CBloomFilter
+{
+private:
+    std::vector<unsigned char> vData;
+    unsigned int nHashFuncs;
+
+    unsigned int Hash(unsigned int nHashNum, const std::vector<unsigned char>& vDataToHash) const;
+
+public:
+    // Creates a new bloom filter which will provide the given fp rate when filled with the given number of elements
+    // Note that if the given parameters will result in a filter outside the bounds of the protocol limits,
+    // the filter created will be as close to the given parameters as possible within the protocol limits.
+    // This will apply if nFPRate is very low or nElements is unreasonably high.
+    CBloomFilter(unsigned int nElements, double nFPRate);
+    // Using a filter initialized with this results in undefined behavior
+    // Should only be used for deserialization
+    CBloomFilter() {}
+
+    IMPLEMENT_SERIALIZE
+    (
+        READWRITE(vData);
+        READWRITE(nHashFuncs);
+    )
+
+    void insert(const std::vector<unsigned char>& vKey);
+    void insert(const COutPoint& outpoint);
+    void insert(const uint256& hash);
+
+    bool contains(const std::vector<unsigned char>& vKey) const;
+    bool contains(const COutPoint& outpoint) const;
+    bool contains(const uint256& hash) const;
+
+    // True if the size is <= MAX_BLOOM_FILTER_SIZE and the number of hash functions is <= MAX_HASH_FUNCS
+    // (catch a filter which was just deserialized which was too big)
+    bool IsWithinSizeConstraints() const;
+
+    bool IsTransactionRelevantToFilter(const CTransaction& tx) const;
+};
+
+#endif /* BITCOIN_BLOOM_H */
diff --git a/src/makefile.linux-mingw b/src/makefile.linux-mingw
index 95e7e83cd0..ff565f2a2d 100644
--- a/src/makefile.linux-mingw
+++ b/src/makefile.linux-mingw
@@ -84,6 +84,7 @@ OBJS= \
     obj/walletdb.o \
     obj/noui.o \
     obj/hash.o \
+    obj/bloom.o \
     obj/leveldb.o \
     obj/txdb.o
 
diff --git a/src/makefile.mingw b/src/makefile.mingw
index 2abc34478d..9a6680bf40 100644
--- a/src/makefile.mingw
+++ b/src/makefile.mingw
@@ -79,6 +79,7 @@ OBJS= \
     obj/wallet.o \
     obj/walletdb.o \
     obj/hash.o \
+    obj/bloom.o \
     obj/noui.o \
     obj/leveldb.o \
     obj/txdb.o
diff --git a/src/makefile.osx b/src/makefile.osx
index 474f17b735..8b7c559fa1 100644
--- a/src/makefile.osx
+++ b/src/makefile.osx
@@ -97,6 +97,7 @@ OBJS= \
     obj/wallet.o \
     obj/walletdb.o \
     obj/hash.o \
+    obj/bloom.o \
     obj/noui.o \
     obj/leveldb.o \
     obj/txdb.o
diff --git a/src/makefile.unix b/src/makefile.unix
index 979a4a6520..14cf1b8fa5 100644
--- a/src/makefile.unix
+++ b/src/makefile.unix
@@ -128,6 +128,7 @@ OBJS= \
     obj/wallet.o \
     obj/walletdb.o \
     obj/hash.o \
+    obj/bloom.o \
     obj/noui.o \
     obj/leveldb.o \
     obj/txdb.o