Merge bitcoin#20464: refactor: Treat CDataStream bytes as uint8_t

fa29272 Remove redundant MakeUCharSpan wrappers (MarcoFalke)
faf4aa2 Remove CDataStream::Init in favor of C++11 member initialization (MarcoFalke)
fada14b Treat CDataStream bytes as uint8_t (MarcoFalke)
fa8bdb0 refactor: Drop CDataStream constructors in favor of one taking a Span of bytes (MarcoFalke)
faa96f8 Remove unused CDataStream methods (MarcoFalke)

Pull request description:

  Using `uint8_t` for raw bytes has a style benefit:
  * The signedness is clear from reading the code, as it does not depend on the architecture

  Other clean-ups in this pull include:
  * Remove unused methods
  * Constructor is simplified with `Span`
  * Remove `Init()` member in favor of C++11 member initialization

ACKs for top commit:
  laanwj:
    code review ACK fa29272
  theStack:
    ACK fa29272 🍾

Tree-SHA512: 931ee28bd99843d7e894b48e90e1187ffb0278677c267044b3c0c255069d9bbd9298ab2e539b1002a30b543d240450eaec718ef4ee95a7fd4be0a295e926343f

Author: laanwj

src/bench/prevector.cpp

@@ -76,7 +76,7 @@ static void PrevectorDeserialize(benchmark::Bench& bench)
         for (auto x = 0; x < 1000; ++x) {
             s0 >> t1;
         }
-        s0.Init(SER_NETWORK, 0);
+        s0.Rewind();
     });
 }
 

src/dbwrapper.h

@@ -8,9 +8,10 @@
 #include <clientversion.h>
 #include <fs.h>
 #include <serialize.h>
+#include <span.h>
 #include <streams.h>
-#include <util/system.h>
 #include <util/strencodings.h>
+#include <util/system.h>
 
 #include <leveldb/db.h>
 #include <leveldb/write_batch.h>
@@ -73,12 +74,12 @@ class CDBBatch
     {
         ssKey.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey << key;
-        leveldb::Slice slKey(ssKey.data(), ssKey.size());
+        leveldb::Slice slKey((const char*)ssKey.data(), ssKey.size());
 
         ssValue.reserve(DBWRAPPER_PREALLOC_VALUE_SIZE);
         ssValue << value;
         ssValue.Xor(dbwrapper_private::GetObfuscateKey(parent));
-        leveldb::Slice slValue(ssValue.data(), ssValue.size());
+        leveldb::Slice slValue((const char*)ssValue.data(), ssValue.size());
 
         batch.Put(slKey, slValue);
         // LevelDB serializes writes as:
@@ -98,7 +99,7 @@ class CDBBatch
     {
         ssKey.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey << key;
-        leveldb::Slice slKey(ssKey.data(), ssKey.size());
+        leveldb::Slice slKey((const char*)ssKey.data(), ssKey.size());
 
         batch.Delete(slKey);
         // LevelDB serializes erases as:
@@ -137,7 +138,7 @@ class CDBIterator
         CDataStream ssKey(SER_DISK, CLIENT_VERSION);
         ssKey.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey << key;
-        leveldb::Slice slKey(ssKey.data(), ssKey.size());
+        leveldb::Slice slKey((const char*)ssKey.data(), ssKey.size());
         piter->Seek(slKey);
     }
 
@@ -146,7 +147,7 @@ class CDBIterator
     template<typename K> bool GetKey(K& key) {
         leveldb::Slice slKey = piter->key();
         try {
-            CDataStream ssKey(slKey.data(), slKey.data() + slKey.size(), SER_DISK, CLIENT_VERSION);
+            CDataStream ssKey(MakeUCharSpan(slKey), SER_DISK, CLIENT_VERSION);
             ssKey >> key;
         } catch (const std::exception&) {
             return false;
@@ -157,7 +158,7 @@ class CDBIterator
     template<typename V> bool GetValue(V& value) {
         leveldb::Slice slValue = piter->value();
         try {
-            CDataStream ssValue(slValue.data(), slValue.data() + slValue.size(), SER_DISK, CLIENT_VERSION);
+            CDataStream ssValue(MakeUCharSpan(slValue), SER_DISK, CLIENT_VERSION);
             ssValue.Xor(dbwrapper_private::GetObfuscateKey(parent));
             ssValue >> value;
         } catch (const std::exception&) {
@@ -232,7 +233,7 @@ class CDBWrapper
         CDataStream ssKey(SER_DISK, CLIENT_VERSION);
         ssKey.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey << key;
-        leveldb::Slice slKey(ssKey.data(), ssKey.size());
+        leveldb::Slice slKey((const char*)ssKey.data(), ssKey.size());
 
         std::string strValue;
         leveldb::Status status = pdb->Get(readoptions, slKey, &strValue);
@@ -243,7 +244,7 @@ class CDBWrapper
             dbwrapper_private::HandleError(status);
         }
         try {
-            CDataStream ssValue(strValue.data(), strValue.data() + strValue.size(), SER_DISK, CLIENT_VERSION);
+            CDataStream ssValue(MakeUCharSpan(strValue), SER_DISK, CLIENT_VERSION);
             ssValue.Xor(obfuscate_key);
             ssValue >> value;
         } catch (const std::exception&) {
@@ -266,7 +267,7 @@ class CDBWrapper
         CDataStream ssKey(SER_DISK, CLIENT_VERSION);
         ssKey.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey << key;
-        leveldb::Slice slKey(ssKey.data(), ssKey.size());
+        leveldb::Slice slKey((const char*)ssKey.data(), ssKey.size());
 
         std::string strValue;
         leveldb::Status status = pdb->Get(readoptions, slKey, &strValue);
@@ -310,8 +311,8 @@ class CDBWrapper
         ssKey2.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey1 << key_begin;
         ssKey2 << key_end;
-        leveldb::Slice slKey1(ssKey1.data(), ssKey1.size());
-        leveldb::Slice slKey2(ssKey2.data(), ssKey2.size());
+        leveldb::Slice slKey1((const char*)ssKey1.data(), ssKey1.size());
+        leveldb::Slice slKey2((const char*)ssKey2.data(), ssKey2.size());
         uint64_t size = 0;
         leveldb::Range range(slKey1, slKey2);
         pdb->GetApproximateSizes(&range, 1, &size);
@@ -329,11 +330,10 @@ class CDBWrapper
         ssKey2.reserve(DBWRAPPER_PREALLOC_KEY_SIZE);
         ssKey1 << key_begin;
         ssKey2 << key_end;
-        leveldb::Slice slKey1(ssKey1.data(), ssKey1.size());
-        leveldb::Slice slKey2(ssKey2.data(), ssKey2.size());
+        leveldb::Slice slKey1((const char*)ssKey1.data(), ssKey1.size());
+        leveldb::Slice slKey2((const char*)ssKey2.data(), ssKey2.size());
         pdb->CompactRange(&slKey1, &slKey2);
     }
-
 };
 
 #endif // BITCOIN_DBWRAPPER_H

src/psbt.cpp

@@ -363,7 +363,7 @@ bool DecodeBase64PSBT(PartiallySignedTransaction& psbt, const std::string& base6
 
 bool DecodeRawPSBT(PartiallySignedTransaction& psbt, const std::string& tx_data, std::string& error)
 {
-    CDataStream ss_data(tx_data.data(), tx_data.data() + tx_data.size(), SER_NETWORK, PROTOCOL_VERSION);
+    CDataStream ss_data(MakeUCharSpan(tx_data), SER_NETWORK, PROTOCOL_VERSION);
     try {
         ss_data >> psbt;
         if (!ss_data.empty()) {

src/qt/recentrequeststablemodel.cpp

@@ -181,7 +181,7 @@ void RecentRequestsTableModel::addNewRequest(const SendCoinsRecipient &recipient
 // called from ctor when loading from wallet
 void RecentRequestsTableModel::addNewRequest(const std::string &recipient)
 {
-    std::vector<char> data(recipient.begin(), recipient.end());
+    std::vector<uint8_t> data(recipient.begin(), recipient.end());
     CDataStream ss(data, SER_DISK, CLIENT_VERSION);
 
     RecentRequestEntry entry;

src/qt/walletmodel.cpp

@@ -245,7 +245,7 @@ WalletModel::SendCoinsReturn WalletModel::sendCoins(WalletModelTransaction &tran
 
         CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
         ssTx << *newTx;
-        transaction_array.append(&(ssTx[0]), ssTx.size());
+        transaction_array.append((const char*)&(ssTx[0]), ssTx.size());
     }
 
     // Add addresses / update labels that we've sent to the address book,

src/rpc/rawtransaction.cpp

@@ -1345,7 +1345,7 @@ static RPCHelpMan combinepsbt()
 
     CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
     ssTx << merged_psbt;
-    return EncodeBase64(MakeUCharSpan(ssTx));
+    return EncodeBase64(ssTx);
 },
     };
 }
@@ -1484,7 +1484,7 @@ static RPCHelpMan createpsbt()
     CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
     ssTx << psbtx;
 
-    return EncodeBase64(MakeUCharSpan(ssTx));
+    return EncodeBase64(ssTx);
 },
     };
 }
@@ -1553,7 +1553,7 @@ static RPCHelpMan converttopsbt()
     CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
     ssTx << psbtx;
 
-    return EncodeBase64(MakeUCharSpan(ssTx));
+    return EncodeBase64(ssTx);
 },
     };
 }
@@ -1644,7 +1644,7 @@ static RPCHelpMan utxoupdatepsbt()
 
     CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
     ssTx << psbtx;
-    return EncodeBase64(MakeUCharSpan(ssTx));
+    return EncodeBase64(ssTx);
 },
     };
 }
@@ -1740,7 +1740,7 @@ static RPCHelpMan joinpsbts()
 
     CDataStream ssTx(SER_NETWORK, PROTOCOL_VERSION);
     ssTx << shuffled_psbt;
-    return EncodeBase64(MakeUCharSpan(ssTx));
+    return EncodeBase64(ssTx);
 },
     };
 }

src/streams.h

@@ -6,17 +6,19 @@
 #ifndef BITCOIN_STREAMS_H
 #define BITCOIN_STREAMS_H
 
-#include <support/allocators/zeroafterfree.h>
 #include <serialize.h>
+#include <span.h>
+#include <support/allocators/zeroafterfree.h>
 
 #include <algorithm>
 #include <assert.h>
 #include <ios>
 #include <limits>
+#include <optional>
 #include <stdint.h>
 #include <stdio.h>
-#include <string>
 #include <string.h>
+#include <string>
 #include <utility>
 #include <vector>
 
@@ -202,14 +204,14 @@ class VectorReader
 class CDataStream
 {
 protected:
-    typedef CSerializeData vector_type;
+    using vector_type = SerializeData;
     vector_type vch;
-    unsigned int nReadPos;
+    unsigned int nReadPos{0};
 
     int nType;
     int nVersion;
-public:
 
+public:
     typedef vector_type::allocator_type   allocator_type;
     typedef vector_type::size_type        size_type;
     typedef vector_type::difference_type  difference_type;
@@ -221,62 +223,22 @@ class CDataStream
     typedef vector_type::reverse_iterator reverse_iterator;
 
     explicit CDataStream(int nTypeIn, int nVersionIn)
-    {
-        Init(nTypeIn, nVersionIn);
-    }
-
-    CDataStream(const_iterator pbegin, const_iterator pend, int nTypeIn, int nVersionIn) : vch(pbegin, pend)
-    {
-        Init(nTypeIn, nVersionIn);
-    }
-
-    CDataStream(const char* pbegin, const char* pend, int nTypeIn, int nVersionIn) : vch(pbegin, pend)
-    {
-        Init(nTypeIn, nVersionIn);
-    }
-
-    CDataStream(const vector_type& vchIn, int nTypeIn, int nVersionIn) : vch(vchIn.begin(), vchIn.end())
-    {
-        Init(nTypeIn, nVersionIn);
-    }
-
-    CDataStream(const std::vector<char>& vchIn, int nTypeIn, int nVersionIn) : vch(vchIn.begin(), vchIn.end())
-    {
-        Init(nTypeIn, nVersionIn);
-    }
+        : nType{nTypeIn},
+          nVersion{nVersionIn} {}
 
-    CDataStream(const std::vector<unsigned char>& vchIn, int nTypeIn, int nVersionIn) : vch(vchIn.begin(), vchIn.end())
-    {
-        Init(nTypeIn, nVersionIn);
-    }
+    explicit CDataStream(Span<const uint8_t> sp, int nTypeIn, int nVersionIn)
+        : vch(sp.data(), sp.data() + sp.size()),
+          nType{nTypeIn},
+          nVersion{nVersionIn} {}
 
     template <typename... Args>
     CDataStream(int nTypeIn, int nVersionIn, Args&&... args)
+        : nType{nTypeIn},
+          nVersion{nVersionIn}
     {
-        Init(nTypeIn, nVersionIn);
         ::SerializeMany(*this, std::forward<Args>(args)...);
     }
 
-    void Init(int nTypeIn, int nVersionIn)
-    {
-        nReadPos = 0;
-        nType = nTypeIn;
-        nVersion = nVersionIn;
-    }
-
-    CDataStream& operator+=(const CDataStream& b)
-    {
-        vch.insert(vch.end(), b.begin(), b.end());
-        return *this;
-    }
-
-    friend CDataStream operator+(const CDataStream& a, const CDataStream& b)
-    {
-        CDataStream ret = a;
-        ret += b;
-        return (ret);
-    }
-
     std::string str() const
     {
         return (std::string(begin(), end()));
@@ -297,12 +259,12 @@ class CDataStream
     const_reference operator[](size_type pos) const  { return vch[pos + nReadPos]; }
     reference operator[](size_type pos)              { return vch[pos + nReadPos]; }
     void clear()                                     { vch.clear(); nReadPos = 0; }
-    iterator insert(iterator it, const char x=char()) { return vch.insert(it, x); }
-    void insert(iterator it, size_type n, const char x) { vch.insert(it, n, x); }
+    iterator insert(iterator it, const uint8_t x) { return vch.insert(it, x); }
+    void insert(iterator it, size_type n, const uint8_t x) { vch.insert(it, n, x); }
     value_type* data()                               { return vch.data() + nReadPos; }
     const value_type* data() const                   { return vch.data() + nReadPos; }
 
-    void insert(iterator it, std::vector<char>::const_iterator first, std::vector<char>::const_iterator last)
+    void insert(iterator it, std::vector<uint8_t>::const_iterator first, std::vector<uint8_t>::const_iterator last)
     {
         if (last == first) return;
         assert(last - first > 0);
@@ -373,12 +335,17 @@ class CDataStream
         nReadPos = 0;
     }
 
-    bool Rewind(size_type n)
+    bool Rewind(std::optional<size_type> n = std::nullopt)
     {
+        // Total rewind if no size is passed
+        if (!n) {
+            nReadPos = 0;
+            return true;
+        }
         // Rewind by n characters if the buffer hasn't been compacted yet
-        if (n > nReadPos)
+        if (*n > nReadPos)
             return false;
-        nReadPos -= n;
+        nReadPos -= *n;
         return true;
     }
 
@@ -462,11 +429,6 @@ class CDataStream
         return (*this);
     }
 
-    void GetAndClear(CSerializeData &d) {
-        d.insert(d.end(), begin(), end());
-        clear();
-    }
-
     /**
      * XOR the contents of this stream with a certain key.
      *

src/support/allocators/zeroafterfree.h

@@ -42,7 +42,7 @@ struct zero_after_free_allocator : public std::allocator<T> {
     }
 };
 
-// Byte-vector that clears its contents before deletion.
-typedef std::vector<char, zero_after_free_allocator<char> > CSerializeData;
+/** Byte-vector that clears its contents before deletion. */
+using SerializeData = std::vector<uint8_t, zero_after_free_allocator<uint8_t>>;
 
 #endif // BITCOIN_SUPPORT_ALLOCATORS_ZEROAFTERFREE_H