refactor: change the return type of mulDiv to std::optional (XRPLF#4243)

- Previously, mulDiv had `std::pair<bool, uint64_t>` as the output type. - This is an error-prone interface as it is easy to ignore when overflow occurs. - Using a return type of `std::optional` should decrease the likelihood of ignoring overflow. - It also allows for the use of optional::value_or() as a way to explicitly recover from overflow. - Include limits.h header file preprocessing directive in order to satisfy gcc's numeric_limits incomplete_type requirement. Fix XRPLF#3495 --------- Co-authored-by: John Freeman <[email protected]>
ckeshava · Jul 10, 2023 · 95d67eb · 95d67eb
1 parent a6664eb
commit 95d67eb
Show file tree

Hide file tree

Showing 12 changed files with 93 additions and 83 deletions.
diff --git a/src/ripple/app/misc/NetworkOPs.cpp b/src/ripple/app/misc/NetworkOPs.cpp
@@ -2058,7 +2058,7 @@ NetworkOPsImp::pubServer()
                     f.em->openLedgerFeeLevel,
                     f.loadBaseServer,
                     f.em->referenceFeeLevel)
-                    .second);
+                    .value_or(ripple::muldiv_max));
 
             jvObj[jss::load_factor] = trunc32(loadFactor);
             jvObj[jss::load_factor_fee_escalation] =
@@ -2506,7 +2506,7 @@ NetworkOPsImp::getServerInfo(bool human, bool admin, bool counters)
                 escalationMetrics.openLedgerFeeLevel,
                 loadBaseServer,
                 escalationMetrics.referenceFeeLevel)
-                .second;
+                .value_or(ripple::muldiv_max);
 
         auto const loadFactor = std::max(
             safe_cast<std::uint64_t>(loadFactorServer),

diff --git a/src/ripple/app/misc/TxQ.h b/src/ripple/app/misc/TxQ.h
@@ -492,7 +492,7 @@ class TxQ
             @param seriesSize Total number of transactions in the series to be
                 processed.
 
-            @return A `std::pair` as returned from @ref `mulDiv` indicating
+            @return A `std::pair` indicating
                 whether the calculation result overflows.
         */
         static std::pair<bool, FeeLevel64>
@@ -862,20 +862,15 @@ template <class T>
 XRPAmount
 toDrops(FeeLevel<T> const& level, XRPAmount baseFee)
 {
-    if (auto const drops = mulDiv(level, baseFee, TxQ::baseLevel); drops.first)
-        return drops.second;
-
-    return XRPAmount(STAmount::cMaxNativeN);
+    return mulDiv(level, baseFee, TxQ::baseLevel)
+        .value_or(XRPAmount(STAmount::cMaxNativeN));
 }
 
 inline FeeLevel64
 toFeeLevel(XRPAmount const& drops, XRPAmount const& baseFee)
 {
-    if (auto const feeLevel = mulDiv(drops, TxQ::baseLevel, baseFee);
-        feeLevel.first)
-        return feeLevel.second;
-
-    return FeeLevel64(std::numeric_limits<std::uint64_t>::max());
+    return mulDiv(drops, TxQ::baseLevel, baseFee)
+        .value_or(FeeLevel64(std::numeric_limits<std::uint64_t>::max()));
 }
 
 }  // namespace ripple

diff --git a/src/ripple/app/misc/impl/LoadFeeTrack.cpp b/src/ripple/app/misc/impl/LoadFeeTrack.cpp
@@ -109,9 +109,9 @@ scaleFeeLoad(
 
     auto const result = mulDiv(
         fee, feeFactor, safe_cast<std::uint64_t>(feeTrack.getLoadBase()));
-    if (!result.first)
+    if (!result)
         Throw<std::overflow_error>("scaleFeeLoad");
-    return result.second;
+    return *result;
 }
 
 }  // namespace ripple
diff --git a/src/ripple/app/misc/impl/TxQ.cpp b/src/ripple/app/misc/impl/TxQ.cpp
@@ -58,12 +58,8 @@ getFeeLevelPaid(ReadView const& view, STTx const& tx)
         return FeeLevel64(0);
     }
 
-    if (std::pair<bool, FeeLevel64> const feeLevelPaid =
-            mulDiv(effectiveFeePaid, TxQ::baseLevel, baseFee);
-        feeLevelPaid.first)
-        return feeLevelPaid.second;
-
-    return FeeLevel64(std::numeric_limits<std::uint64_t>::max());
+    return mulDiv(effectiveFeePaid, TxQ::baseLevel, baseFee)
+        .value_or(FeeLevel64(std::numeric_limits<std::uint64_t>::max()));
 }
 
 static std::optional<LedgerIndex>
@@ -77,7 +73,8 @@ getLastLedgerSequence(STTx const& tx)
 static FeeLevel64
 increase(FeeLevel64 level, std::uint32_t increasePercent)
 {
-    return mulDiv(level, 100 + increasePercent, 100).second;
+    return mulDiv(level, 100 + increasePercent, 100)
+        .value_or(static_cast<FeeLevel64>(ripple::muldiv_max));
 }
 
 //////////////////////////////////////////////////////////////////////////
@@ -114,16 +111,19 @@ TxQ::FeeMetrics::update(
         // upperLimit must be >= minimumTxnCount_ or std::clamp can give
         // unexpected results
         auto const upperLimit = std::max<std::uint64_t>(
-            mulDiv(txnsExpected_, cutPct, 100).second, minimumTxnCount_);
+            mulDiv(txnsExpected_, cutPct, 100).value_or(ripple::muldiv_max),
+            minimumTxnCount_);
         txnsExpected_ = std::clamp<std::uint64_t>(
-            mulDiv(size, cutPct, 100).second, minimumTxnCount_, upperLimit);
+            mulDiv(size, cutPct, 100).value_or(ripple::muldiv_max),
+            minimumTxnCount_,
+            upperLimit);
         recentTxnCounts_.clear();
     }
     else if (size > txnsExpected_ || size > targetTxnCount_)
     {
         recentTxnCounts_.push_back(
             mulDiv(size, 100 + setup.normalConsensusIncreasePercent, 100)
-                .second);
+                .value_or(ripple::muldiv_max));
         auto const iter =
             std::max_element(recentTxnCounts_.begin(), recentTxnCounts_.end());
         BOOST_ASSERT(iter != recentTxnCounts_.end());
@@ -181,7 +181,8 @@ TxQ::FeeMetrics::scaleFeeLevel(Snapshot const& snapshot, OpenView const& view)
     {
         // Compute escalated fee level
         // Don't care about the overflow flag
-        return mulDiv(multiplier, current * current, target * target).second;
+        return mulDiv(multiplier, current * current, target * target)
+            .value_or(static_cast<FeeLevel64>(ripple::muldiv_max));
     }
 
     return baseLevel;
@@ -264,7 +265,7 @@ TxQ::FeeMetrics::escalatedSeriesFeeLevel(
     auto const totalFeeLevel = mulDiv(
         multiplier, sumNlast.second - sumNcurrent.second, target * target);
 
-    return totalFeeLevel;
+    return {totalFeeLevel.has_value(), *totalFeeLevel};
 }
 
 LedgerHash TxQ::MaybeTx::parentHashComp{};
@@ -1782,8 +1783,11 @@ TxQ::getTxRequiredFeeAndSeq(
 
     std::uint32_t const accountSeq = sle ? (*sle)[sfSequence] : 0;
     std::uint32_t const availableSeq = nextQueuableSeqImpl(sle, lock).value();
-
-    return {mulDiv(fee, baseFee, baseLevel).second, accountSeq, availableSeq};
+    return {
+        mulDiv(fee, baseFee, baseLevel)
+            .value_or(XRPAmount(std::numeric_limits<std::int64_t>::max())),
+        accountSeq,
+        availableSeq};
 }
 
 std::vector<TxQ::TxDetails>

diff --git a/src/ripple/basics/FeeUnits.h b/src/ripple/basics/FeeUnits.h
@@ -409,7 +409,7 @@ template <
     class Source2,
     class Dest,
     class = enable_muldiv_t<Source1, Source2, Dest>>
-std::pair<bool, Dest>
+std::optional<Dest>
 mulDivU(Source1 value, Dest mul, Source2 div)
 {
     // Fees can never be negative in any context.
@@ -420,20 +420,20 @@ mulDivU(Source1 value, Dest mul, Source2 div)
         assert(value.value() >= 0);
         assert(mul.value() >= 0);
         assert(div.value() >= 0);
-        return {false, Dest{0}};
+        return std::nullopt;
     }
 
     using desttype = typename Dest::value_type;
     constexpr auto max = std::numeric_limits<desttype>::max();
 
     // Shortcuts, since these happen a lot in the real world
     if (value == div)
-        return {true, mul};
+        return mul;
     if (mul.value() == div.value())
     {
         if (value.value() > max)
-            return {false, Dest{max}};
-        return {true, Dest{static_cast<desttype>(value.value())}};
+            return std::nullopt;
+        return Dest{static_cast<desttype>(value.value())};
     }
 
     using namespace boost::multiprecision;
@@ -447,9 +447,9 @@ mulDivU(Source1 value, Dest mul, Source2 div)
     auto quotient = product / div.value();
 
     if (quotient > max)
-        return {false, Dest{max}};
+        return std::nullopt;
 
-    return {true, Dest{static_cast<desttype>(quotient)}};
+    return Dest{static_cast<desttype>(quotient)};
 }
 
 }  // namespace feeunit
@@ -464,7 +464,7 @@ template <
     class Source2,
     class Dest,
     class = feeunit::enable_muldiv_t<Source1, Source2, Dest>>
-std::pair<bool, Dest>
+std::optional<Dest>
 mulDiv(Source1 value, Dest mul, Source2 div)
 {
     return feeunit::mulDivU(value, mul, div);
@@ -475,15 +475,15 @@ template <
     class Source2,
     class Dest,
     class = feeunit::enable_muldiv_commute_t<Source1, Source2, Dest>>
-std::pair<bool, Dest>
+std::optional<Dest>
 mulDiv(Dest value, Source1 mul, Source2 div)
 {
     // Multiplication is commutative
     return feeunit::mulDivU(mul, value, div);
 }
 
 template <class Dest, class = feeunit::enable_muldiv_dest_t<Dest>>
-std::pair<bool, Dest>
+std::optional<Dest>
 mulDiv(std::uint64_t value, Dest mul, std::uint64_t div)
 {
     // Give the scalars a non-tag so the
@@ -492,7 +492,7 @@ mulDiv(std::uint64_t value, Dest mul, std::uint64_t div)
 }
 
 template <class Dest, class = feeunit::enable_muldiv_dest_t<Dest>>
-std::pair<bool, Dest>
+std::optional<Dest>
 mulDiv(Dest value, std::uint64_t mul, std::uint64_t div)
 {
     // Multiplication is commutative
@@ -503,20 +503,24 @@ template <
     class Source1,
     class Source2,
     class = feeunit::enable_muldiv_sources_t<Source1, Source2>>
-std::pair<bool, std::uint64_t>
+std::optional<std::uint64_t>
 mulDiv(Source1 value, std::uint64_t mul, Source2 div)
 {
     // Give the scalars a dimensionless unit so the
     // unit-handling version gets called.
     auto unitresult = feeunit::mulDivU(value, feeunit::scalar(mul), div);
-    return {unitresult.first, unitresult.second.value()};
+
+    if (!unitresult)
+        return std::nullopt;
+
+    return unitresult->value();
 }
 
 template <
     class Source1,
     class Source2,
     class = feeunit::enable_muldiv_sources_t<Source1, Source2>>
-std::pair<bool, std::uint64_t>
+std::optional<std::uint64_t>
 mulDiv(std::uint64_t value, Source1 mul, Source2 div)
 {
     // Multiplication is commutative

diff --git a/src/ripple/basics/impl/mulDiv.cpp b/src/ripple/basics/impl/mulDiv.cpp
@@ -17,15 +17,13 @@
 */
 //==============================================================================
 
-#include <ripple/basics/contract.h>
 #include <ripple/basics/mulDiv.h>
 #include <boost/multiprecision/cpp_int.hpp>
-#include <limits>
-#include <utility>
+#include <optional>
 
 namespace ripple {
 
-std::pair<bool, std::uint64_t>
+std::optional<std::uint64_t>
 mulDiv(std::uint64_t value, std::uint64_t mul, std::uint64_t div)
 {
     using namespace boost::multiprecision;
@@ -35,12 +33,10 @@ mulDiv(std::uint64_t value, std::uint64_t mul, std::uint64_t div)
 
     result /= div;
 
-    auto constexpr limit = std::numeric_limits<std::uint64_t>::max();
+    if (result > ripple::muldiv_max)
+        return std::nullopt;
 
-    if (result > limit)
-        return {false, limit};
-
-    return {true, static_cast<std::uint64_t>(result)};
+    return static_cast<std::uint64_t>(result);
 }
 
 }  // namespace ripple
diff --git a/src/ripple/basics/mulDiv.h b/src/ripple/basics/mulDiv.h
@@ -21,24 +21,24 @@
 #define RIPPLE_BASICS_MULDIV_H_INCLUDED
 
 #include <cstdint>
+#include <limits>
+#include <optional>
 #include <utility>
 
 namespace ripple {
+auto constexpr muldiv_max = std::numeric_limits<std::uint64_t>::max();
 
 /** Return value*mul/div accurately.
     Computes the result of the multiplication and division in
     a single step, avoiding overflow and retaining precision.
     Throws:
         None
     Returns:
-        `std::pair`:
-            `first` is `false` if the calculation overflows,
-                `true` if the calculation is safe.
-            `second` is the result of the calculation if
-                `first` is `false`, max value of `uint64_t`
-                if `true`.
+        `std::optional`:
+            `std::nullopt` if the calculation overflows. Otherwise, `value * mul
+   / div`.
 */
-std::pair<bool, std::uint64_t>
+std::optional<std::uint64_t>
 mulDiv(std::uint64_t value, std::uint64_t mul, std::uint64_t div);
 
 }  // namespace ripple

diff --git a/src/ripple/rpc/handlers/Fee1.cpp b/src/ripple/rpc/handlers/Fee1.cpp
@@ -20,7 +20,6 @@
 #include <ripple/app/ledger/OpenLedger.h>
 #include <ripple/app/main/Application.h>
 #include <ripple/app/misc/TxQ.h>
-#include <ripple/basics/mulDiv.h>
 #include <ripple/protocol/ErrorCodes.h>
 #include <ripple/protocol/Feature.h>
 #include <ripple/rpc/Context.h>

diff --git a/src/ripple/rpc/impl/TransactionSign.cpp b/src/ripple/rpc/impl/TransactionSign.cpp
@@ -738,9 +738,9 @@ checkFee(
     auto const limit = [&]() {
         // Scale fee units to drops:
         auto const result = mulDiv(feeDefault, mult, div);
-        if (!result.first)
+        if (!result)
             Throw<std::overflow_error>("mulDiv");
-        return result.second;
+        return *result;
     }();
 
     if (fee > limit)

diff --git a/src/test/app/TxQ_test.cpp b/src/test/app/TxQ_test.cpp
@@ -22,7 +22,6 @@
 #include <ripple/app/misc/TxQ.h>
 #include <ripple/app/tx/apply.h>
 #include <ripple/basics/Log.h>
-#include <ripple/basics/mulDiv.h>
 #include <ripple/protocol/ErrorCodes.h>
 #include <ripple/protocol/jss.h>
 #include <ripple/protocol/st.h>