Cxx: Initial version of the sources form C++ to C migration.

java/nativeWrappers/build.gradle

@@ -16,7 +16,7 @@ task makeNativeWrappersDfp(dependsOn: compileJava, type: JavaExec) {
     args = [ver["three"], ver["suffix"], ver["sha"],
             "ddfp" + ver["suffix"] + "_",
             "com_epam_deltix_dfp_NativeImpl_",
-            "$rootDir/native/NativeImpl.cpp",
+            "$rootDir/native/NativeImpl.c",
             "$rootDir/java/dfp/build/generated/sources/nativeWrappers/com/epam/deltix/dfp/NativeImpl.java",
             "$rootDir/csharp/EPAM.Deltix.DFP/NativeImpl.cs",
             "$rootDir/native"

native/CMakeLists.txt

@@ -50,7 +50,7 @@ set(CMAKE_POSITION_INDEPENDENT_CODE ON)
 
 add_subdirectory (../thirdparty/IntelRDFPMathLib20U2/LIBRARY LIBRARY)
 
-add_library(dfp SHARED NativeImpl.cpp)
+add_library(dfp SHARED NativeImpl.c)
 target_compile_definitions(dfp PRIVATE API_PREFIX=ddfp${VERSION_SUFFIX}_ JAVA_PREFIX=com_epam_deltix_dfp_NativeImpl_ DECIMALNATIVE_SHARED_LIBRARY DECIMALNATIVE_EXPORTS)
 target_link_libraries(dfp LINK_PUBLIC bid)
 set_target_properties(dfp PROPERTIES OUTPUT_NAME "ddfp${VERSION_SUFFIX}")
@@ -84,14 +84,14 @@ file(GLOB BID_SRCS_STATIC
 		../thirdparty/IntelRDFPMathLib20U2/LIBRARY/float128/dpml_four_over_pi.c
 		../thirdparty/IntelRDFPMathLib20U2/LIBRARY/float128/dpml_exception.c
 		../thirdparty/IntelRDFPMathLib20U2/LIBRARY/float128/sqrt_tab_t.c
-		NativeImplToString.cpp)
+		NativeImplToString.c)
 
-add_library(dfpStatic STATIC NativeImpl.cpp ${BID_SRCS_STATIC})
+add_library(dfpStatic STATIC NativeImpl.c ${BID_SRCS_STATIC})
 include_directories(../thirdparty/IntelRDFPMathLib20U2/LIBRARY/src)
 target_compile_definitions(dfpStatic PRIVATE API_PREFIX=ddfp${VERSION_SUFFIX}_ NOJAVA)
 set_target_properties(dfpStatic PROPERTIES OUTPUT_NAME "ddfpStatic${VERSION_SUFFIX}")
 install(TARGETS dfpStatic DESTINATION ./bin/${CMAKE_BUILD_TYPE}/${INSTALL_SUFFIX})
 
 add_executable(demo demo.cpp)
 #target_link_libraries(demo LINK_PRIVATE dfp stdc++)
-target_link_libraries(demo LINK_PRIVATE dfpStatic stdc++)
+target_link_libraries(demo LINK_PRIVATE dfpStatic) # stdc++)

native/NativeImpl.c

@@ -0,0 +1,161 @@
+#include <string.h>
+#include <stdint.h>
+
+#include "NativeImpl.h"
+#include "NativeImplToString.h"
+
+#ifdef _MSC_VER
+#define strcmpIgnoreCase stricmp
+#else
+#define strcmpIgnoreCase strcasecmp
+#endif
+
+//region Conversion
+static const uint64_t DECIMAL_NATIVE_UNDERLYING_NULL = 0xFFFFFFFFFFFFFF80ULL;	// = -0x80
+
+BID_UINT64 dfp64_try_parse(const char* str, _IDEC_flags* exception) {
+	if (!strcmpIgnoreCase(str, "null"))
+		return DECIMAL_NATIVE_UNDERLYING_NULL;
+	auto ret = bid64_from_string((char*)str);
+	if (exception)
+		*exception = _IDEC_glbflags;
+	return ret;
+}
+
+BID_UINT64 dfp64_parse(const char* str) {
+	return dfp64_try_parse(str, 0 /*nullptr*/);
+}
+
+OPN(fromFloat64, binary64_to_bid64(x), double x)
+OPN(fromFloat32, binary32_to_bid64(x), float x)
+OPN(fromInt64, bid64_from_int64(x), int64 x)
+OPN(fromUInt64, bid64_from_uint64(x), uint64 x)
+OPNR(toFloat64, double, bid64_to_binary64(x), BID_UINT64 x)
+OPNR(toFloat32, float, bid64_to_binary32(x), BID_UINT64 x)
+OPNR(toInt64, int64, bid64_to_int64_xint(x), BID_UINT64 x)
+OPNR(toUInt64, uint64, bid64_to_uint64_xint(x), BID_UINT64 x)
+OPN(fromFixedPoint64, bid64_scalbn(bid64_from_int64(mantissa), -tenPowerFactor), int64 mantissa, int32 tenPowerFactor)
+OPNR(toFixedPoint, int64, bid64_to_int64_xint(bid64_scalbn(value, numberOfDigits)), BID_UINT64 value, int32 numberOfDigits)
+
+OPN(parse, dfp64_parse(str), const char* str)
+OPN(tryParse, dfp64_try_parse(str, exception), const char* str, uint32* exception)
+
+OPNR(to_string, const char*, dfp64_to_string(x), BID_UINT64 x)
+OPNR(to_string_2, const char*, dfp64_to_string_2(x, decimalMark), BID_UINT64 x, char decimalMark)
+OPNR(to_string_3, const char*, dfp64_to_string_3(x, decimalMark, buffer512), BID_UINT64 x, char decimalMark, char* buffer512)
+
+OPNR(to_scientific_string, const char*, dfp64_to_scientific_string(x), BID_UINT64 x)
+OPNR(to_scientific_string_2, const char*, dfp64_to_scientific_string_2(x, decimalMark), BID_UINT64 x, char decimalMark)
+OPNR(to_scientific_string_3, const char*, dfp64_to_scientific_string_3(x, decimalMark, buffer512), BID_UINT64 x, char decimalMark, char* buffer512)
+
+
+//endregion
+
+//region Classification
+
+OPN_BOOL(isNaN, bid64_isNaN(x), BID_UINT64 x)
+OPN_BOOL(isInfinity, bid64_isInf(x), BID_UINT64 x)
+OPN_BOOL(isPositiveInfinity, (intBool)(bid64_isInf(x) && !bid64_isSigned(x)), BID_UINT64 x)
+OPN_BOOL(isNegativeInfinity, (intBool)(bid64_isInf(x) && bid64_isSigned(x)), BID_UINT64 x)
+OPN_BOOL(isFinite, bid64_isFinite(x), BID_UINT64 x)
+OPN_BOOL(isNormal, bid64_isNormal(x), BID_UINT64 x)
+OPN_BOOL(signBit, bid64_isSigned(x), BID_UINT64 x)
+
+//endregion
+
+//region Comparison
+
+DDFP_API(int32) PPCAT(API_PREFIX, compare) (BID_UINT64 a, BID_UINT64 b) {
+	if (bid64_quiet_less(a, b))
+		return -1;
+	if (bid64_quiet_greater(a, b))
+		return 1;
+	if (bid64_quiet_equal(a, b))
+		return 0;
+	return bid64_isNaN(b) - bid64_isNaN(a);
+}
+JAVA_API_IMPL(
+	JNI_API(int32) PPCAT(PPCAT(Java_, JAVA_PREFIX), compare) (void* jEnv, void* jClass, BID_UINT64 a, BID_UINT64 b) {
+		if (bid64_quiet_less(a, b))
+			return -1;
+		if (bid64_quiet_greater(a, b))
+			return 1;
+		if (bid64_quiet_equal(a, b))
+			return 0;
+		return bid64_isNaN(a) - bid64_isNaN(b);
+	}
+	JNI_API(int32) PPCAT(PPCAT(JavaCritical_, JAVA_PREFIX), compare) (BID_UINT64 a, BID_UINT64 b) {
+		if (bid64_quiet_less(a, b))
+			return -1;
+		if (bid64_quiet_greater(a, b))
+			return 1;
+		if (bid64_quiet_equal(a, b))
+			return 0;
+		return bid64_isNaN(a) - bid64_isNaN(b);
+	}
+)
+
+OPN_BOOL(isEqual, bid64_quiet_equal(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN_BOOL(isNotEqual, bid64_quiet_not_equal(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN_BOOL(isLess, bid64_quiet_less(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN_BOOL(isLessOrEqual, bid64_quiet_less_equal(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN_BOOL(isGreater, bid64_quiet_greater(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN_BOOL(isGreaterOrEqual, bid64_quiet_greater_equal(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN_BOOL(isZero, bid64_isZero(a), BID_UINT64 a)
+OPN_BOOL(isNonZero, bid64_quiet_not_equal(a, zeroConst), BID_UINT64 a)
+OPN_BOOL(isPositive, bid64_quiet_greater(a, zeroConst), BID_UINT64 a)
+OPN_BOOL(isNegative, bid64_quiet_less(a, zeroConst), BID_UINT64 a)
+OPN_BOOL(isNonPositive, bid64_quiet_less_equal(a, zeroConst), BID_UINT64 a)
+OPN_BOOL(isNonNegative, bid64_quiet_greater_equal(a, zeroConst), BID_UINT64 a)
+
+//endregion
+
+//region Rounding
+
+OPN(roundTowardsPositiveInfinity, bid64_round_integral_positive(x), BID_UINT64 x)
+OPN(roundTowardsNegativeInfinity, bid64_round_integral_negative(x), BID_UINT64 x)
+OPN(roundTowardsZero, bid64_round_integral_zero(x), BID_UINT64 x)
+OPN(roundToNearestTiesAwayFromZero, bid64_round_integral_nearest_away(x), BID_UINT64 x)
+OPN(roundToNearestTiesToEven, bid64_round_integral_nearest_even(x), BID_UINT64 x)
+
+//endregion
+
+//region Minimum & Maximum
+
+OPN(max2, bid64_isNaN(a) || bid64_isNaN(b) ? nanConst : bid64_maxnum(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN(max3, bid64_isNaN(a) || bid64_isNaN(b) || bid64_isNaN(c) ? nanConst : bid64_maxnum(bid64_maxnum(a, b), c), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c)
+OPN(max4, bid64_isNaN(a) || bid64_isNaN(b) || bid64_isNaN(c) || bid64_isNaN(d) ? nanConst : bid64_maxnum(bid64_maxnum(a, b), bid64_maxnum(c, d)), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c, BID_UINT64 d)
+OPN(min2, bid64_isNaN(a) || bid64_isNaN(b) ? nanConst : bid64_minnum(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN(min3, bid64_isNaN(a) || bid64_isNaN(b) || bid64_isNaN(c) ? nanConst : bid64_minnum(bid64_minnum(a, b), c), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c)
+OPN(min4, bid64_isNaN(a) || bid64_isNaN(b) || bid64_isNaN(c) || bid64_isNaN(d) ? nanConst : bid64_minnum(bid64_minnum(a, b), bid64_minnum(c, d)), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c, BID_UINT64 d)
+
+//endregion
+
+//region Arithmetic
+
+OPN(negate, bid64_negate(x), BID_UINT64 x)
+OPN(abs, bid64_abs(x), BID_UINT64 x)
+OPN(add2, bid64_add(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN(add3, bid64_add(bid64_add(a, b), c), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c)
+OPN(add4, bid64_add(bid64_add(a, b), bid64_add(c, d)), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c, BID_UINT64 d)
+OPN(subtract, bid64_sub(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN(multiply2, bid64_mul(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN(multiply3, bid64_mul(bid64_mul(a, b), c), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c)
+OPN(multiply4, bid64_mul(bid64_mul(a, b), bid64_mul(c, d)), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c, BID_UINT64 d)
+OPN(multiplyByInt32, bid64_mul(a, bid64_from_int32(integer)), BID_UINT64 a, int32 integer)
+OPN(multiplyByInt64, bid64_mul(a, bid64_from_int64(integer)), BID_UINT64 a, int64 integer)
+OPN(divide, bid64_div(a, b), BID_UINT64 a, BID_UINT64 b)
+OPN(divideByInt32, bid64_div(x, bid64_from_int32(integer)), BID_UINT64 x, int32 integer)
+OPN(divideByInt64, bid64_div(x, bid64_from_int64(integer)), BID_UINT64 x, int64 integer)
+OPN(multiplyAndAdd, bid64_fma(a, b, c), BID_UINT64 a, BID_UINT64 b, BID_UINT64 c)
+OPN(scaleByPowerOfTen, bid64_scalbn(a, n), BID_UINT64 a, int32 n)
+OPN(mean2, bid64_div(bid64_add(a, b), twoConst), BID_UINT64 a, BID_UINT64 b)
+
+//endregion
+
+//region Special
+
+OPN(nextUp, bid64_nextup(x), BID_UINT64 x)
+OPN(nextDown, bid64_nextdown(x), BID_UINT64 x)
+
+//endregion