#include <qcbor_main_decode.h>

Data Fields
int64_t	nExponent

union {

int64_t nInt

uint64_t uInt

UsefulBufC bigNum

}	Mantissa

Detailed Description

This holds the value for big floats and decimal fractions, as an exponent and mantissa. For big floats the base for exponentiation is 2. For decimal fractions it is 10. Whether an instance is a big float or decimal fraction is known by context, usually by uDataType in QCBORItem which might be QCBOR_TYPE_DECIMAL_FRACTION, QCBOR_TYPE_BIGFLOAT, ...

The mantissa may be an int64_t or a big number. This is again determined by context, usually uDataType in QCBORItem which might be QCBOR_TYPE_DECIMAL_FRACTION, QCBOR_TYPE_DECIMAL_FRACTION_POS_BIGNUM, ... The sign of the big number also comes from the context (QCBOR_TYPE_DECIMAL_FRACTION_POS_BIGNUM, QCBOR_TYPE_DECIMAL_FRACTION_NEG_BIGNUM,...).

bigNum is big endian or network byte order. The most significant byte is first.

When Mantissa is int64_t, it represents the true value of the mantissa with the offset of 1 for CBOR negative values applied. When it is a negative big number (QCBOR_TYPE_DECIMAL_FRACTION_NEG_BIGNUM or QCBOR_TYPE_BIGFLOAT_NEG_BIGNUM), the offset of 1 has NOT been applied (doing so requires somewhat complex big number arithmetic and may increase the length of the big number). To get the correct value bigNum must be incremented by one before use.

Also see QCBOREncode_AddTDecimalFraction(), QCBOREncode_AddTBigFloat(), QCBOREncode_AddTDecimalFractionBigNum() and QCBOREncode_AddTBigFloatBigNum().

The documentation for this struct was generated from the following file:

/Users/lgl/Code/QCBOR/dev/inc/qcbor/qcbor_main_decode.h

Data Fields

Detailed Description