We currently always print bit-vector constants using binary representation. This can lead to quite large values in models when using 32- or 64-bit values that are essentially unreadable. When the bit-vector width is a multiple of 4, this patch uses hexadecimal notation instead, as allowed by the SMT-LIB specification.
Note that this only applies to bit-vector terms; bit-vector semantic values are always printed in their binary representation for debugging purposes.
We currently always print bit-vector constants using binary representation. This can lead to quite large values in models when using 32- or 64-bit values that are essentially unreadable. When the bit-vector width is a multiple of 4, this patch uses hexadecimal notation instead, as allowed by the SMT-LIB specification.
Note that this only applies to bit-vector terms; bit-vector semantic values are always printed in their binary representation for debugging purposes.