Question

I’m developing a C++ library with oracle OCI library. I’m loading from a Oracle Database 11g database. Here I need to load large values with decimal places. In this case I need to use long double instead of double. I’m not sure whether OCI supports for long double.
According to the documentation SQLT_FLT is for float and double only. Can someone let me know whether OCI supports for long double and if so how to retrieve them

Answer 1

Yes and no.

According to the server docs on floats, the ‘native’ C types support up to double precision (BINARY_DOUBLE). However, the NUMBER type can store:

• Positive numbers in the range 1 x 10-130 to 9.99…9 x 10125 with up to 38 significant digits
• Negative numbers from -1 x 10-130 to 9.99…99 x 10125 with up to 38 significant digits

which is more precision than long double on x86/amd64.

So, you’d need to use that type instead of BINARY_*. Sadly, the developer docs say:

You should not need to use NUMBER as an external datatype. If you do use it, Oracle returns numeric values in its internal 21-byte binary format and will expect this format on input. The following discussion is included for completeness only.

On the other hand, there are also docs on:

• using the OCINumber type;
• OCI NUMBER Functions.

AFAICS OCINumberFromReal() supports long double; and if you want even more precision, then you could use OCINumberFromText() with decimal string.