binary_float(double) support?

phoenix610

Member²
Hello.

As the subject, Does the DOA 4.0.6 support Oracle10g binary_float(double) data type?

Using the DOA 4.0.6, the binary_float data type is binded to the Delphi's TStringField, but
it seems that the binary_float value is truncated or rounded.
the binary_float value using dbGo differs from
that using the DOA.
Which accuires correct values from DB, the DOA or the dbGo?
 
There is currently no explicit support for the binary_float data type, but this will be added in the next patch release. It will be treated as an otFloat field or variable.
 
Hello.

I'm checking out the feature of DOA binary_float support.
I have one question about DOA binary_float suport.

Now, I have a table that has a number(8) field, a number field and a binary_float field.

I want to treat a number(8) field as TIntegerField, a number field as TStringField,
and a binary_float field as TFloatField.

If I set TOracleSession.Preferences.FloatPrecision property to 15 or 16, a binary_float field binds to Delphi TStringField.

What should I do??

I want to treat a Oracle numeric field as a Delphi numeric field as far as I can do so
because I want to sort a data in a "corret" order at the client side by using Hi-Featured Grid such as DevExpress QuauntumGrid.

I'd like to say that a "corret" order means not a "string" order but a "numeric" order.
 
The precision of a binary_float is less than 15 or 16 digits, so it logically it should have been treated as a TFloatField with this setup. My guess is that the binary_float precision is not formally reported by Oracle though, and this results in the behavior you found. We'll try to fix it.
 
Hi.

It's no problem to wait for next release.

I'd like only to know whether this issue can be fixed or not.

I'm looking forward to next release.

Thanks, Marco.
 
Back
Top