Subject Re: [Firebird-Architect] RFC: Proposal for the implementation
Author Nando Dessena
Ann,

>> Not to mention that stuffing all
>>that info in a single column violates the most basic normalization
>>rules, which is ironic for a field whose name starts with RDB. ;-)

A> Would you prefer to see a series of fields like these?

A> RDB$PERMANENT_RELATION char(1)CHECK (VALUE IN ('Y', 'N');
A> RDB$ON_COMMIT_PRESERVE char(1)CHECK (VALUE IN ('Y', 'N');
A> RDB$GLOBAL char (1)CHECK (VALUE IN ('Y', 'N');

if it's a matter of preference, then I'd like the bitmask, the
translation in RDB$TYPES (or a set of predefined constants) and the
bit_and and bit_or operators.

Ciao
--
Nando Dessena
http://www.flamerobin.org