Subject | Re: [ib-support] Problems with gbak |
---|---|
Author | Ann W. Harrison |
Post date | 2002-03-21T15:32:14Z |
At 12:27 AM 3/21/2002 -0400, Claudio Valderrama C. wrote:
alignment, starting with quads and doubles, then longs, then shorts
and varchars, then chars.
What I have seen, on occasion, is a column (rdb$fields record) with
a type (rdb$field_type) that evaluates to long (8) and a length
(rdb$field_length) of 2, or, conversely, a type of short (7) and
a length of 4. Gbak naively believes the length, but the engine
believes the type. I believe there's a bug report suggesting fixing
gbak to check length/type and the engine to do the same one updates.
Regards,
Ann
www.ibphoenix.com
We have answers.
> > > Watch this chunk, what's the value of offset? And length?Actually, gbak is very careful to order fields in its messages by
> >
> > record_length = 46;
> > length = 48;
>
>Nothing special. The code in that function found that it needed an alignment
>for some field and this raised the length from 46 to 48.
alignment, starting with quads and doubles, then longs, then shorts
and varchars, then chars.
What I have seen, on occasion, is a column (rdb$fields record) with
a type (rdb$field_type) that evaluates to long (8) and a length
(rdb$field_length) of 2, or, conversely, a type of short (7) and
a length of 4. Gbak naively believes the length, but the engine
believes the type. I believe there's a bug report suggesting fixing
gbak to check length/type and the engine to do the same one updates.
Regards,
Ann
www.ibphoenix.com
We have answers.