Re: [HACKERS] just another standards question

Thomas G. Lockhart writes:
> I had found a draft of the SQL92 standard on-line somewhere many months
> ago; don't remember where. It is ~1.6MB uncompressed, probably much
> smaller compressed; do you want me to send you a copy?

Yes, please.

> Anyway, it looks like Informix and Sybase have it right:
>
> 4.18.3 Indicator parameters
>
> An indicator parameter is an integer parameter that is specified
> immediately following another parameter. Its primary use is to
> indicate whether the value that the other parameter assumes or
> supplies is a null value. An indicator parameter cannot immediately
> follow another indicator parameter.

Okay, that works completely.

> The other use for indicator parameters is to indicate whether
> string data truncation occurred during a transfer between a host
> program and an SQL-implementation in parameters or host variables.
> If a non-null string value is transferred and the length of the
> target data item is sufficient to accept the entire source data
> item, then the indicator parameter or variable is set to 0 to in-
> dicate that truncation did not occur. However, if the length of
> the target data item is insufficient, then the indicator parame-
> ter or variable is set to the length of the source data item (in
> characters or bits, as appropriate) to indicate that truncation
> occurred and to indicate the original length in characters or bits,
> as appropriate, of the source.

So I have to set the indicator also during writes? That is if I insert
string "foobar" into char(3), do I have to set the indicator to 3?

Michael

--
Dr. Michael Meskes, Project-Manager | topsystem Systemhaus GmbH
meskes(at)topsystem(dot)de | Europark A2, Adenauerstr. 20
meskes(at)debian(dot)org | 52146 Wuerselen
Go SF49ers! Go Rhein Fire! | Tel: (+49) 2405/4670-44
Use Debian GNU/Linux! | Fax: (+49) 2405/4670-10