Re: Bug? Determining lengths of CLOB's in trigger/Oracle 8.1.7

"Sybrand Bakker" <gooiditweg_at_sybrandb.nospam.demon.nl> insulted another person in message news:adpsrvcfv4b3tpo9j5maf8o58uv4j1bahp_at_4ax.com...

> connecting to a 8i database, and lousy trigger code. You shouldn't
> insert extra records in the table you are firing a trigger on, or you

He isn't. Reread his code.

> I have the strange feeling when you just add an extra number column to
> the table, and manipulate the length column in a before insert
> trigger, as it should have been done in the first place, your

Why should he have done this in the first place? There isn't enough information in his post to determine why he wants to do what he does. There is more than enough information to see things are not working as I would expect and as he has documented. And if as you suggest he does add a new column, and it works, why should it work. You need to explain this.

P.F. Ferdaz.

"Sybrand Bakker" <gooiditweg_at_sybrandb.nospam.demon.nl> wrote in message news:adpsrvcfv4b3tpo9j5maf8o58uv4j1bahp_at_4ax.com...
> On 21 Nov 2003 09:02:02 -0800, bbulsara23_at_hotmail.com (barry) wrote:
>
> >I'd
> >appreciate someone shining a light here if at all possible, or
> >confirmation of the bug.
>
>
> Likely the 'bug' consists of the combination of a 9i sqlloader
> connecting to a 8i database, and lousy trigger code. You shouldn't
> insert extra records in the table you are firing a trigger on, or you
> must at least make sure you use an autonomous transaction.
> I have the strange feeling when you just add an extra number column to
> the table, and manipulate the length column in a before insert
> trigger, as it should have been done in the first place, your
> so-called 'bug' disappears instantly.
>
>
> --
> Sybrand Bakker, Senior Oracle DBA
Received on Fri Nov 21 2003 - 16:05:16 CST