From: | Stephen Davies <scldad(at)sdc(dot)com(dot)au> |
---|---|
To: | Sbragion Denis <infotecn(at)tin(dot)it> |
Cc: | pgsql-interfaces(at)postgreSQL(dot)org |
Subject: | Re: [INTERFACES] VB and ODBC: found |
Date: | 1998-04-27 12:11:18 |
Message-ID: | 199804271211.VAA24591@mustang.sdc.com.au |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-interfaces |
> Hello,
>
> after a fast look I found a reference in the MS knowledge base:
>
> http://support.microsoft.com/support/kb/articles/q109/8/00.asp
>
> little excerpt:
>
> Before you can edit an external ODBC table, the table must contain a unique
> index. If you get the following error message, you might not have a unique
> index on the table:
>
> Can't perform operation; it is illegal.
>
> You could also receive this error if the Data1.Recordset.Updatable flag is
> not set to True. Also, if you set the Data1.Options property to 64
> (SQL_PASSTHROUGH), the data control will not be updatable.
>
OK. Maybe that is the difference. I never use data controls to access ODBC
databases: the performance is awful.
There also seem to be strange constraints placed by MS on some types of
recordsets: the "safest" seem to be tables rather than dynasets etc.
I always use direct calls to the ODBC DLL. This is quite easy and, in general,
gives performance comparable with native SQL commands - and does not require
unique indexes or other strange MS policies..
Having said that, my personal preference is to have a unique key in every
table and if there is no obvious candidate, to make that key a
system-generated sequence number.
Cheers,
Stephen.
========================================================================
Stephen Davies Consulting scldad(at)sdc(dot)com(dot)au
Adelaide, South Australia. Voice: 61-8-82728863
Computing & Network solutions. Fax: 61-8-82741015
From | Date | Subject | |
---|---|---|---|
Next Message | David Hartwig | 1998-04-27 12:14:28 | Re: [INTERFACES] Odbc and Visual Basic |
Previous Message | David Hartwig | 1998-04-27 12:00:57 | Re: [INTERFACES] graphic-objects |