Re: Access2000 & sequence as primary key in view : #DELETED

From: Arnaud Lesauvage <thewild(at)freesurf(dot)fr>
To: Hiroshi Inoue <inoue(at)tpf(dot)co(dot)jp>
Cc: pgsql-odbc(at)postgresql(dot)org
Subject: Re: Access2000 & sequence as primary key in view : #DELETED
Date: 2006-10-12 07:48:50
Message-ID: 452DF362.50307@freesurf.fr
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-odbc

Hiroshi Inoue a écrit :
> Arnaud Lesauvage wrote:
>> Hiroshi Inoue a écrit :
>>> Though I don't understand the problem wholly, the virtual column "index"
>>> doesn't seem appropriate for the identifier. How do you find the row
>>> using the "index" ?
>>
>> In Access, "index" is fine as a field identifier. You just have to
>> reference it between brackets : [index].
>
> What I meant is, for example
>
> 1. Issue the query "select * from my_view".
> 2. You may get some data whose "index" = 1, 2, 3, .....
> 3. Then issue the query "select * from my_view where index = 1"
> 4. The query would return no data.

This is absolutely true, but actually this field is not really meant
to be used as a primary key. I need it because this view is the base
table of a csv export (and I need a unique "index" field in this export).
I chose it as the primary key in access only because I had no other
"correct" field (integer, unique), and I know that I need this kind of
field for a linked table to work correctly in Access.
Now I think that this #DELETED probleme does come from the fact that
the index changes whenever you query the view.
For the moment, I use the view's "real" primary key in Access and it
works OK (it is a 2 columns primary key though, but on two integer
columns).

In response to

Browse pgsql-odbc by date

  From Date Subject
Next Message ioguix 2006-10-12 08:00:24 Re: LATIN1/9 conversion....
Previous Message Hiroshi Inoue 2006-10-12 07:42:34 Re: Access2000 & sequence as primary key in view : #DELETED