Re: Table's Field Info

From: Ludek Finstrle <luf(at)pzkagis(dot)cz>
To: Dmitry Samokhin <sdld(at)mail(dot)ru>
Cc: pgsql-odbc(at)postgresql(dot)org
Subject: Re: Table's Field Info
Date: 2006-01-20 16:46:37
Message-ID: 20060120164637.GB20633@soptik.pzkagis.cz
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-odbc

> > Working in the following environment:
> >
> > Borland VCL ADO Components -> Microsoft OLE DB Provider for ODBC Drivers ->
> > PostgreSQL ODBC Driver -> PostgreSQL Server
>
> We need more :-) What about psqlODBC version? What's datasource setings?

I see no psqlODBC version and no datasource settings in your reply.

> > and executing SQL statements like this:
> >
> > SELECT * FROM mytable WHERE ...
> >
> > I was surprised the driver issues the statement "SELECT * FROM mytable"
> > just before executing my custom statement. I guess it tries to retreive
> > the field info for "mytable". But what if "mytable" contains mios of
> > rows?! The
>
> I'm not familiar with this part of code. And I don't have enough time
> until end of week.

I take a look and the problem is quite beside the point. Not psqlODBC
driver issues statement ;-) The psqlODBC get SQLPrepare from your
test application. So I can't modify this behaviour.
But ...
The statement "SELECT * FROM mytable" is issued becouse the driver returns
empty table name for field (I don't know why BC++ have to get table name)
but it could be changed when you turn on option Parse Statement.
It could you help Use declare/fetch option too.
There is bug in Parse statement until 08.01.0104 (maybe later).
So please try 08.01.0107 development snapshot from pgfoundry.org.

> > execution will continue forever... Yes, I am aware of the option "Disallow
> > Premature". But my Borland TADOQuery object crashes on

Where is the option? Do you mean something in psqlodbc or in BC++?

> > begin;declare cursor .. for select ...;fetch backward in ..;close ..;commit
> > (unexpected EOF of client connection)
>
> How can we fast reproduce your problem? Could you post mylog output?
> It's ideal to post one mylog with "SELECT * FROM mytable" and second
> one with crash. Or you could post me example source or executable with
> data specification.

How can I reproduce this failure with your test app?

Regards,

Luf

In response to

Responses

Browse pgsql-odbc by date

  From Date Subject
Next Message Ludek Finstrle 2006-01-20 16:54:58 Re: PsqlODBC slow on UNION queries
Previous Message Ludek Finstrle 2006-01-20 16:30:27 Re: Updates on updates