Skip site navigation (1) Skip section navigation (2)

Problems with PGSQL ODBC drivers

From: Harry Broomhall <harry(dot)broomhall(at)uk(dot)easynet(dot)net>
To: pgsql-odbc(at)postgresql(dot)org
Subject: Problems with PGSQL ODBC drivers
Date: 2003-01-08 17:33:47
Message-ID: 200301081733.RAA72769@haeb.noc.uk.easynet.net (view raw or flat)
Thread:
Lists: pgsql-odbc
    I posted this to the general newsgroup
'comp.databases.postgresql.questions' a little while ago, and got no takers!

<grin>

    As I am fairly sure this is a ODBC driver issue I am reposting it here,
somewhat ammended, in the hope of enlightenment.

===================================

   I have run up against a problem trying to use the PostgreSQL ODBC
Windows driver, and wondered if anyone can point me in the right
direction.

   Basics:

Server:  PostgreSQL 7.2.1 on i386-unknown-freebsd4.5,
         compiled by GCC 2.95.3

ODBC driver:  PostgreSQL (not beta) 7.02.00.05

Windows:  NT4  - SP 4

Excel 2000

  I have a view on the server from which I wish to import data into
Excel, using MS Query.

  If I type the SQL query direct into MSQuery it works fine, and the
data is fetched.

  If I attempt to use the MSQuery 'wizard' then when I get to 'Finish'
and ask to return the data to Excel then I get an Error box with the
message:  "Error:  No such attribute or function 'oid'".  This message
is echoed in the PostgreSQL logs.

  I enabled the log files in the ODBC driver, and noticed that an
attribute 'oid' was added to the query that I would otherwise have
expected the MSQuery app to send.  AFAIK views don't have oids, so it
would seem to me that the driver is misidentifying the view as a table
with oids?

   Looking further at the logs, and going though the driver source, it
*looks* to me as if the code for the ODBC function SQLSpecialColumns
when called with IdentifierType = SQL_BEST_ROWID may be the culprit.

   It *always* (AFAICT) returns oid for this call.

   Looking through info.c I note the following:

  There is a comment early in the function where it says 'Create the
query to find out if this is a view or not...', and fetches
pg_class.relhasrules.  However in the code for SQLTables is the
comment 'view is represented by its relkind since 7.1'.

   Any suggestions as to work-arounds, patches, or tests to try would
be welcomed.

========================

   Regards,
          Harry.
 


Responses

pgsql-odbc by date

Next:From: Patrick HatcherDate: 2003-01-08 18:36:20
Subject: Unable to download from new site
Previous:From: Tambet MatiisenDate: 2003-01-08 10:42:11
Subject: ADO resync and views

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group