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

NEW ODBC DRIVER

From: Byron Nikolaidis <byronn(at)insightdist(dot)com>
To: "pgsql-interfaces(at)postgreSQL(dot)org" <pgsql-interfaces(at)postgreSQL(dot)org>
Cc: Benoit Foucher <bfoucher(at)pratique(dot)fr>, Stephen Davies <scldad(at)sdc(dot)com(dot)au>, "Jose' Soares Da Silva" <sferac(at)bo(dot)nettuno(dot)it>, Hannu Krosing <hannu(at)trust(dot)ee>, Tom Ivar Helbekkmo <tih+mail(at)Hamartun(dot)Priv(dot)NO>, Sbragion Denis <infotecn(at)tin(dot)it>, Dario Fumagalli <dfumagalli(at)art-media(dot)it>, Olaf Mittelstaedt <mstaedt(at)va-sigi(dot)va(dot)fh-ulm(dot)de>, Paul Lisewski <paull(at)techone(dot)com(dot)au>, Christopher Luk <wluk(at)hkcccc(dot)org>, Aleksey Demakov <avd(at)avd(dot)garsib(dot)ru>, Tony Cowderoy <tony(at)nthfen(dot)demon(dot)co(dot)uk>
Subject: NEW ODBC DRIVER
Date: 1998-05-08 16:35:19
Message-ID: 35533446.92762625@insightdist.com (view raw or flat)
Thread:
Lists: pgsql-hackerspgsql-interfaces
Hello all,

There is a new odbc driver (version 6.30.0237) and source code at our
website (http://www.insightdist.com/psqlodbc).  This one contains some
bug fixes AND alot of setup options!  You can now configure the driver
more closely to match your needs hopefully.  Please click on the version
link and/or the "dialog boxes" link for more detailed information.  Some
quick highlights are:

- By democratic vote, uses ISO datestyle, period!  Automatically sets
this datestyle to backend on connection.
- Advanced options for Driver and Datasource (and lots of them).
- Ability to control whether cursors are used, thus emulating the old
driver/libpq behavior.
- Ability to control how unknown sizes are reported: "Longest" emulates
old driver/libpq behavior.
- Recognizing Unique indexes is a driver option.
- Some data type mappings and data type sizes are configurable.
- SQLExtendedFetch implemented when not using cursors.

There is also a "Defaults" button to set all these options back to
optimum settings.

Note, there are performance issues involved with not using cursors,
since the driver must suck down all the rows in the result set.
However, there are some advantages when it comes to updating tables
since the tables are not kept locked by the backend.  If you are having
problems with locked tables or the driver hanging, try setting Use
Cursors to false.  This is a workaround until whenever the backend
locking improves.  Also, when not using cursors, the sizes of character
data types varchar and text can be known since all the tuples are
retrieved in the result set.  So if you set the option "Longest" on the
Advanced Options (Driver),  this will be possible.


Feedback on this new driver is appreciated.

Regards,

Byron.


Responses

pgsql-hackers by date

Next:From: Jackson, DeJuanDate: 1998-05-08 16:50:23
Subject: RE: [QUESTIONS] Important questions about PostgreSQL
Previous:From: Jackson, DeJuanDate: 1998-05-08 16:24:21
Subject: RE: [QUESTIONS] Wrong include/config.h --> Glibc2.0.7pre1 + linux 2.0.33

pgsql-interfaces by date

Next:From: Chris OsbornDate: 1998-05-08 17:49:36
Subject: Re: [INTERFACES] NEW ODBC DRIVER
Previous:From: Constantin TeodorescuDate: 1998-05-07 17:25:21
Subject: Precompiled libraries for PgAccess, Linux RedHat 5.0 i386 and PostgreSQL 6.3.2

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