Resent-From: POSTGRES mailing list <postman@postgres.Berkeley.EDU>
Resent-Message-Id: <199402152306.PAA23883@nobozo.CS.Berkeley.EDU>
Sender: owner-postman@postgres.Berkeley.EDU
Date: Tue, 15 Feb 1994 11:53:52 -0800
From: Bruce Lowe <lowe@zso.dec.com>
Message-Id: <9402151953.AA29417@slugbt.zso.dec.com>
To: postgres@postgres.Berkeley.EDU
Subject: Request for info update
Resent-To: postgres-dist@postgres.Berkeley.EDU
Resent-Date: Tue, 15 Feb 94 15:06:08 -0800
Resent-XMts: smtp

About a year ago, I tried Postgres (Ultrix) and liked it, but I had
certain reservations and questions. I posted the following note and
received a reply, the text is included below.

Have there been any more recent developments regarding my questions in
particular, and regarding Postgres in general?

-Bruce Lowe

-----------------------------------------------------------------------------

Return-Path: lowe
To: postgres@postgres.berkeley.edu
Cc: lowe@rust
Subject: some questions from a new user
Date: Fri, 15 Jan 93 10:42:40 -0800
From: lowe
X-Mts: smtp

I am a software developer with DEC, with experience in commercial Ingres 
application development (I grew up on version 5 using QUEL). I have recently
downloaded and installed Postgres and have been doing a little experimenting
with it.

I have a few questions for anyone who might know and answer:

- The biggest obstacle I have found for getting good use from Postgres is the
  apparent lack of a facility for retrieving status/error information. There
  is a reference in the file "c.h" about future plans for a global variable
  to contain the status of the last operation (retrieving from the ERROR 
  relation is mentioned, but I can find no such relation). My question is
  this: is there a way that the status/error state for the last operation can
  be manually retrieved from some system relation (ala INQUIRE_INGRES)? Could
  error text info by subsequently looked up?

- Has anyone written an embedded quel preprocessor ?

- Has anyone done any porting work to OSF ? Specifically OSF running on a DEC
  Alpha box ? Would there be any significant difficulties due to 32/64 bit
  issues ?

- How about NT ? I gather that the port to NT would by of much
  greater difficulty since the sockets and shared memory mechanisms are so 
  different.

I will look forward to any replies. THank you.

-Bruce Lowe

-----------------------------------------------------------------------------

Return-Path: aoki@postgres.Berkeley.EDU
Received: by inet-gw-1.pa.dec.com; id AA17140; Fri, 15 Jan 93 13:19:30 -0800
Received: by faerie.CS.Berkeley.EDU (5.57/Ultrix3.0-C) id AA13016; Fri, 15 Jan 93 13:19:48 -0800
From: <aoki@postgres.berkeley.edu>
Message-Id: <9301152119.AA13016@faerie.CS.Berkeley.EDU>
To: lowe@rust.zso.dec.com
Cc: postgres@postgres.berkeley.edu
Subject: Re: some questions from a new user 
In-Reply-To: Your message of Fri, 15 Jan 93 10:42:40 -0800 
	     <9301151840.AA07965@postgres.Berkeley.EDU> 
Date: Fri, 15 Jan 93 13:19:48 PST

I don't have a good answer for the error-handling question.  There aren't
any current plans that I know of to change it.

> - Has anyone written an embedded quel preprocessor ?

libpq is the only application interface that I know of.

> - Has anyone done any porting work to OSF ? Specifically OSF running on a DEC
>   Alpha box ? Would there be any significant difficulties due to 32/64 bit
>   issues ?

We are working on an Alpha port.  Yes, there are difficulties with 32/64-bit 
issues (though they appear to be of a "massive pain-in-the-rear" nature rather
than really "significant").  In fact, if you have any pull with the DEC
compiler guys, perhaps you could suggest that they provide a command-line 
switch that makes "int" 8 bytes for the OSF/1 C compiler..

> - How about NT ? I gather that the port to NT would by of much
>   greater difficulty since the sockets and shared memory mechanisms are so 
>   different.

A Berkeley-written NT port is still very much in the discussion/politics 
phase.

Disclaimer: I'm just a post_hacker.  I don't speak for Mike Stonebraker.
--
  Paul M. Aoki  |  CS Div., Dept. of EECS, UCB  |  aoki@postgres.Berkeley.EDU
                |  Berkeley, CA 94720           |  ...!uunet!ucbvax!aoki