pg.ddx.io  pgsql-sql@postgresql.org mailing list archive  
help / color / mirror / Atom feed
Issues w/ 7.1 to 7.2 upgrade.
3+ messages / 3 participants
[nested] [flat]

* Issues w/ 7.1 to 7.2 upgrade.
@ 2002-02-19 20:59 Diehl, Jeffrey <jdiehl@sandia.gov>
  2002-02-20 03:38 ` Re: Issues w/ 7.1 to 7.2 upgrade. Josh Berkus <josh@agliodbs.com>
  2002-02-20 13:11 ` Re: Issues w/ 7.1 to 7.2 upgrade. Thomas Good <tomg@admin.nrnet.org>
  0 siblings, 2 replies; 3+ messages in thread

From: Diehl, Jeffrey @ 2002-02-19 20:59 UTC (permalink / raw)
  To: pgsql-sql

Hi all.

I just upgraded from 7.1 to 7.2, mainly because of the facts that vacuum
doesn't do table locks and I thought that "replace into" would be in 7.2.

Anyway, I don't see that "replace into" is there...

Also, I noticed a slight difference that caught me off guard:

If I try to insert '1234567890' into a varchar(5) under 7.1, the value of
the field will be '12345'.

If I try it under 7.2, I get an error.

Needless to say, I learned a bit more about my dataset this morning...
<grin>

Mike Diehl,
Network Monitoring Tool Devl.
Sandia National Laboratories.
(505) 284-3137
jdiehl@sandia.gov




^ permalink  raw  reply  [nested|flat] 3+ messages in thread

* Re: Issues w/ 7.1 to 7.2 upgrade.
  2002-02-19 20:59 Issues w/ 7.1 to 7.2 upgrade. Diehl, Jeffrey <jdiehl@sandia.gov>
@ 2002-02-20 03:38 ` Josh Berkus <josh@agliodbs.com>
  1 sibling, 0 replies; 3+ messages in thread

From: Josh Berkus @ 2002-02-20 03:38 UTC (permalink / raw)
  To: Diehl, Jeffrey <jdiehl@sandia.gov>; pgsql-sql

Mike,

> I just upgraded from 7.1 to 7.2, mainly because of the facts that
>  vacuum
> doesn't do table locks and I thought that "replace into" would be in
>  7.2.
> 
> Anyway, I don't see that "replace into" is there...

Nope.  I hadn't had the impression from the list traffic that the
 developers ever considered letting in this bit of non-standard SQL.
  While it's undoubtedly useful, it's not in the SQL 92 spec and a
 *lot* of extra programming would be required to support it (for
 example, those of us who use UPDATE and INSERT triggers might find
 ourselves re-writing them all).

Remember:  If you really, really want "replace into", Postgres *is* an
 open-source database.  You are always welcome to hack your own
 "replace into" patch.

> Also, I noticed a slight difference that caught me off guard:
> 
> If I try to insert '1234567890' into a varchar(5) under 7.1, the
>  value of
> the field will be '12345'.
> 
> If I try it under 7.2, I get an error.

Another annoying change per the spec, I believe.  Personally, this
 change made me happy because unexpected truncations had previously
 escaped my notice.  But it did lead to a spate of emergency bug fixes
 in my code.

-Josh

______AGLIO DATABASE SOLUTIONS___________________________
                                       Josh Berkus
  Complete information technology      josh@agliodbs.com
   and data management solutions       (415) 565-7293
  for law firms, small businesses        fax 621-2533
    and non-profit organizations.      San Francisco



^ permalink  raw  reply  [nested|flat] 3+ messages in thread

* Re: Issues w/ 7.1 to 7.2 upgrade.
  2002-02-19 20:59 Issues w/ 7.1 to 7.2 upgrade. Diehl, Jeffrey <jdiehl@sandia.gov>
@ 2002-02-20 13:11 ` Thomas Good <tomg@admin.nrnet.org>
  1 sibling, 0 replies; 3+ messages in thread

From: Thomas Good @ 2002-02-20 13:11 UTC (permalink / raw)
  To: Diehl, Jeffrey <jdiehl@sandia.gov>; +Cc: pgsql-sql

On Tue, 19 Feb 2002, Diehl, Jeffrey wrote:

> Hi all.
> 
> I just upgraded from 7.1 to 7.2, mainly because of the facts that vacuum
> doesn't do table locks and I thought that "replace into" would be in 7.2.
> 
> Anyway, I don't see that "replace into" is there...

Hi Mike,

I'm anxiously awaiting DROP COLUMN (rather than using the workaround
of SELECTing desired attributes into a new relation, dropping the old
rel and renaming the new...)

I was hoping it would be in 7.2 but not yet.  :-(

> Also, I noticed a slight difference that caught me off guard:
> 
> If I try to insert '1234567890' into a varchar(5) under 7.1, the value of
> the field will be '12345'.
> 
> If I try it under 7.2, I get an error.

I found this in using 7.2 beta - it mimics Oracle's response to a string
that exceeds the len of the attribute.  I kinda liked the unstated substr()
behaviour.  But I guess it was a bit sloppy.  

------------------------------------------------------------------------
Thomas Good                                         tomg@admin.nrnet.org
Programmer/Analyst                                  thomas@sqlclinic.net
Residential Services, Behavioral Health Services
Saint Vincent Catholic Medical Centers                                      
--
SQL Clinic - An Open Source Clinical Record            www.sqlclinic.net
------------------------------------------------------------------------
 






^ permalink  raw  reply  [nested|flat] 3+ messages in thread


end of thread, other threads:[~2002-02-20 13:11 UTC | newest]

Thread overview: 3+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2002-02-19 20:59 Issues w/ 7.1 to 7.2 upgrade. Diehl, Jeffrey <jdiehl@sandia.gov>
2002-02-20 03:38 ` Josh Berkus <josh@agliodbs.com>
2002-02-20 13:11 ` Thomas Good <tomg@admin.nrnet.org>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox