Re: [HACKERS] (Fwd) Re: Any Oracle 9 users? A test please...

2002-10-04 Thread Roland Roberts

 Mike == Mike Mascari [EMAIL PROTECTED] writes:

Mike Tom Lane wrote:
 Yury Bokhoncovich [EMAIL PROTECTED] writes:

 As reported by my friend: Oracle 8.1.7 (ver.9 behaves the same way):

 [ to_char(sysdate) advances in a transaction ]

 Now I'm really confused; this directly contradicts the report
 of Oracle 8's behavior that we had earlier from Roland Roberts.
 Can someone explain why the different results?

Mike Roland used an anonymous PL/SQL procedure:

You're right and I didn't think enough about what was happening.  This
also explains why I so often see the same timestamp throughout a
transaction---the transaction is all taking place inside a PL/SQL
procedure.

roland
-- 
   PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD RL Enterprises
[EMAIL PROTECTED] 76-15 113th Street, Apt 3B
[EMAIL PROTECTED]   Forest Hills, NY 11375

---(end of broadcast)---
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to [EMAIL PROTECTED] so that your
message can get through to the mailing list cleanly



Re: [HACKERS] Re: New Linux xfs/reiser file systems

2001-05-04 Thread Roland Roberts

 Bruce == Bruce Momjian [EMAIL PROTECTED] writes:

 Well, arguably if you're setting up a database server then a
 reasonable DBA should think about such things...

Bruce Yes, but people have trouble installing PostgreSQL.  I
Bruce can't imagine walking them through a newfs.

In most of linux-land, the DBA is probably also the sysadmin.  In
bigger shops, and those which currently run, say Oracle or Sybase, the
two roles are separate.  When they are separate, you don't have to
walk the DBA through it; he just walks over to the sysadmin and says
I need X megabytes of space on a new Y filesystem.

roland
-- 
   PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD RL Enterprises
[EMAIL PROTECTED] 76-15 113th Street, Apt 3B
[EMAIL PROTECTED]  Forest Hills, NY 11375

---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster



Re: [HACKERS] Is there no DESCRIBE TABLE; on PGSQL? help!!!

2001-10-23 Thread Roland Roberts

 Ron == Ron de Jong [EMAIL PROTECTED] writes:

Ron Any idea to get a human readable list with column
Ron descriptions like type,size,key,default,null.

Ron It would be nice if it would look simular to the mysql
Ron variant:

You'll need to write your own query to get it to look like mysql.
From psql, you can do

\d employee

roland
-- 
   PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD RL Enterprises
[EMAIL PROTECTED] 76-15 113th Street, Apt 3B
[EMAIL PROTECTED]   Forest Hills, NY 11375

---(end of broadcast)---
TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]



Re: [HACKERS] Re: Week number

2001-03-14 Thread Roland Roberts

 "Peter" == Peter Eisentraut [EMAIL PROTECTED] writes:

Peter The POSIX numbering (0-6) is actually pretty slick because
Peter it allows both versions to work: In the U.S. (e.g.) you get
Peter a natural order starting at 0, in Germany (e.g.) you get
Peter Monday as #1.

Oracle's to_char() supports format IW for the ISO week of the year,
but there is no equivalent ID for the ISO day of the week.  Perhaps
this should be a PostgreSQL extension?

roland
-- 
   PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD RL Enterprises
[EMAIL PROTECTED] 76-15 113th Street, Apt 3B
[EMAIL PROTECTED]  Forest Hills, NY 11375

---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])



Re: AW: [HACKERS] Re: Week number

2001-03-14 Thread Roland Roberts

 "AZ" == Zeugswetter Andreas SB [EMAIL PROTECTED] writes:

 Unix day-of-week starts on Sunday, not Monday, which is what
 date_trunc('dow',...) returns. Presumably this is modeled on
 the traditional notion (at least in the US; I suspect this is
 true in most European countries at least) of Sunday being "the
 first day of week".

AZ Germany and Austria have Monday as first day of week, I think
AZ most of Europe also.

I believe the goal was to have a to_char() that was complete and
Oracle-compatible.  Perhaps we need to also have a trunc() which is
Oracle compatible.  I haven't been playing with 7.1beta, but 7.0
trunc() doesn't like timestamps.  In Oracle, I can say

select trunc(sysdate) - trunc(sysdate,'ww') + 1 from dual;

to get Monday=1.

roland
-- 
   PGP Key ID: 66 BC 3B CD
Roland B. Roberts, PhD RL Enterprises
[EMAIL PROTECTED] 76-15 113th Street, Apt 3B
[EMAIL PROTECTED]  Forest Hills, NY 11375

---(end of broadcast)---
TIP 6: Have you searched our list archives?

http://www.postgresql.org/search.mpl