Re: [SQL] Sql ORDER BY and ASC/DESC question

2008-01-30 Thread A. Kretschmer
am  Wed, dem 30.01.2008, um 11:35:51 +0100 mailte Jaroslav Sivy folgendes:
> Hello everyone,
> 
> I have following problem: am using pl/sql functions to trigger some
> sql code and i need to pass ORDER_BY column name and ASC/DESC sorting
> order as an input parameters into that function and order the result
> based on these input parameters.
> 
> The problem is, that the only way is to create query as "string" and
> then execute it as "string".

Right.


> 
> Is there any other way around how to avoid that "string query"?

No.


> If it's not possible, could anyone implement this feature into one of
> future versions? I think that it would be quite handy to have

Unlikely...


Andreas
-- 
Andreas Kretschmer
Kontakt:  Heynitz: 035242/47150,   D1: 0160/7141639 (mehr: -> Header)
GnuPG-ID:   0x3FFF606C, privat 0x7F4584DA   http://wwwkeys.de.pgp.net

---(end of broadcast)---
TIP 9: In versions below 8.0, the planner will ignore your desire to
   choose an index scan if your joining column's datatypes do not
   match


[SQL] Sql ORDER BY and ASC/DESC question

2008-01-30 Thread Jaroslav Sivy
Hello everyone,

I have following problem: am using pl/sql functions to trigger some sql code 
and i need to pass ORDER_BY column name and ASC/DESC sorting order as an input 
parameters into that function
and order the result based on these input parameters.

The problem is, that the only way is to create query as "string" and then 
execute it as "string".

Is there any other way around how to avoid that "string query"?

(i don't want to use IFs either, because 5 columns x 2 sorting orders would 
require 10 IFfed queries with different column/sorting order combination)


If it's not possible, could anyone implement this feature into one of future 
versions? I think that it would be quite handy to have something like that.

Thank you for your responses.
---
Menu.sk - Denne aktualizovany zoznam obedovych menu a restauracii s presnou 
lokalizaciou. 

  Put the text of the welcome message here.


---(end of broadcast)---
TIP 2: Don't 'kill -9' the postmaster


Re: [SQL] Sql ORDER BY and ASC/DESC question

2008-01-30 Thread Bart Degryse
Actually there might be assuming your function is a set returning function.
This example eg works perfectly and sorts the output of the function without 
having to use execute.
 
CREATE TABLE "public"."error_types" (
  "id" SERIAL, 
  "errdesc" TEXT NOT NULL, 
  "autofix" BOOLEAN DEFAULT false NOT NULL, 
  CONSTRAINT "error_types_errdesc_key" UNIQUE("errdesc"), 
  CONSTRAINT "error_types_pkey" PRIMARY KEY("id")
) WITH (fillfactor = 100, OIDS = FALSE);
 
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Missing 
InvoiceID', True);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Missing 
InventTransID', True);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Illegal 
oneshot dates (start <> end)', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Illegal 
dates (start > end)', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES 
('Accountnumber not defined', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Creditnote 
with a positive amount', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Invoice with 
a negative amount', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Original 
invoice not found', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Update 
reknr', True);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Attempt to 
change reknr', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Update 
transactiondate', True);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Attempt to 
change transactiondate', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Update 
invoiceid', True);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Attempt to 
change invoiceid', False);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Update 
lineamountmst', True);
INSERT INTO "public"."error_types" ("errdesc", "autofix") VALUES ('Attempt to 
change lineamountmst', False);
 
CREATE TYPE test3_type AS (
  myid integer,
  myerrdesc text
);
 
CREATE OR REPLACE FUNCTION test3 () RETURNS SETOF test3_type AS
$body$
DECLARE
  rec test3_type;
BEGIN
  FOR rec IN (
SELECT id, errdesc
FROM error_types)
  LOOP
RETURN NEXT rec;
  END LOOP;
  RETURN;
END;
$body$
LANGUAGE 'plpgsql' VOLATILE CALLED ON NULL INPUT SECURITY INVOKER;
 
select * from test3() order by myerrdesc desc;

myidmyerrdesc
11Update transactiondate
9Update reknr
15Update lineamountmst
13Update invoiceid
8Original invoice not found
1Missing InvoiceID
2Missing InventTransID
7Invoice with a negative amount
3Illegal oneshot dates (start <> end)
4Illegal dates (start > end)
6Creditnote with a positive amount
12Attempt to change transactiondate
10Attempt to change reknr
16Attempt to change lineamountmst
14Attempt to change invoiceid
5Accountnumber not defined
 
I hope this is useful to you.

>>> "A. Kretschmer" <[EMAIL PROTECTED]> 2008-01-30 11:42 >>>
am  Wed, dem 30.01.2008, um 11:35:51 +0100 mailte Jaroslav Sivy folgendes:
> Hello everyone,
> 
> I have following problem: am using pl/sql functions to trigger some
> sql code and i need to pass ORDER_BY column name and ASC/DESC sorting
> order as an input parameters into that function and order the result
> based on these input parameters.
> 
> The problem is, that the only way is to create query as "string" and
> then execute it as "string".

Right.


> 
> Is there any other way around how to avoid that "string query"?

No.


> If it's not possible, could anyone implement this feature into one of
> future versions? I think that it would be quite handy to have

Unlikely...


Andreas
-- 
Andreas Kretschmer
Kontakt:  Heynitz: 035242/47150,   D1: 0160/7141639 (mehr: -> Header)
GnuPG-ID:   0x3FFF606C, privat 0x7F4584DA   http://wwwkeys.de.pgp.net ( 
http://wwwkeys.de.pgp.net/ )

---(end of broadcast)---
TIP 9: In versions below 8.0, the planner will ignore your desire to
   choose an index scan if your joining column's datatypes do not
   match


Re: [SQL] Sql ORDER BY and ASC/DESC question

2008-01-30 Thread Gregory Stark
"A. Kretschmer" <[EMAIL PROTECTED]> writes:

> am  Wed, dem 30.01.2008, um 11:35:51 +0100 mailte Jaroslav Sivy folgendes:
>> Hello everyone,
>> 
>> I have following problem: am using pl/sql functions to trigger some
>> sql code and i need to pass ORDER_BY column name and ASC/DESC sorting
>> order as an input parameters into that function and order the result
>> based on these input parameters.
>> 
>> The problem is, that the only way is to create query as "string" and
>> then execute it as "string".
>
> Right.
>> 
>> Is there any other way around how to avoid that "string query"?

If you're not concerned with the planner being able to find indexes to satisfy
these orderings (ie, you don't mind always doing a sort) you could do
something like:

ORDER BY 
 CASE ? 
 WHEN 1 THEN name ASC
 WHEN 2 THEN name DESC
 WHEN 3 THEN height ASC
 WHEN 4 THEN height DESC
 ELSE id ASC
 END

But to the planner this will be basically an opaque expression. It won't be
able to use any indexes on these columns. Also, incidentally you might want to
use text strings instead of integer labels.

-- 
  Gregory Stark
  EnterpriseDB  http://www.enterprisedb.com
  Get trained by Bruce Momjian - ask me about EnterpriseDB's PostgreSQL 
training!

---(end of broadcast)---
TIP 9: In versions below 8.0, the planner will ignore your desire to
   choose an index scan if your joining column's datatypes do not
   match


Re: [SQL] Sql ORDER BY and ASC/DESC question

2008-01-30 Thread Tom Lane
Gregory Stark <[EMAIL PROTECTED]> writes:
> If you're not concerned with the planner being able to find indexes to satisfy
> these orderings (ie, you don't mind always doing a sort) you could do
> something like:

> ORDER BY 
>  CASE ? 
>  WHEN 1 THEN name ASC
>  WHEN 2 THEN name DESC
>  WHEN 3 THEN height ASC
>  WHEN 4 THEN height DESC
>  ELSE id ASC
>  END

Uh, no, putting the ASC/DESC decoration inside a CASE like that is not
gonna work --- it's only allowed at the top level of an ORDER BY clause.
For numerical sort keys you can cheat by using "-x" in place of
"x DESC", but I'm not aware of any equivalent hack for text keys.

regards, tom lane

---(end of broadcast)---
TIP 5: don't forget to increase your free space map settings


Re: [SQL] Sql ORDER BY and ASC/DESC question

2008-01-30 Thread Gregory Stark
"Tom Lane" <[EMAIL PROTECTED]> writes:

> Gregory Stark <[EMAIL PROTECTED]> writes:
>
>> ORDER BY 
>>  CASE ? 
>>  WHEN 1 THEN name ASC
>
> Uh, no, putting the ASC/DESC decoration inside a CASE like that is not
> gonna work

doh! I had a feeling something was wrong but couldn't put my finger on it
before I hit send. Sigh.

> For numerical sort keys you can cheat by using "-x" in place of
> "x DESC", but I'm not aware of any equivalent hack for text keys.

Yeah, you could do a really kludgy thing with a second sort expression where
you null out one expression or the other depending on the parameter but it
starts to look more and more spaghetti-like.

-- 
  Gregory Stark
  EnterpriseDB  http://www.enterprisedb.com
  Get trained by Bruce Momjian - ask me about EnterpriseDB's PostgreSQL 
training!

---(end of broadcast)---
TIP 6: explain analyze is your friend


[SQL] Problem with the to_timestamp function

2008-01-30 Thread Luca Clementi




So the start_time is a varchar column, which contains a date.


opal_app=# select job_id,start_time from job_status where 
job_id='app1201551799779' ;
 job_id  |start_time
--+--

app1201551799779 | 1 28, 2008 12:23:19 午後
(1 row)

opal_app=# select job_id,to_timestamp(start_time, 'M DD,   HH12:MI:SS 午後') 
from job_status where job_id='app1201551799779';
 job_id  |  to_timestamp  
--+

app1201551799779 | 2008-01-28 02:23:19-08
(1 row)

opal_app=# select job_id,to_timestamp(start_time, 'M DD,   HH12:MI:SS') 
from job_status where job_id='app1201551799779';
 job_id  |  to_timestamp  
--+

app1201551799779 | 2008-01-28 02:23:19-08
(1 row)

opal_app=# select job_id,to_timestamp(start_time, 'M DD,   HH:MI:SS') from 
job_status where job_id='app1201551799779';
 job_id  |  to_timestamp  
--+

app1201551799779 | 2008-01-28 02:23:19-08
(1 row)


It seems that the to_timestamp does not work properly in this case, 
when it comes to parsing the hours. I verified that this problem 
happen no matter what the input hours is and the result is always 
00:min:sec or 02:min:sec.


opal_app=# select job_id,start_time from job_status where 
job_id='app1201563668439';
 job_id  |   start_time
--+-

app1201563668439 | 1 28, 2008 3:41:08 午後
(1 row)

opal_app=# select job_id, to_timestamp(start_time, 'M DD,   HH24:MI:SS') 
from job_status where job_id='app1201563668439';
 job_id  |  to_timestamp  
--+

app1201563668439 | 2008-01-28 00:01:08-08
(1 row)

Or
opal_app=# select job_id,start_time from job_status where 
job_id='app1201565220760';
 job_id  |   start_time
--+-

app1201565220760 | 1 28, 2008 4:07:00 午後
(1 row)

opal_app=# select job_id,to_timestamp(start_time, 'M DD,   HH:MI:SS') from 
job_status where job_id='app1201565220760';
 job_id  |  to_timestamp  
--+

app1201565220760 | 2008-01-28 00:07:00-08
(1 row)



Is this a bug or am I doing something wrong?


Thank you for any help,
Luca



---(end of broadcast)---
TIP 4: Have you searched our list archives?

  http://archives.postgresql.org


Re: [SQL] Problem with the to_timestamp function

2008-01-30 Thread Tom Lane
Luca Clementi <[EMAIL PROTECTED]> writes:
> It seems that the to_timestamp does not work properly in this case, 
> when it comes to parsing the hours.

to_timestamp() is not very robust if the input doesn't exactly match
what it expects for the format string.  I'm not sure if that's the
issue here, but have you tried just casting the string to timestamp?
That would use the standard timestamp input converter, which is
pretty flexible.

(BTW, there is definitely 0 hope of recognizing a timezone name that's
written in Chinese characters, unless maybe you fool around with the
timezone-abbreviations configuration file.)

regards, tom lane

---(end of broadcast)---
TIP 3: Have you checked our extensive FAQ?

   http://www.postgresql.org/docs/faq


Re: [SQL] Problem with the to_timestamp function

2008-01-30 Thread Michael Glaesemann


On Jan 30, 2008, at 19:26 , Tom Lane wrote:


(BTW, there is definitely 0 hope of recognizing a timezone name that's
written in Chinese characters, unless maybe you fool around with the
timezone-abbreviations configuration file.)


Just FYI, the characters in the original email are the equivalent of  
PM, rather than designating a time zone. Not that that affects your  
advice, of course.


Michael Glaesemann
grzm seespotcode net



---(end of broadcast)---
TIP 4: Have you searched our list archives?

  http://archives.postgresql.org


Re: [SQL] Problem with the to_timestamp function

2008-01-30 Thread Luca Clementi

Tom Lane wrote:

Luca Clementi <[EMAIL PROTECTED]> writes:
It seems that the to_timestamp does not work properly in this case, 
when it comes to parsing the hours.


to_timestamp() is not very robust if the input doesn't exactly match
what it expects for the format string.  I'm not sure if that's the
issue here, but have you tried just casting the string to timestamp?
That would use the standard timestamp input converter, which is
pretty flexible.

(BTW, there is definitely 0 hope of recognizing a timezone name that's
written in Chinese characters, unless maybe you fool around with the
timezone-abbreviations configuration file.)



I found the problem!!

to_timestamp(start_time, 'M DD,   HH12:MI:SS')
  ^^
in the formatting string I have two spaces, while in the original there 
is only one:

1 28, 2008 12:23:19 午後

So if I use: 'M DD,  HH12:MI:SS' as a formatting string it works!
I wander how come if the formatting string doesn't match properly you 
get such a bad behavior, would it better just a error?



PS: some e-mail clients do not render properly the spacing in my first 
e-mail.


Thanks again for your help!!
Luca


---(end of broadcast)---
TIP 9: In versions below 8.0, the planner will ignore your desire to
  choose an index scan if your joining column's datatypes do not
  match


Re: [SQL] Problem with the to_timestamp function

2008-01-30 Thread Tom Lane
Luca Clementi <[EMAIL PROTECTED]> writes:
> I found the problem!!
> to_timestamp(start_time, 'M DD,   HH12:MI:SS')
> in the formatting string I have two spaces, while in the original there 
> is only one:
> 1 28, 2008 12:23:19 午後

> So if I use: 'M DD,  HH12:MI:SS' as a formatting string it works!
> I wander how come if the formatting string doesn't match properly you 
> get such a bad behavior, would it better just a error?

Yeah, it should either do something reasonable or throw an error.
Improving to_timestamp's handling of such cases is on the TODO list.
In the meantime, though, the standard input converter is a whole lot
more flexible for slightly-out-of-spec input ... so I ask again,
do you really need to_timestamp at all?

regards, tom lane

---(end of broadcast)---
TIP 6: explain analyze is your friend