Обсуждение: problems with timestamp data type

Hi.

  I have been using PostgreSQL v 7.1.3 without
problems for a year.
I installed the latest version (7.2.1) a week ago and
I have had some problems with a particular query:

///***** QUERY *****///

SELECT date_part('year', fechaacta) AS "A�o de
donaci�n", sexo AS "Sexo", count(*) AS "Cantidad de
donantes"
FROM "Capital"
WHERE
date_part('year', age(timestamp(fechanacimiento)))
BETWEEN 1 AND 100
AND "fecha" BETWEEN '2001/01/01' AND '2002/12/31'
GROUP BY date_part('year', fechaacta), sexo

///***** end QUERY ***////

I obtain the following answer from server:

"Warning: PostgreSQL query failed: ERROR: parser:
parse error at or near "fechanacimiento" in
/var/www/html/ActasPcias/consultas.php on line 471"

Note: "fechanacimiento" is a field of the table
"Capital".


Exactly the same query has been working fine for a
year with PostgreSQL v 7.1.3 in exactly the same
database. more over, the database was created in my
new red hat 7.3 server by using "pg_dumpall" from my
old server (red hat 7.2).


Can anybody tell me what's going on ?

Thanks

Juan Pablo


__________________________________________________
Do You Yahoo!?
Yahoo! Autos - Get free new car price quotes
http://autos.yahoo.com

On Wed, 17 Jul 2002, Juan Pablo wrote:

> SELECT date_part('year', fechaacta) AS "A�o de
> donaci�n", sexo AS "Sexo", count(*) AS "Cantidad de
> donantes"
> FROM "Capital"
> WHERE
> date_part('year', age(timestamp(fechanacimiento)))
> BETWEEN 1 AND 100
> AND "fecha" BETWEEN '2001/01/01' AND '2002/12/31'
> GROUP BY date_part('year', fechaacta), sexo
...
> Exactly the same query has been working fine for a
> year with PostgreSQL v 7.1.3 in exactly the same
> database. more over, the database was created in my
> new red hat 7.3 server by using "pg_dumpall" from my
> old server (red hat 7.2).
>
>
> Can anybody tell me what's going on ?

Well, to fix it, you'll probably need to double quote
timestamp. Timestamp(...) used to be a call to the
type conversion function but is now the specification
of a timestamp type with a given precision (which I
think matches spec).  You might want to consider seeing
if you can do the type conversion using the standard
SQL cast in any case.