I am running PostgreSQL 8.1.5 under Windows Server 2003 Standard
edition. All has been running for weeks now, don't know if it has been
restarted since we installed. Today we installed a program, Paradox
database, which required a restart, and now the PostgreSQL Server
service won't start with the
On Tue, 2007-01-16 at 20:00 -0600, Adam Rich wrote:
> Robert,
> Open "Computer Management" and find the postgres service.
> There should be an item saying "Run As". You want that to be
> a non-privledged account. Typically, postgres asks you what account
> to use when you install it. Just make
On Wed, 2007-01-17 at 14:13 +, Dave Page wrote:
> Raymond O'Donnell wrote:
> > Robert Fitzpatrick wrote:
> >
> >> Thanks, yes, I verified the services has 'postgres' as the account being
> >> used under the Log On tab of the PostgreSQL service.
I have a customer who is wants to migrate his MSSQL database to
PostgreSQL and we'll replace his application ASP with PHP. The issues
should be limited as there are no stored procedures or triggers in
MSSQL, just structure and data should be all that is needed to migrate.
I have never migrated from
On Mon, 2007-01-22 at 16:32 +, Peter Rosenthal wrote:
> Wanting to do something similar I recently submitted a large patch to
> the mysql2pgsql project. It will now handle conversion of a mysqldump
> file complete with data for the quite large and diverse DB I was using
> it with. I'm sure ther
On Mon, 2007-01-22 at 12:13 -0500, Ted Byers wrote:
> Is the original application ASP or SP.NET? It makes a difference,
> particularly if it was developed to take advantage of ASP.NET 2. It might
> conceivably be ASP.NET 3, but since that is brand new I can't see anyone
> paying to replace an
I want to be able to build complex search and reporting capabilities in
to our PHP5 application. We want to be able to save searches for later
use. We also want to build queries from virtually any field in certain
tables from our PHP app. I hope to do as much within postgresql as
possible. Then I s
I am trying to use the crosstab function of the contrib tablefunc.
Reading the README, I believe I am supposed to be using crosstab(sql, N)
for my situation and wondering if the SQL can be based on a view? I have
this view created that gives me each sales rep and their total number of
units sold an
Finally figured out what was wrong with my crosstab that I posted for
help yesterday. I was really close, just need to set the right types. I
have it working using the crosstab(text sql, int N) function. This
displays a crosstab from my view below for units sold by each sales rep
under each month..
I have a query that pulls totals for the month and from there I am
building a crosstab to show all months. My dilemma is that sometimes
there is no data for a month and the crosstab becomes skewed. I made a
table with all the 12 months in it and joined to the query in my view to
get all the months
Haven't done a whole lot of plsql returning records, only those based on
a query. I was wondering, can I build a record from the results of the
query using other values for some fields in the record? I know 'return
next' requires a record and want to build my own record to include some
argument val
I have a function that returns a set of records based on one of my
views. The function takes two arguments of user_id and saved search
name, looks up the search values from a table previously saved and
performs a SELECT query on my view to return my set of records found.
However, I don't want to SE
We have a MS SQL server db that we successfully migrated to pgsql 8.2
and I am now working on some views and notice select queries being very
slow. I have vacuumed last night and running the exact same query (with
minor changes to syntax of course), it runs in just a few seconds
compared to several
On Sat, 2007-05-19 at 17:05 -0400, Tom Lane wrote:
> Robert Fitzpatrick <[EMAIL PROTECTED]> writes:
> > I am running the following query on a linux server with comparable
> > processor and memory as the windows server.
>
> Show us the table definitions and the EXP
On Sat, 2007-05-19 at 19:19 -0400, Tom Lane wrote:
> You're comparing fields of distinct types, which not only incurs
> run-time type conversions but can interfere with the ability to
> use some plan types at all. Looking at the table definitions,
> you've got primary keys declared as SERIAL (ie,
I posted an issue I was having with a migration from MS SQL server to
pgsql over the weekend. Tom's suggestion for the query I posted was
right on, I made the appropriate updates to column types by dumping,
changing and restoring the database. I then analyze'd the db and my
query performance was eq
On Tue, 2007-05-22 at 17:21 +0100, Richard Huxton wrote:
> Robert Fitzpatrick wrote:
> > I have developed a view in pgsql that takes over 160K ms to execute, but
> > when copied into MS SQL against the old database (with syntax mods of
> > course), runs in a few seconds.
>
On Tue, 2007-05-22 at 19:04 +0200, PFC wrote:
> > I thought the same thing. While I'm not a MS SQL guru, I put 'TOP 100
> > PERCENT' after SELECT in the query. The Enterprise Manager does not
> > indicate how many rows come back. I save it as a VIEW in MS SQL and do a
> > 'select count(*)...' and,
On Tue, 2007-05-22 at 13:10 -0400, Tom Lane wrote:
> Robert Fitzpatrick <[EMAIL PROTECTED]> writes:
> > On Tue, 2007-05-22 at 17:21 +0100, Richard Huxton wrote:
> >> Your query seems to produce 41.8 million rows. Are you sure MS-SQL is
> >> returning that man
On Tue, 2007-05-22 at 19:03 +0100, Richard Huxton wrote:
> 4. We're still 5 x slower than MS-SQL (with the count). That might
> well
> be down to having to check visibility on each row with our MVCC
> rather
> than just going to the index.
Tips? I'd love to know how to see inside MVCC. I really
On Tue, 2007-05-22 at 14:30 -0400, Robert Fitzpatrick wrote:
> On Tue, 2007-05-22 at 19:03 +0100, Richard Huxton wrote:
> >
> > Hmm... How much of your machine is PG getting to use vs. MS-SQL? What
> > are your shared_buffers, work_mem, effective_cache_size (and how much
On Tue, 2007-05-22 at 20:13 +0100, Richard Huxton wrote:
> Robert Fitzpatrick wrote:
> > I did some googling and came up with some ideas, I have it now with
> > these settings and after restarting PG, no help.
> >
> > work_mem = 5MB
>
> My last post missed yo
Thanks to Richard for the help earlier this week with performance
questions, once I was able to get pgsql and mssql using the same
resources and doing some tuning, I was able to get comparable results.
The issue still though, I have this view that I designed with the
thought in mind to provide all
On Wed, 2007-05-23 at 18:53 -0400, Ericson Smith wrote:
> Yeah, we've used Tsearch with joins and searches on other fields on
> the tsearch table no problem. Tsearches are just another part of a
> WHERE clause.
>
And can there be Tsearch fields built based on fields in different
tables? Where ca
On Wed, 2007-05-23 at 19:48 -0400, Robert Fitzpatrick wrote:
> On Wed, 2007-05-23 at 18:53 -0400, Ericson Smith wrote:
> > Yeah, we've used Tsearch with joins and searches on other fields on
> > the tsearch table no problem. Tsearches are just another part of a
> > WHERE
How can I reference any NEW field in an insert/update trigger function?
When someone inserts or updates any field with a single asterisk (*), I
need it to become '%%%'. But if they use an asterisk in any combination
with other fields, then I want to TRANSLATE those asterisks to a single
'%'.
I was
On Fri, 2007-05-25 at 11:17 -0400, Robert Fitzpatrick wrote:
> But if they use an asterisk in any combination
> with other fields
I meant to say 'But if they use an asterisk in any combination with
other *values* in the field...'. For instance, if they enter '*test*',
On Fri, 2007-05-25 at 12:00 -0400, Alvaro Herrera wrote:
> Robert Fitzpatrick wrote:
> > How can I reference any NEW field in an insert/update trigger function?
> > When someone inserts or updates any field with a single asterisk (*), I
> > need it to become '%%%'.
On Fri, 2007-05-25 at 13:45 -0400, Alvaro Herrera wrote:
> Robert Fitzpatrick wrote:
>
> > But still, how would I reference all fields using the pl/perl? Can I
> > specify column numbers versus names as in '$_TD->{new}{1}' for the first
> > column and loop or
I have a dump from 8.2 restored to file that was pg_dump'd with format c
from a production server. I want to know if it is possible for me to
restore this to a 8.0 development server where I am not able to upgrade
at this time. Trying to do some testing, but this is the only other
server that's ava
On Fri, 2007-06-01 at 19:24 +0200, RW wrote:
> I've never tried this but maybe it works if you use pg_dump
> from 8.0 to do the dump.
>
> Greetings
> Robert
>
> Robert Fitzpatrick wrote:
> > I have a dump from 8.2 restored to file that was pg_dump'd with for
Can anyone suggest how one might be able to do this? I want to be able
to generate an 8 character random password for users in their password
field. Perhaps through the default setting of the field or a trigger
function. I found the following, but is there anything that can be used
on both Windows
Anyone know a package that can do this? Perferrably a Unix/Linux
package.
--
Robert
---(end of broadcast)---
TIP 3: Have you checked our extensive FAQ?
http://www.postgresql.org/docs/faq
I have a field with 'AA-BB-CC-DD' and I want to pull those four values
into an array and then loop through the array inserting records into a
table for each element. Can you someone point me to an example of this
in pl/pgsql?
--
Robert
---(end of broadcast)---
I am having a problem gettig a percent via division. Below is the first
part of my trigger function where pct returns 0.00, instead of the
expected 0.50. If I try without dimensions to the numeric variable, I
just get 0. What is the correct way to accomplish the percent?
CREATE OR REPLACE FUNCTION
I have a column that I want to sort by certain values. The values are
Unit, Exterior and Common. I want all the records with Unit first,
Common second and Exterior last in the sort order. These are the only 3
possible values, is there a way to sort manually like that with the
alphanumeric values?
I have a database that is used every day for the past year and all of a
sudden, this morning, I get a report that a user cannot login. I have
doubled checked the pg_hba.conf file, which has not been changed in
several months. But only this one user even after resetting the
password, any other user
Is it possible to use COALESCE function inside a function as a cursor
variable?
test cursor (myvar varchar) for
(coalesce(SELECT...,0));
I get a syntax error when trying this...ERROR: syntax error at or near
"COALESCE"...is there a way to do this?
--
Robert
I have a function that returns a set of a custom type...
trax=# \d client_search
Composite type "public.client_search"
Column | Type
+
clientno | bigint
client | character varying(100)
contactno | bigint
c
I have a trigger function that I want to apply to several tables, hence
my use of TG_RELNAME. I just want the record to get inserted if an
UPDATE comes from my view rule if the record for the client doesn't
already exist. This is what I have, but I'm finding the FOUND is not
returned for EXECUTE. H
I have a large database used with our mail filter. The pg_dumpall
results in about 3GB with this being the only database in the system
besides templates and postgres.
I do a vacuum every night after backup and it takes about an hour, is
this normal for this size db? I also have autovacuum enabled
On Thu, 2007-09-20 at 13:22 -0400, Bill Moran wrote:
> In response to Robert Fitzpatrick <[EMAIL PROTECTED]>:
>
> > I have a large database used with our mail filter. The pg_dumpall
> > results in about 3GB with this being the only database in the system
> >
On Thu, 2007-09-20 at 16:38 -0400, Bill Moran wrote:
> In response to Robert Fitzpatrick <[EMAIL PROTECTED]>:
> Why does everyone leave of the IO subsystem? It's almost as if many
> people don't realize that disks exist ...
>
> With 2G of RAM, and a DB that's
Trying to format our pg_dump files so that we can take advantage of
rsync delta copies and the problem seems to be the sequences and maybe
very active tables not allowing much matched data in the transfer. I
have a 3GB dump file in plain text and broke off the head of two dump
files into their own
I have a project where it is necessary to determine Outliers of lab
results and looking for some pointers on the best way to handle this
type of calculation with PostgreSQL. Possibly an operator? I have no
experience with that. I found some info on the web for calculating
Outliers, here is one of t
I have a PHP script that was having problems using the COPY command with
files around 1500 lines in size. The script will build the copy data
from incoming CSV file into a temp file, then start a COPY command and
loop through the copy data using pg_put_line to insert and then
pg_end_copy after post
Thanks to some help here on the list, I've been able to get addresses
sorting pretty well, but now I have a issue with same addresses on
different streets not grouping the streets. This is what I'm using a
substring search in the ORDER BY statement now like in this view:
SELECT tblhudsimilargroups
I searched through the net quickly and the list archives, but could not
find anything doing this. Is it possible? There is a Perl module for
doing this, but I guess cannot use modules for security reasons.
--
Robert
---(end of broadcast)---
TIP 2:
On Fri, 2004-11-05 at 16:48, Allen Landsidel wrote:
> On Fri, 5 Nov 2004 16:31:21 -0500, Goutam Paruchuri
> <[EMAIL PROTECTED]> wrote:
> >
> > Iam trying to import data from ms-sql server to postgres. I export the data
> > which has datetime columns in sql server using BCP. I use the following to
What does it mean when you drop a database and then recreate with the
same name and all the objects are still there. I want to wipe out the db
and put back from pg_restore. After I re-create the db, all the old
tables are back before I run pg_restore.
--
Robert
---(end o
On Thu, 2004-11-11 at 14:03, Doug McNaught wrote:
> Robert Fitzpatrick <[EMAIL PROTECTED]> writes:
>
> > What does it mean when you drop a database and then recreate with the
> > same name and all the objects are still there. I want to wipe out the db
> > and put ba
I have a view that used union all to merge three tables together. I was
hoping to create a rule, one for each table using the WHERE condition of
the rule to determine which table gets updated. Is this possible?
This is what I have, assume the view here is a merge of three tables
using union all:
On Wed, 2004-11-17 at 11:49, Michael Fuhr wrote:
> On Wed, Nov 17, 2004 at 11:20:41AM -0500, Robert Fitzpatrick wrote:
>
> > I have a view that used union all to merge three tables together. I was
> > hoping to create a rule, one for each table using the WHERE condition
On Wed, 2004-11-17 at 12:49, Robert Fitzpatrick wrote:
> On Wed, 2004-11-17 at 11:49, Michael Fuhr wrote:
> > On Wed, Nov 17, 2004 at 11:20:41AM -0500, Robert Fitzpatrick wrote:
> >
> > > I have a view that used union all to merge three tables together. I was
> >
I share the same sequences across 3 tables and have a COPY that send
data to one of those tables. That table has a before insert trigger
function that looks for a condition and redirect records to the other
two depending on that condition result. Sharing the sequence is needed
to act as a primary k
I have changed the default sequence on a primary key integer (created as
SERIAL) field in a table, but it will not let me drop the old sequence
and continues to tell me that the sequence is required by the table. Can
someone tell me what is required to get this dropped? I am just paranoid
that it c
I have a PHP 5 app using pgsql 8.2 and a HTML table of clients now grown
to almost 10,000 loading 25 per page. There is a filter feature atop all
seven columns in the table listing (all varchar except one date column).
Also, sorting can be done by clicking any column header. Some complain
of speed
I have a couple of servers running Postfix with amavisd-maia+SA+clamav
on FreeBSD 6.2 which use a central db server running PgSQL 8.2.4. My
issue is the bayes database causing SA TIMED OUT in the logs and want to
make sure I am getting everything into memory. The disk activity is high
on the db ser
On Thu, 2008-01-24 at 10:46 -0500, Bill Moran wrote:
> In response to Robert Fitzpatrick <[EMAIL PROTECTED]>:
>
> > How can I tell if PgSQL is using memory or not and
> > how much?
>
> Well, top is helpful. Also, consider installing the pg_buffercache addon
>
Been searching for a way to do this, but haven't found what I was hoping
to find. Is there any way in pl/pgsql to test a text value to see if it
would be a valid interval without having to try and store in a field? In
a trigger, I'd like to test a NEW text type field. Right now, I have
just the fol
On Thu, 2008-02-07 at 16:58 -0800, Jeff Davis wrote:
> On Thu, 2008-02-07 at 19:37 -0500, Robert Fitzpatrick wrote:
> > Been searching for a way to do this, but haven't found what I was hoping
> > to find. Is there any way in pl/pgsql to test a text value to see if it
> >
I have built posgresql 8.3.1 on a FreeBSD 6.1 via the ports system
WITH_DEBUG=true and also on a Windows machine with the debugger
selected. I see the plugin_debugger.dll under the lib/plugins folder on
the Windows machine as well. But when I look for the procs, they are not
there
template1=#
On Fri, 2008-05-16 at 14:27 +0100, Dave Page wrote:
> There is a README with the plugin which tells you how to modify
> postgresql.conf to preload the libraries at server start (which you
> must do). Once you've done that, just run the pldbgapi.sql script
> that's also included to load the api fun
We have PostgreSQL 8.1 running on Windows 2000 for a few weeks now, when
we try to start the service, it could not start claiming no error
returned. So, I go to the command prompt and run the following:
"C:\Program Files\PostgreSQL\8.1\bin\pg_ctl.exe" runservice -N
"pgsql-8.1" -D "C:\Program Files
We have PostgreSQL 8.1 running on Windows 2000 for a few weeks now. Don't know
what happened, the users reported a connection issue to the database and I
found the service will not start. When
we try to start the service, it could not start claiming no error
returned. So, I go to the command prom
I was wondering if someone could help with my new task of trying to
import from an API into PostgreSQL 9.0. An example query would respond
like this...
transactid=164d8355b0f4fc2eb051344d3b6b0b5f
status=SUCCESS
domaincount=3
domain_0_name=example1.com
domain_0_expiration=2011/06/16
domain_0_status
I am upgrading a Windows install for a client of mine from 8.2.x to
9.0.3 and understand the pginstaller does not provide plperl for this
version. ActivePerl 5.8 was already installed and after uninstalling 8.2
and installing 9.0.3, there is no plperl.dll in the lib folder. I
thought this was due t
I am upgrading a Windows install for a client of mine from 8.2.x to
9.0.3 and understand the pginstaller does not provide plperl for this
version. ActivePerl 5.8 was already installed and after uninstalling 8.2
and installing 9.0.3, there is no plperl.dll in the lib folder. I
thought this was due t
I am upgrading a Windows install from 8.2.x to 9.0.3. ActivePerl 5.8 was
already installed, but when I run the createlang command to install into
my database, I get...
could not load library “C:/Program Files/PostgreSQL/8.3/lib/plperl.dll”:
The specified module could not be found. SQL state: 58P01
On 2/8/2011 11:58 AM, Robert Fitzpatrick wrote:
> could not load library “C:/Program Files/PostgreSQL/8.3/lib/plperl.dll”:
> The specified module could not be found. SQL state: 58P01
Sorry the correct error I am getting is...
could not load library “C:/Program Files/PostgreSQL/9.0/lib/plpe
On 2/8/2011 12:03 PM, Robert Fitzpatrick wrote:
> On 2/8/2011 11:58 AM, Robert Fitzpatrick wrote:
>> could not load library “C:/Program Files/PostgreSQL/8.3/lib/plperl.dll”:
>> The specified module could not be found. SQL state: 58P01
>
> Sorry the correct error I am getting
I have a FreeBSD server with Postfix that filters mail using
amavisd-maia+SA+ClamAV. It crashed when we received an SMTP attack that
traced back to a compromised user login and a flood a messages were sent
to this smarthost. After getting it back up, I find this in the logs...
Nov 4 08:09:50 esmt
I've worked with pgsql for a while, but never needed to post from a
database trigger to another db. Is this possible? And if so, can someone
offer a pointer to the docs on how to refer to other db's in my script,
etc?
--
Robert
--
Sent via pgsql-general mailing list (pgsql-general@postgresql.o
On Wed, 2008-11-12 at 18:02 +0100, [EMAIL PROTECTED] wrote:
> > I've worked with pgsql for a while, but never needed to post from a
> > database trigger to another db. Is this possible? And if so, can someone
> > offer a pointer to the docs on how to refer to other db's in my script,
> > etc?
>
>
I try to use dblink_connect while logged into the database as the
superuser and it returns OK, but if I disconnect and use the same
command as follows, it give and error, password is required. Even if I
try to use the superuser login in the connect string below while
connected as a non-superuser, I
I am trying to develop a trigger that will post a new account into a
table in another db sing dblink that is part of the egroupware web app
that uses tripledes as the algorithm. I can't seem to find a combination
for gen_salt that produces the correct crypt password, however, my
knowledge in this a
Is it not possible to set the value of a field by reference of NEW.field
in an AFTER trigger? I have a trigger where I set NEW.field :=
myfunction and NEW.field shows the resulting value in my RAISE NOTICE as
shown here...
CREATE OR REPLACE FUNCTION "public"."registrants_tr_test_func" () RETURNS
I am getting the following error when trying to migrate a db from
8.4.0/FreeBSD 7.1 to 9.0.1/FreeBSD 8.1, both are VMware virtual machines
with FreeBSD amd64...
data1# psql -U pgsql template1 < /data/maia.sql
template1 < /data/incoming.sql
ERROR: invalid byte sequence for encoding "UTF8": 0x
My contacts table has a FK with cascade delete to foreign table
companies using the company_id column.
I have a DELETE AFTER trigger on my contacts table that checks to see if
there are any contacts left with an email address or it won't allow you
to delete the record for a company. However, if th
On 10/13/2011 5:45 PM, David Johnston wrote:
> the company record should not be visible
> if you execute a SELECT against the companies table using the given
> company_id value. The previous is not tested and I am not totally sure
> about the visibility rules in this situation (mainly whether the
I get this error after doing the upgrade on FreeBSD 5.2.1:
The data directory was initialized by PostgreSQL version 7.3, which is
not compatible with this version 7.4.2
Should I just re-initdb and the import my dump file?
--
Robert
---(end of broadcast)
Trying to portupgrade my PHP 4.3.6
package with PostgreSQL 7.4.2 support and getting this error below. Trying
to setup postgresql with Heimdal Kerberos 5 support, believe I have, but not
tried anything with it, yet.
/usr/local/lib/libpq.so: undefined reference to
`krb5_cc_get_principal'/us
I have plperl installed my PostgreSQL 7.4.2 server, but from what I
understand in chapter 39.3 of the docs, you cannot access the databases
without DBD::PgSPI. According to the readme for that module, it will
only run on the untrusted plperlu.
Is this the only way to run queries (SELECT, INSERT, U
On Wed, 2004-06-16 at 19:05, Paul Thomas wrote:
> On 16/06/2004 21:27 Robert Fitzpatrick wrote:
> > I have plperl installed my PostgreSQL 7.4.2 server, but from what I
> > understand in chapter 39.3 of the docs, you cannot access the databases
> > without DBD::PgSPI. Accord
On 7.4.2 I have a trigger that I want to update any existing boolean
values to false if a new one in that group is declare true by inserting
a new record or updating an existing record:
ohc=# CREATE OR REPLACE FUNCTION "public"."clear_common_groups" ()
RETURNS trigger AS'
ohc'# BEGIN
ohc'# IF NE
On Fri, 2004-06-18 at 13:17, Robert Fitzpatrick wrote:
> ohc=# CREATE OR REPLACE FUNCTION "public"."clear_common_groups" ()
> RETURNS trigger AS'
> ohc'# BEGIN
> ohc'# IF NEW.common_area = ''t'' THEN
> ohc&#x
I have a function that tells me if a record is positive and negative
based on several field values. I use it in select statements:
ohc=> SELECT sample_id, is_wipe_positive(tblleadwipe.sample_id) AS
positive FROM tblleadwipe WHERE hud_building_id IS NOT NULL;
sample_id | positive
---+-
I tried posting this to PHP-DB list, but no answers since yesterday. I'm
thinking it is not possible, but just trying to confirm. I am using
pg_exec($dbh, "COPY sometable FROM stdin USING DELIMITERS"). Once this
the connectino is made and this line has been issued, my script begins
analyzing each l
I have a function that does varous things, I can post it if necessary.
Basically, it deletes any related similar_group_id in tblhud74b, sets
sort_order field in tblhudunits to NULL and then proceeds to re-populate
tblhud74b based on certain calculations. The function runs fine through
all loops to
I have tblhud74b that has a foreign key to tblhudunits with Restrict
Updates. I have a function that DELETEs corresponding records before
doing an UPDATE right after the DELETE statement and receive the
following error:
ERROR: update or delete on "tblhudunits" violates foreign key
constraint "tbl
I have varchar column with both numbers and letters, like 1 thru 10 and
5A thru 5G, they are unit numbers for apartments. If I have 1 thru 100,
since it is a varchar field, it sorts like 1,10,11... instead of
1,2,3...
Is there any way to handle this without having to make a sort order
column?
--
I am getting 'postmaster does not shut down' when trying to stop the
database with '.../pg_ctl -D /path/to/datadir stop -m fast. How should I
proceed to get the database shut down and restarted? Are there other
options besides immediate shutdown flag? I dont' want to lose everything
I've worked on
I am running PostgreSQL 7.4.5 and have a trigger on a table called
tblriskassessors which inserts, updates or delete a corresponding record
in tblinspectors by lookup of a contact id and license number match. The
INSERT and DELETE work fine. The UPDATE works good unless I update the
license number.
Can someone point me to some more information or perhaps show an example
of returning a recordset from a plpgsql function. I'd like to send an
argument or arguments to the function, do some queries to return a set
of records. I've done several functions that return one value of one
type, but nothin
On Thu, 2004-09-23 at 18:28, Tim Penhey wrote:
> Robert Fitzpatrick wrote:
>
> >Can someone point me to some more information or perhaps show an example
> >of returning a recordset from a plpgsql function. I'd like to send an
> >argument or arguments to the function
Is it possible to return the field names of a composite type using a
function? Any examples or pointers?
--
Robert
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddress
96 matches
Mail list logo
