Weird "could not determine which collation to use for string comparison" with LEAST/GREATEST on PG11 procedure

Started by Voillequin, Jean-Marcover 7 years ago5 messagesbugs
Jump to latest
#1Voillequin, Jean-Marc
Jean-Marc.Voillequin@moodys.com

Hello,

When I run the following script on a PG 11 psql:

select version();
show lc_collate;
create or replace function same_values_func(a text, b text) returns void as $body$
begin
assert a = b;
end;$body$ language plpgsql;
select same_values_func(least('a','b'),'a');
create or replace procedure same_values_proc(a text, b text) as $body$
begin
assert a = b;
end;$body$ language plpgsql;
call same_values_proc(least('a','b'),'a');

I get the following output and error at the end:

SIMPLE=> select version();
version
--------------------------------------------------------------------------------------------------------
PostgreSQL 11.0 on x86_64-pc-mingw64, compiled by gcc.exe (Rev5, Built by MSYS2 project) 4.9.2, 64-bit
(1 row)

SIMPLE=>
SIMPLE=> show lc_collate;
lc_collate
------------
C
(1 row)

SIMPLE=>
SIMPLE=> create or replace function same_values_func(a text, b text) returns void as $body$
SIMPLE$> begin
SIMPLE$> assert a = b;
SIMPLE$> end;$body$ language plpgsql;
CREATE FUNCTION
SIMPLE=>
SIMPLE=> select same_values_func(least('a','b'),'a');
same_values_func
------------------

(1 row)

SIMPLE=>
SIMPLE=> create or replace procedure same_values_proc(a text, b text) as $body$
SIMPLE$> begin
SIMPLE$> assert a = b;
SIMPLE$> end;$body$ language plpgsql;
CREATE PROCEDURE
SIMPLE=>
SIMPLE=> call same_values_proc(least('a','b'),'a');
ERROR: could not determine which collation to use for string comparison
HINT: Use the COLLATE clause to set the collation explicitly.
SIMPLE=>

I tried to convert all void functions we have on a PG10 DB to procedures and I get this strange error.
The same error occurs with GREATEST.
Any idea?
Regards.

-----------------------------------------

Moody's monitors email communications through its networks for regulatory compliance purposes and to protect its customers, employees and business and where allowed to do so by applicable law. The information contained in this e-mail message, and any attachment thereto, is confidential and may not be disclosed without our express permission. If you are not the intended recipient or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that you have received this message in error and that any review, dissemination, distribution or copying of this message, or any attachment thereto, in whole or in part, is strictly prohibited. If you have received this message in error, please immediately notify us by telephone, fax or e-mail and delete the message and all of its attachments. Every effort is made to keep our network free from viruses. You should, however, review this e-mail message, as well as any attachment thereto, for viruses. We take no responsibility and have no liability for any computer virus which may be transferred via this e-mail message.

This email was sent to you by Moody’s Investors Service EMEA Limited
Registered office address:
One Canada Square
Canary Wharf
London, E14 5FA
Registered in England and Wales No: 8922701

-----------------------------------------

#2Tom Lane
tgl@sss.pgh.pa.us
In reply to: Voillequin, Jean-Marc (#1)
Re: Weird "could not determine which collation to use for string comparison" with LEAST/GREATEST on PG11 procedure

"Voillequin, Jean-Marc" <Jean-Marc.Voillequin@moodys.com> writes:

SIMPLE=> create or replace procedure same_values_proc(a text, b text) as $body$
SIMPLE$> begin
SIMPLE$> assert a = b;
SIMPLE$> end;$body$ language plpgsql;
CREATE PROCEDURE
SIMPLE=>
SIMPLE=> call same_values_proc(least('a','b'),'a');
ERROR: could not determine which collation to use for string comparison
HINT: Use the COLLATE clause to set the collation explicitly.

Yeah, same here. I think somebody forgot to run assign_expr_collations()
on CALL arguments.

regards, tom lane

#3Peter Eisentraut
peter_e@gmx.net
In reply to: Tom Lane (#2)
Re: Weird "could not determine which collation to use for string comparison" with LEAST/GREATEST on PG11 procedure

On 21/11/2018 19:19, Tom Lane wrote:

"Voillequin, Jean-Marc" <Jean-Marc.Voillequin@moodys.com> writes:

SIMPLE=> create or replace procedure same_values_proc(a text, b text) as $body$
SIMPLE$> begin
SIMPLE$> assert a = b;
SIMPLE$> end;$body$ language plpgsql;
CREATE PROCEDURE
SIMPLE=>
SIMPLE=> call same_values_proc(least('a','b'),'a');
ERROR: could not determine which collation to use for string comparison
HINT: Use the COLLATE clause to set the collation explicitly.

Yeah, same here. I think somebody forgot to run assign_expr_collations()
on CALL arguments.

This appears to fix it.

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Attachments:

0001-Add-collation-assignment-to-CALL-statement.patchtext/plain; charset=UTF-8; name=0001-Add-collation-assignment-to-CALL-statement.patch; x-mac-creator=0; x-mac-type=0Download+20-1
#4Tom Lane
tgl@sss.pgh.pa.us
In reply to: Peter Eisentraut (#3)
Re: Weird "could not determine which collation to use for string comparison" with LEAST/GREATEST on PG11 procedure

Peter Eisentraut <peter.eisentraut@2ndquadrant.com> writes:

On 21/11/2018 19:19, Tom Lane wrote:

Yeah, same here. I think somebody forgot to run assign_expr_collations()
on CALL arguments.

This appears to fix it.

I think this should be fine as a band-aid patch. As I mentioned
previously, I'm not really happy with our generally-unprincipled
approach to where collation assignment is called from ... but a
bug-fix patch should probably not be tasked with making that
better. Especially not with less than a week till 11.2.

regards, tom lane

#5Peter Eisentraut
peter_e@gmx.net
In reply to: Tom Lane (#4)
Re: Weird "could not determine which collation to use for string comparison" with LEAST/GREATEST on PG11 procedure

On 05/02/2019 15:46, Tom Lane wrote:

Peter Eisentraut <peter.eisentraut@2ndquadrant.com> writes:

On 21/11/2018 19:19, Tom Lane wrote:

Yeah, same here. I think somebody forgot to run assign_expr_collations()
on CALL arguments.

This appears to fix it.

I think this should be fine as a band-aid patch. As I mentioned
previously, I'm not really happy with our generally-unprincipled
approach to where collation assignment is called from ... but a
bug-fix patch should probably not be tasked with making that
better. Especially not with less than a week till 11.2.

committed

--
Peter Eisentraut http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services