[BUGS] BUG #14754: ecpg SQL parsing error

Поиск
Список
Период
Сортировка
От zuberre@gmail.com
Тема [BUGS] BUG #14754: ecpg SQL parsing error
Дата
Msg-id 20170721141640.19351.88347@wrigleys.postgresql.org
обсуждение исходный текст
Ответы Re: [BUGS] BUG #14754: ecpg SQL parsing error  (Greg Stark <stark@mit.edu>)
Re: [BUGS] BUG #14754: ecpg SQL parsing error  (Michael Meskes <meskes@postgresql.org>)
Re: [BUGS] BUG #14754: ecpg SQL parsing error  (Michael Meskes <meskes@postgresql.org>)
Список pgsql-bugs
The following bug has been logged on the website:

Bug reference:      14754
Logged by:          Richard Zuber
Email address:      zuberre@gmail.com
PostgreSQL version: 9.5.7
Operating system:   CentOS 7.3.1611
Description:

Hello,

I’ve been using ecpg as part of my automated test framework to ensure my
various SQL migrations have proper syntax.  I believe I have run into an
error where ECPG is reporting a syntax error that does not in fact exist.

Assume the following table is defined (though this should not be relevant to
the issue):

CREATE TABLE foo
(  id   SERIAL NOT NULL,   name text NOT NULL,  PRIMARY KEY (id)
);

There are two files to be fed into ecpg (version - ecpg (PostgreSQL 9.5.7)
4.11.0) :

--doesnt_work.sql
EXEC SQL WITH cte AS
(  INSERT INTO foo(name)  VALUES  ('bar')  RETURNING id
)
SELECT * FROM foo;
--EOF

The output of “doesn’t_work.sql” is:

$ ecpg doesnt_work.sql
doesnt_work.sql:7: ERROR: syntax error at or near ")"
$ echo $?
3

Since this is a valid statement from a syntax perspective (ignoring what the
statement does), I would expect no error.  

The following appears to work as expected:

--works.sql
EXEC SQL WITH cte AS
(  INSERT INTO foo(name)  VALUES  ('bar')
)
SELECT * FROM foo;
--EOF

$ ecpg works.sql
$ echo $?
0

The difference between the two is the use of the “RETURNING” statement in
the CTE.  I have reviewed section 33.5 of the manual.  While I took note of
33.5.3 in the manual, this statement does not actually return a result set.
That said, I may be misinterpreting the documentation in this matter.  

I have reviewed release notes from later versions of postgres as well as the
open TODOs, but have not seen anything that appears to relate to this
behavior.

I appreciate your attention,
Richard Zuber


--
Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-bugs

В списке pgsql-bugs по дате отправления:

Предыдущее
От: buhhunyx@tut.by
Дата:
Сообщение: Re: [BUGS] BUG #14750: Seq Scan instead of Index Scan works without limitation
Следующее
От: kes-kes@yandex.ru
Дата:
Сообщение: [BUGS] BUG #14755: unexpected error