Re: possible/feasible to specify field and value in error msg?

Поиск
Список
Период
Сортировка
От Bruce Momjian
Тема Re: possible/feasible to specify field and value in error msg?
Дата
Msg-id 20130703151855.GB6521@momjian.us
обсуждение исходный текст
Ответ на Re: possible/feasible to specify field and value in error msg?  (Bruce Momjian <bruce@momjian.us>)
Ответы Re: possible/feasible to specify field and value in error msg?  (Willy-Bas Loos <willybas@gmail.com>)
Список pgsql-hackers
On Wed, Jul  3, 2013 at 11:14:18AM -0400, Bruce Momjian wrote:
> On Wed, Jul  3, 2013 at 10:54:48AM +0200, Willy-Bas Loos wrote:
> > Hi,
> > 
> > I have some complicated query that truncates and fills a table and i get this
> > message:
> > ERROR:  smallint out of range
> > STATEMENT: <my huge query>
> > This is in postgres 8.4
> > I don't know where the error is, and the query takes rather long. So it is
> > going to be a bit cumbersome for me to debug this.
> > 
> > Would it be possible/feasible to specify, in future versions of postgres:
> > * what value
> > * which field (of which table)
> > * the offending tuple? (possibly truncated to some threshold nr of characters)
> > 
> > I ask because i can imagine that, inside the code that handles this, you might
> > not have access to that information and adding access to it might be
> > inefficient.
> > 
> > I do get the whole query of course, and that is very handy for automated
> > things. But in this case, it doesn't help me.
> 
> We will add optional error details in Postgres 9.3:
> 
>     http://momjian.us/main/blogs/pgblog/2013.html#April_11_2013
> 
> I don't know if an out-of-range error would generate the column name.

I just tested this and it doesn't show the offending column name; 
sorry:
test=> CREATE TABLE test(x smallint);CREATE TABLEtest=> \set VERBOSITY verbosetest=> INSERT INTO test VALUES
(10000000);ERROR: 22003: smallint out of rangeLOCATION:  i4toi2, int.c:349
 

--  Bruce Momjian  <bruce@momjian.us>        http://momjian.us EnterpriseDB
http://enterprisedb.com
 + It's impossible for everything to be true. +



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Alvaro Herrera
Дата:
Сообщение: Re: dynamic background workers
Следующее
От: Robert Haas
Дата:
Сообщение: Re: Patch to add regression tests for SCHEMA