Re: INTEGER range ("-2147483648" is not accepted.)
От | Satoshi Nagayasu |
---|---|
Тема | Re: INTEGER range ("-2147483648" is not accepted.) |
Дата | |
Msg-id | 4C20BCA9.9050400@gmail.com обсуждение исходный текст |
Ответ на | Re: INTEGER range ("-2147483648" is not accepted.) (Thom Brown <thombrown@gmail.com>) |
Список | pgsql-docs |
Thom, > Actually, come to think of it, shouldn't we have a gotchas page on the wiki? I agree with that it should be described in some tech document, but I don't have any good idea where/how it should be written. Basically, it's a parser issue, but app developers may meet it on their type casting (my guess), and it's a bit tricky. Regards, On 2010/06/22 18:57, Thom Brown wrote: > On 22 June 2010 10:46, Thom Brown<thombrown@gmail.com> wrote: >> On 22 June 2010 09:59, Satoshi Nagayasu<satoshi.nagayasu@gmail.com> wrote: >>> Magnus, >>> >>> Thanks for your advice. I've understood how it happens. >>> >>> However, it looks tricky and difficult to understand, >>> so I hope that the message could be more understandable >>> as Thom mentioned. >>> >>> Regards, >>> >> >> This does appear to be a gotcha, as the following returns a negative >> integer as expected: >> >> postgres=# SELECT -2147483648; >> ?column? >> ------------- >> -2147483648 >> (1 row) >> >> postgres=# SELECT pg_typeof(-2147483648); >> pg_typeof >> ----------- >> integer >> (1 row) >> >> And just in case... >> >> postgres=# SELECT pg_typeof(test.my_num) FROM (SELECT -2147483648) AS >> test(my_num); >> pg_typeof >> ----------- >> integer >> (1 row) >> >> So it's affected by the cast operator? >> >> Thom >> > > Actually, come to think of it, shouldn't we have a gotchas page on the wiki? > > Thom > -- NAGAYASU Satoshi <satoshi.nagayasu@gmail.com>
В списке pgsql-docs по дате отправления: