Re: BUG #3037: strange behave of CHECK constraint
От | Pavel Stehule |
---|---|
Тема | Re: BUG #3037: strange behave of CHECK constraint |
Дата | |
Msg-id | BAY114-F3882FD09D541A289EBC55BF98E0@phx.gbl обсуждение исходный текст |
Ответ на | Re: BUG #3037: strange behave of CHECK constraint (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-bugs |
>"Pavel Stehule" <pavel.stehule@hotmail.com> writes: > > root=# create table products( > > barcode char(13) NOT NULL > > check (barcode NOT similar to '%[^0-9]%') > > ); > > > root=# insert into products values('22'); > > ERROR: new row for relation "products" violates check constraint > > "products_barcode_check" > > > root=# select '22' NOT similar to '%[^0-9]%'; > > ?column? > > ---------- > > t > >That isn't a counterexample, because you forgot about the trailing >spaces, which can match [^0-9]. > >regression=# select '22' NOT similar to '%[^0-9]%'; > ?column? >---------- > t >(1 row) > >regression=# select '22'::char(13) NOT similar to '%[^0-9]%'; > ?column? >---------- > f >(1 row) > > I see it now. Thank You. It's one trick from Joe Celko and I too late understand it. Nice day Pavel Stehule _________________________________________________________________ Emotikony a pozadi programu MSN Messenger ozivi vasi konverzaci. http://messenger.msn.cz/
В списке pgsql-bugs по дате отправления: