Re: BUG #2905: min and max return incorrect text type
От | Bruce Momjian |
---|---|
Тема | Re: BUG #2905: min and max return incorrect text type |
Дата | |
Msg-id | 200701232302.l0NN2wl27035@momjian.us обсуждение исходный текст |
Ответ на | Re: BUG #2905: min and max return incorrect text type (Alvaro Herrera <alvherre@alvh.no-ip.org>) |
Список | pgsql-bugs |
Alvaro Herrera wrote: > Adriaan van Os wrote: > > Tom Lane wrote: > > >Bruce Momjian <bruce@momjian.us> writes: > > >>Peter Eisentraut wrote: > > >>>Aside from the apparent discrepancy between the documentation and the > > >>>actual behavior, is there an actual use case where this is a problem? > > > > > >>No, I don't think so, and I am reluctant to adjust the documentation to > > >>say "or similar". > > > > > >The documentation is correct as it stands: max(text) returns text. > > >Adriaan's complaint about max(varchar) is off base because there is > > >no such function. > > > > No, the documentation says that the Argument Type of max and min can be > > "any array, numeric, string, or date/time type" and that the Return type is > > the "same as argument type". The functions min and max applied to a field > > of type varchar return a function result of type text. So, if a > > max(varchar) function is missing and the cause of the text result type is > > implicit type casting, then the fact that there is no max(varchar) function > > is exactly the bug. > > Do you have a specific situation on which this causes a problem for you? > I mean, are you asking because it really bugs you, or just for the sake > of being pedantic? And what suggestion do you have for a change? How would you like the documentation wording changed? -- Bruce Momjian bruce@momjian.us EnterpriseDB http://www.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +
В списке pgsql-bugs по дате отправления: