Re: [PATCHES] to_date() validation
| От | Alex Hunsaker |
|---|---|
| Тема | Re: [PATCHES] to_date() validation |
| Дата | |
| Msg-id | 34d269d40809091827le21056co509469c0684f6d55@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: [PATCHES] to_date() validation ("Brendan Jurd" <direvus@gmail.com>) |
| Список | pgsql-hackers |
On Mon, Sep 8, 2008 at 2:24 AM, Brendan Jurd <direvus@gmail.com> wrote: > HEAD actually gets this one wrong; in defiance of the documentation it > returns 2000-09-07. So, it seems to me that the patch shifts the > behaviour in the right direction. > > Barring actually teaching the code that some nodes (like YYYY) can > take an open-ended number of characters, while others (like MM) must > take an exact number of characters, I'm not sure what can be done to > improve this. Perhaps something for a later patch? Sound good to me and I would probably argue that things like MM should not be hard coded to take only 2 chars... But then again to play devils advocate I can just as easily do things like to_char(...) + '30 months'::interval;
В списке pgsql-hackers по дате отправления: