Re: JDBC and positioned updates
От | Kevin Wooten |
---|---|
Тема | Re: JDBC and positioned updates |
Дата | |
Msg-id | AD4D45AA-DA82-42FC-9F48-EA1033F1BF0F@me.com обсуждение исходный текст |
Ответ на | Re: JDBC and positioned updates (Hédi HACHENI <hacheni@kopileft.com>) |
Список | pgsql-jdbc |
https://github.com/impossibl/pgjdbc-ng supports ‘setCursorName’.
PostgreSQL doesn’t support all capabilities defined for cursors but what you’ve referenced should work.
On Dec 26, 2015, at 1:53 AM, Hédi HACHENI <hacheni@kopileft.com> wrote:Basically we will not perform positioned updates on the hole table data but for a data portion that it would not exceed 30 rows. That's why I said it depends on what table we're going through.
But we take care about queries performances.
On 12/26/2015 09:21 AM, Vladimir Sitnikov wrote:we cannot achieve positioned updates using the JDBC implementationThis is "functional requirement"All of these parameters depends on what table we're going throughThose a "non-functional requirements" (see [1]).
You'd better collect NFRs *before* you finalize design and write the code.Our main problem isSuppose you somehow achieved "positional updates". What if it turns
out to be super-slow? Believe me, there are good reasons for
"positional updates" to be super slow.
You would have to rewrite the whole thing from scratch to make it fast.
This is why I do not want to help you to shoot into your own foot.
Anyway, pull requests are welcome.
[1]: https://en.wikipedia.org/wiki/Non-functional_requirement
Vladimir
--
Sent via pgsql-jdbc mailing list (pgsql-jdbc@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-jdbc
В списке pgsql-jdbc по дате отправления: