Re: ResultSet storing all rows with defaulftFetchSize to 5000
От | Vladimir Sitnikov |
---|---|
Тема | Re: ResultSet storing all rows with defaulftFetchSize to 5000 |
Дата | |
Msg-id | CAB=Je-G-_m46JH85Sg=PAg3BqExF-oELutUxNipaJMW-v_CemA@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: ResultSet storing all rows with defaulftFetchSize to 5000 (Tom Lane <tgl@sss.pgh.pa.us>) |
Ответы |
Re: ResultSet storing all rows with defaulftFetchSize to 5000
|
Список | pgsql-jdbc |
>(Yes, we do support WITH HOLD.)
Oh, is it something available through extended query protocol?
The documentation says "If successfully created, a named portal object lasts till the end of the current transaction, unless explicitly destroyed", however I just hope it is a documentation bug and one can create a named portal that can survive transaction end.
I don't think explicit "fetch ... from ..." is a workable solution for pgjdbc as it would incur high overhead (both maintenance, and performance):
1) declare ... binary is either full-binary or full text. pgjdbc implements just a subset of binary formats
2) `fetch ...` cannot be prepared, so I expect more network traffic and parse/execute overhead
3) Reimplementing "frontend" from "ExtendedQuery" to "Declare/fetch" is a great amount of work
Vladimir
В списке pgsql-jdbc по дате отправления: