Re: temporal support patch

Поиск
Список
Период
Сортировка
От Jim Nasby
Тема Re: temporal support patch
Дата
Msg-id 4FC56139.4080304@nasby.net
обсуждение исходный текст
Ответ на Re: temporal support patch  (Miroslav Šimulčík <simulcik.miro@gmail.com>)
Ответы Re: temporal support patch  (Miroslav Šimulčík <simulcik.miro@gmail.com>)
Список pgsql-hackers
On 5/18/12 2:06 AM, Miroslav Šimulčík wrote:
> - no data redundancy - in my extension current versions of entries are stored only once in original table (in
table_log- entries are inserted to both original and log table) 

That's not necessarily a benefit... it makes querying for both history *and* current data a lot more complex. Table
inheritancemight be an elegant solution to that, but I doubt you could just bolt that on top of what you've created. 

The timestamp fields need to have timezone info. If you change the timezone for a connection you will get inconsistent
resultswithout it. 

_sys_end should either be NULLable or if it's going to have a magic value that magic value should be "Infinity":

decibel@workbook.local=# create table t(t timestamptz);
CREATE TABLE
decibel@workbook.local=# insert into t values('infinity');
INSERT 0 1
decibel@workbook.local=# select * from t;    t
---------- infinity
(1 row)

decibel@workbook.local=#
--
Jim C. Nasby, Database Architect                   jim@nasby.net
512.569.9461 (cell)                         http://jim.nasby.net


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Jeff Davis
Дата:
Сообщение: Re: Add primary key/unique constraint using prefix columns of an index
Следующее
От: Robert Haas
Дата:
Сообщение: hash index concurrency