Re: VARIANT / ANYTYPE datatype
От | Tom Lane |
---|---|
Тема | Re: VARIANT / ANYTYPE datatype |
Дата | |
Msg-id | 20824.1304557387@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Re: VARIANT / ANYTYPE datatype (Alvaro Herrera <alvherre@commandprompt.com>) |
Ответы |
Re: VARIANT / ANYTYPE datatype
|
Список | pgsql-hackers |
Alvaro Herrera <alvherre@commandprompt.com> writes: > As a followup idea there exists the desire to store records as records > and not text representation of same (given differing record types, of > course), for which it'd be more worthwhile. Maybe. The conventional wisdom is that text representation of data is more compact than PG's internal representation by a significant factor --- our FAQ says up to 5x, in fact. I know that that's including row overhead and indexes and so on, but I still don't find it to be a given that you're going to win on space with this sort of trick. Another point here is that once the values in question get to be any significant number of bytes apiece, both text and the proposed VARIANT representation could be subject to compression, which would greatly reduce any nominal advantage of one over the other. Any representational inefficiency in either would just be grist for the compressor. regards, tom lane
В списке pgsql-hackers по дате отправления: