Re: json api WIP patch
От | james |
---|---|
Тема | Re: json api WIP patch |
Дата | |
Msg-id | 50EBC09E.8070303@mansionfamily.plus.com обсуждение исходный текст |
Ответ на | Re: json api WIP patch (Andrew Dunstan <andrew@dunslane.net>) |
Ответы |
Re: json api WIP patch
|
Список | pgsql-hackers |
> The processing functions have been extended to provide populate_record() and populate_recordset() functions.The latterin particular could be useful in decomposing a piece of json representing an array of flat objects (a fairly commonpattern) into a set of Postgres records in a single pass. So this would allow an 'insert into ... select ... from <unpack-the-JSON>(...)'? I had been wondering how to do such an insertion efficiently in the context of SPI, but it seems that there is no SPI_copy equiv that would allow a query parse and plan to be avoided. Is this mechanism likely to be as fast as we can get at the moment in contexts where copy is not feasible?
В списке pgsql-hackers по дате отправления: