Re: Proposal: TABLE functions
От | Pavel Stehule |
---|---|
Тема | Re: Proposal: TABLE functions |
Дата | |
Msg-id | BAY114-F385390560C897143BEBF04F99C0@phx.gbl обсуждение исходный текст |
Ответ на | Re: Proposal: TABLE functions (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-hackers |
> >> I thought you said this was just syntactic sugar for capabilities we > >> already had? > > > My mistake. I am sorry. I have to store somewhere flag. One bit, which > > signalise "don't use OUT arguments as function's parameters". > >Huh? What exactly is the meaning of the arguments then? > >It sounds to me like this might be better thought of as a new >proargmode value, but I'm quite unsure what you're talking about ... > My basic idea was: CREATE FUNCTION aaa(IN a1, OUT a, OUT b) RETURNS SETOF RECORD AS $$ .. is similar CREATE FUNCTION aaa(IN a1) RETURNS SETOF RECORD AS $$ from executor perspective there isn't any difference. But PL languages have to create only IN variables. It's protection before identifier's name colision. With special flag I don't need any changes in executor. And small change in PL compile rutines. Special proargmode can be solution too. I don't need new column in pg_proc, but have to modify executor and need more changes in output rutines in PL. I'll go on the way to spec. proargmode. It's good idea. Thank You Pavel Stehule _________________________________________________________________ Emotikony a pozadi programu MSN Messenger ozivi vasi konverzaci. http://messenger.msn.cz/
В списке pgsql-hackers по дате отправления: