Re: CREATE LANGUAGE
От | Tom Lane |
---|---|
Тема | Re: CREATE LANGUAGE |
Дата | |
Msg-id | 15546.997542570@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | CREATE LANGUAGE (Peter Eisentraut <peter_e@gmx.net>) |
Ответы |
Re: CREATE LANGUAGE
|
Список | pgsql-hackers |
Peter Eisentraut <peter_e@gmx.net> writes: > Is there a reason why it's CREATE LANGUAGE 'string' and not CREATE > LANGUAGE "ident"? I'd like to allow both to make it consistent with the > other create commands. SQL also uses identifier syntax for language > names. Seems reasonable to me. Don't overlook the CREATE FUNCTION ... LANGUAGE clause, too. > Also, what is LANCOMPILER for? If it's just a comment we ought to use > pg_description. I think the original idea might have been to someday support auto-building of compiled functions. (Though a setup allowing one to invoke make with suitable arguments would probably be far more useful than a bare compiler name.) I'm not in a hurry to rip it out until we have designed such a facility, anyway. There are a lot of vestigial features in Postgres that we might someday resurrect/implement. I tend to view those things as TODO markers, not stuff to rip out to save a byte or two. regards, tom lane
В списке pgsql-hackers по дате отправления: