Re: Rename nodes/relation.h => nodes/pathnodes.h ?

Поиск
Список
Период
Сортировка
От Amit Langote
Тема Re: Rename nodes/relation.h => nodes/pathnodes.h ?
Дата
Msg-id CA+HiwqHLzq4WVbQHuAgyud20Fn3TqANDP6sw=wRvO8GesPT7iA@mail.gmail.com
обсуждение исходный текст
Ответ на Rename nodes/relation.h => nodes/pathnodes.h ?  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
On Tue, Jan 29, 2019 at 12:18 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> In the pluggable-storage discussion, there was some talk of renaming
> nodes/relation.h to avoid confusion with the new access/relation.h
> header.  I think this is a fine idea, not only because of that conflict
> but because "relation.h" has never made a lot of sense as the file's
> name.
>
> After a bit of thought, I propose "pathnodes.h" as the new name.
> That fits in with the other major headers in that directory
> (primnodes.h, parsenodes.h, plannodes.h, execnodes.h), and it seems
> like a reasonable summary of what's in it.  Admittedly, Path nodes
> as such are barely a third of the file's bulk; but I don't see any
> equally pithy way to describe the rest of it, unless something like
> planner_data.h, which is pretty unmelodious.

optnodes.h, as in optimization-related nodes?  I like pathnodes.h too though.

Thanks,
Amit


В списке pgsql-hackers по дате отправления:

Предыдущее
От: "Daniel Verite"
Дата:
Сообщение: Re: Alternative to \copy in psql modelled after \g
Следующее
От: Tom Lane
Дата:
Сообщение: Re: Alternative to \copy in psql modelled after \g