From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Alvaro Herrera <alvherre(at)2ndquadrant(dot)com> |
Cc: | Stephen Frost <sfrost(at)snowman(dot)net>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: COPY planning |
Date: | 2015-09-28 18:29:38 |
Message-ID: | 25695.1443464978@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Alvaro Herrera <alvherre(at)2ndquadrant(dot)com> writes:
> Yup. However I notice that there are a few other callers of
> expression_planner() that do not involve the optimizer for anything
> else. Maybe it makes sense to have a separate header file that's just
> #include "nodes/primnodes.h"
> extern Expr *expression_planner(Expr *expr);
> extern bool plan_cluster_use_sort(Oid tableOid, Oid indexOid);
> Seems it could be used by a large percentage of files currently
> including planner.h.
Hmm. I'd be a bit inclined to define such a file as "planner's
externally visible API", which would mean it should also include
planner() itself, and maybe also eval_const_expressions --- are
there external uses of that?
If we wanted to get rid of external uses of clauses.h, we'd probably
want to move some things like contain_mutable_functions() to nodeFuncs.c,
rather than put them into this hypothetical new header.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Josh Berkus | 2015-09-28 18:34:29 | Re: 9.3.9 and pg_multixact corruption |
Previous Message | Alvaro Herrera | 2015-09-28 18:22:34 | Re: COPY planning |