From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Simon Riggs <simon(at)2ndQuadrant(dot)com> |
Cc: | pgsql-hackers(at)postgreSQL(dot)org |
Subject: | Re: Another refactoring proposal: move stuff into nodes/nodeFuncs.[ch] |
Date: | 2008-08-27 13:40:56 |
Message-ID: | 22496.1219844456@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Simon Riggs <simon(at)2ndQuadrant(dot)com> writes:
> On Mon, 2008-08-25 at 11:24 -0400, Tom Lane wrote:
>> The advantages of doing this would be (a) reduce the number of places
>> to look in when implementing a new node type; (b) eliminate some
>> cross-subsystem #inclusions that weaken modularity of the backend.
> Are we doing either of those things in this release?
Yes, it's done already.
> Might these changes screw up patches already in progress? Can we hold
> off making these changes until we're sure the latter isn't true?
Some patches might need small adjustments (to find the code in a
different file) but that does not strike me as an argument for not
changing things. We have applied far more invasive patches in the past
and undoubtedly will do so again in future. In any case, it's still a
long way until beta freeze, so there's plenty of time to deal with
fallout. I don't see that some other time would be better.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Heikki Linnakangas | 2008-08-27 13:51:02 | Re: Is it really such a good thing for newNode() to be a macro? |
Previous Message | Alvaro Herrera | 2008-08-27 13:05:54 | Re: TODO <-> Commitfest |