Re: WIP list rewrite - Mailing list pgsql-patches

From Neil Conway
Subject Re: WIP list rewrite
Date
Msg-id 40B16BF9.6010307@samurai.com
Whole thread Raw
In response to Re: WIP list rewrite  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: WIP list rewrite
List pgsql-patches
Tom Lane wrote:
> I'm surprised you did that much work and did not fix things to
> distinguish ListCell * from List * ... we're gonna have to make
> another pass over all these places to do that.

That would mean changing _every_ usage of foreach(), and a bunch of
other places besides; plus, once one changes an iteration variable to be
ListCell*, you get compiler warnings if you don't disable the list
compat API for the entire file. Since the hope was to make the changes
minimally invasive, I figured I could do this once the initial work is
in CVS.

Also, I'm not so worried about changes that are effectively
s/List/ListCell/ in a bunch of places -- they can largely be automated...

> Offer of help is still open, if you want to try to divvy up the
> remaining work.

I'd be happy to accept some help (Alvaro was kind enough to offer as
well), if we can figure out a reasonable way to divide the work up (say,
if this work were being done in a CVS branch... too late for that now
though). Any thoughts on how best to do this?

At any rate, once the initial work is in CVS then anyone can submit
patches to update particular parts of the tree to use the new API.

> I really want to see this get in.

As do I.

BTW, should I take your lack of comment on linitial() as "no objection"?

-Neil


pgsql-patches by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] pg_autovacuum fixes
Next
From: Bruno Wolff III
Date:
Subject: Re: Tuple sampling