From: | Warren Turkal <turkal(at)google(dot)com> |
---|---|
To: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Fixed patch for timestamp refactor effort |
Date: | 2008-01-22 21:07:58 |
Message-ID: | 12010360793012-git-send-email-turkal@google.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Here is a fixed patch for the timestamp refactor. I apologize for sending the
wrong patch earlier.
I would like to get some comments on if this patch is heading in the right
direction. Here's the proposal for what I am doing:
Proposal for Refactoring of Timestamp Datatype
Goal:
The primary goal of this effort is to remove as many of the #ifdef tests for
HAVE_INT64_TIMESTAMP from timestamp.c[1] as possible. I have played with
PostgreSQL for many years and I look at this as an opportunity to give back to
the community.
Plan:
1. Replace datatypes selected by value of HAVE_INT64_TIMESTAMP with typedefs.
The typedefs will most likely be placed in timestamp.h[2].
2. Replace the places where the logic is determined by the value of
HAVE_INT64_TIMESTAMP with macros or function calls.
Further Investigation:
It may also be useful to investigate if PostgreSQL still needs two different
representations for timestamps (int64 or double).
[1]$(srcdir)/src/backend/utils/adt/timestamp.c
[2]$(srcdir)/src/include/utils/timestamp.h
Thanks,
wt
From | Date | Subject | |
---|---|---|---|
Next Message | Warren Turkal | 2008-01-22 21:07:59 | [PATCH] Add TimeOffset and DateOffset typedefs |
Previous Message | Alvaro Herrera | 2008-01-22 20:52:57 | Re: Suboptimal plan choice problem with 8.3RC2 |