Re: Revert back to standard AC_STRUCT_TIMEZONE Autoconf macro

Peter Eisentraut <peter.eisentraut@2ndquadrant.com>

From: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2019-10-02T05:30:43Z
Lists: pgsql-hackers

Attachments

On 2019-09-30 21:36, Tom Lane wrote:
> Peter Eisentraut <peter.eisentraut@2ndquadrant.com> writes:
>> Instead of AC_STRUCT_TIMEZONE we use our own variant called
>> PGAC_STRUCT_TIMEZONE that checks for tzname even if other variants were
>> found first.  But since 63bd0db12199c5df043e1dea0f2b574f622b3a4c we
>> don't use tzname anymore, so we don't need this anymore.
> 
> Hmm.  I wonder if we need AC_STRUCT_TIMEZONE either?  Seems like
> we should only be using our own struct pg_tm.

There are a few places that seem to need it, such as initdb/findtimezone.c.

> If we could get
> rid of that configure macro altogether, we could remove some dubious
> junk like plpython.h's "#undef HAVE_TZNAME".

We could keep just the part of AC_STRUCT_TIMEZONE that we need, namely
the check for tm_zone, and remove the part about tzname.

New patch attached.

-- 
Peter Eisentraut              http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

Commits

  1. Simplify PGAC_STRUCT_TIMEZONE Autoconf macro

  2. Integrate src/timezone library for all platforms. There is more we can