Обсуждение: pgsql (configure configure.in)

  Date: Wednesday, July 12, 2000 @ 18:58:55
Author: petere

Update of /home/projects/pgsql/cvsroot/pgsql
     from hub.org:/home/projects/pgsql/tmp/cvs-serv72398

Modified Files:
    configure configure.in

-----------------------------  Log Message  -----------------------------

Remove a bunch of unused configure tests, in particular cases where
* the result is not recorded anywhere
* the result is not used anywhere
* the result is only used in some places, whereas others have been getting away with it
* the result is used improperly

Also make command line options handling a little better (e.g., --disable-locale,
while redundant, should really still *dis*able).

Peter Eisentraut - PostgreSQL <petere@hub.org> writes:
> Remove a bunch of unused configure tests, in particular cases where
> * the result is only used in some places, whereas others have been getting away with it
> * the result is used improperly

Hmm.  Would you mind explaining the cases that fell in those categories?

In particular, I'd not want to see failure-to-obey-a-config-item in
places like ODBC be mistaken as proof that it's OK.  Some parts of our
code are more portable than others, and dumbing down to the least common
denominator is not the direction to move in...

            regards, tom lane

On Wed, 12 Jul 2000, Tom Lane wrote:

> > Remove a bunch of unused configure tests, in particular cases where
> > * the result is only used in some places, whereas others have been getting away with it
> > * the result is used improperly
>
> Hmm.  Would you mind explaining the cases that fell in those categories?

Not at all.

HEADER_STDC    -- result not used anywhere (except c.h, which used it wrongly)
arpa/inet.h    -- result not used
float.h        -- result not used everywhere
limits.h    -- result not used everywhere
netdb.h        -- result not used
netinet/in.h    -- result not used
sys/param.h    -- result not used

AC_TYPE_UID_T, AC_TYPE_MODE_T, AC_TYPE_OFF_T, AC_TYPE_SIZE_T
        -- results not recorded at all

AC_FUNC_MEMCMP, AC_TYPE_SIGNAL, AC_FUNC_VPRINTF
        -- ditto

All of this was strictly backend and libpq, and stuff that has existed at
least since 6.5.1.


--
Peter Eisentraut                  Sernanders vaeg 10:115
peter_e@gmx.net                   75262 Uppsala
http://yi.org/peter-e/            Sweden

eisentrp@csis.gvsu.edu writes:
> HEADER_STDC    -- result not used anywhere (except c.h, which used it wrongly)
> arpa/inet.h    -- result not used
> float.h        -- result not used everywhere
> limits.h    -- result not used everywhere
> netdb.h        -- result not used
> netinet/in.h    -- result not used
> sys/param.h    -- result not used

Hmm.  float.h and limits.h are both required by ANSI C, so it doesn't
surprise me that it's no longer necessary to test for their
availability.  (I guess not-quite-ANSI environments are getting scarcer
and scarcer.)  I am surprised that we're not seeing problems from
non-availability of the networking headers on some machines.  Oh well...
we can always add back the configure tests if we see proof they're
really needed, and in the meantime might as well save the cycles.

            regards, tom lane