Thread: mingw autoconf again

So I was fixing my MingW environment to test and fix that issue with the
functions missing. In doing so, I came across the error previously
discussed that gettimeofday() is now suddently defined in the latest
versions of mingw, but wasn't before.

So I figured I'd have to fix that. Attached is a patch for configure to
check this. I'd like comments from someone who knows autoconf better than
me on:
1) does it look ok?
2) is it ok to put checks for win32-only this way?


It does work on my mingw build, and with the new version of mingw the
functions aren't missing so I can actualliy build CVS HEAD to working state
with that one. Which just means I'll have to write another autoconf test
for the missing functions ni older mingw, but I'd like to get comments on
this one first.

//Magnus

Magnus Hagander wrote:
> So I was fixing my MingW environment to test and fix that issue with the
> functions missing. In doing so, I came across the error previously
> discussed that gettimeofday() is now suddently defined in the latest
> versions of mingw, but wasn't before.

Is it just missing from header files, or does it really not exist?

--
  Heikki Linnakangas
  EnterpriseDB   http://www.enterprisedb.com

Heikki Linnakangas wrote:
> Magnus Hagander wrote:
>> So I was fixing my MingW environment to test and fix that issue with the
>> functions missing. In doing so, I came across the error previously
>> discussed that gettimeofday() is now suddently defined in the latest
>> versions of mingw, but wasn't before.
>
> Is it just missing from header files, or does it really not exist?

It really does not exist. We have a replacement in port/.

//Magnus

Am Montag, 29. Oktober 2007 schrieb Magnus Hagander:
> 1) does it look ok?

yes

> 2) is it ok to put checks for win32-only this way?

Well, the available replacement implementation only works on win32, so this is
about as correct as we should care to be at the moment.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/