httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ralf S. Engelschall" <...@engelschall.com>
Subject Re: [PATCH] Resolve BUILD warnings under ULTRIX/mips (take 4)
Date Tue, 14 Apr 1998 17:38:29 GMT

In article <Pine.BSF.3.95q.980414092034.23554B-100000@valis.worldgate.com> you wrote:
> On Tue, 14 Apr 1998, Ralf S. Engelschall wrote:

>[...]
>> + * Some patches for braindead Ultrix platforms

> I wouldn't say they are patches for Ultrix, since a patch is only
> something that modifies existing source.  This comment goes in the final
> source, not the patch.

Yeah, sorry, should read "kludge". It was a 1-minute-patch for Ben...

>> +#ifdef ULTRIX
>> +#undef fdopen
>> +FILE *ap_os_fdopen(int filedes, char *mode)
>> +{
>> +	return fdopen(filedes, mode);
>> +}
>> +#endif

> I don't like this.  It is ugly to have to go undefing fdopen and, if in
> theory, on Ultrix fdopen was a define to begin with (it could be) you
> would lose that.  

> In general, I like to avoid redefining standard routines.  If we need to
> create our own--even if it is only for one platform--we should do so and
> be done with it.  Why can't it all fit in a define like ap_select?

It can, no problem. But this needs more code changes which I wanted to avoid.
But feel free to make a patch for a ap_fdopen which makes Ben happy, too.  I
personally want to stay out of this issue in the following because I have no
Ultrix at hand to test and have no problem with a complaining braindead Ultrix
compiler...
                                       Ralf S. Engelschall
                                       rse@engelschall.com
                                       www.engelschall.com

Mime
View raw message