httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From r..@engelschall.com (Ralf S. Engelschall)
Subject Re: cvs commit: apache-1.3/src/include compat.h
Date Mon, 13 Apr 1998 14:29:01 GMT

In article <3531F6DD.BD5CABC4@algroup.co.uk> you wrote:
> rse@hyperreal.org wrote:
>>   - renamed remaining os_ stuff in os/unix/ and os/bs2000/ to ap_ variants

> I don't remember buying into removal of os_ as part of adding ap_.
> Surely the correct thing to do is call it ap_os_ (or apos_, but I
> suppose that would be against the rules)? I think it is a definite
> backwards step to lose sight of which functions actually change on a
> per-OS basis.

I changed it to ap_xxx instead of ap_os_xxx because of two reasons:

1. Other os_xx functions were already renamed to ap_xx (e.g.
   os_is_path_absolute, os_escape_path, os_canonical_filename) and these
   changes were a long time in rename.cf without any complaining.

2. Our current consensus is that we just want homogen prefixes
   and that the meaningful prefixes will be done at a later time (Apache
   2.0)?.

So it seemed obvious to me that these remaining os_ functions were renamed to
ap_, too.  But I accept ap_os_xxx, too. The only think I personally want to
have in Apache 1.3 is a 100% consistent way. Whether ap_, ap_os_ or whatever
is not important at this stage (where we still don't have meaningful
prefixes), but partly ap_ and partly os_ for operating-dependend functions is
bogus. So renaming the remaining os_ ones to ap_ is just consequent, isn't it?

                                       Ralf S. Engelschall
                                       rse@engelschall.com
                                       www.engelschall.com

Mime
View raw message