httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <dgau...@arctic.org>
Subject Re: cvs commit: apache-1.3/src/include hide.h
Date Wed, 04 Mar 1998 20:07:13 GMT
This is so silly.  It can't break any modules which -DHIDE and #include
hide.h.  If we declare that to be the default (which I think we're
agreeing on) then 3rd party modules have no problem following suit. 
Rasmus is broken right now because he's not doing that. 

Dean

On Wed, 4 Mar 1998, Jim Jagielski wrote:

> Ralf S. Engelschall wrote:
> > 
> > Yes, sounds nice. Even we are doing it at the wrong location, I think.  But to
> > make this we need a decision scheme for the prefixing.  We can do it this way:
> > Have a list of apapi-functions and all functions which are not in this list
> > gets appriv as there prefix.
> > 
> > Hmmmm... seems like the HIDE-feature becomes a general symbol changing
> > feature, which IMHO is not for what it was intended and written... But +0.
> > 
> 
> This sounds wrong to me... We already have symbols which are
> "hidden", meaning only that they are designed to avoid name-space
> collisions. These are the real API functions. These should NOT
> be then re-hidden. This makes no sense and is SURE to completely
> hose external modules. Only those functions which are non-static
> and not API should be "hidden" and even then, I'm sure there are
> some that are being used. I think that our 1st cut of hidden
> should be those collisions that we _know_ about.
> 
> Soooo to make it official, I'm -1 on any hiding of already
> hidden, ap_* functions...
> -- 
> ====================================================================
>       Jim Jagielski            |       jaguNET Access Services
>      jim@jaguNET.com           |       http://www.jaguNET.com/
>             "Look at me! I'm wearing a cardboard belt!"
> 


Mime
View raw message