apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Thom May <t...@planetarytramp.net>
Subject Re: Symbol name consistency
Date Sun, 17 Mar 2002 23:39:55 GMT
* Cliff Woolley (jwoolley@virginia.edu) wrote :
> On Sun, 17 Mar 2002, Thom May wrote:
> 
> > > > +apr_network_service_byname_get   from apr_getservbyname
> > > > +apr_network_ip_subnet_create     from apr_ipsubnet_create
> > > > +apr_network_ip_subnet_test       from apr_ipsubnet_test
> > > > +apr_network_socket_inherit_set   from apr_socket_set_inherit
> > > > +apr_network_socket_inherit_unset from apr_socket_unset_inherit
> > >
> > > Why put the word network in?
> > >
> > Agreed, but again in terms of consistency I felt I should list them.
> 
> On some of them, maybe.  But on the apr_socket_*, no.  Our naming
> convention is that the function is named after the apr_foo_t type it
> operates on, not what header file it's in.  apr_socket_set_inherit
> operates on an apr_socket_t, so the name is correct.
> 
Ah, ok. Thanks for clearing this up :-)
-Thom

Mime
View raw message