On Thu, Jul 11, 2002 at 11:17:50AM -0700, Justin Erenkrantz wrote:
> On Thu, Jul 11, 2002 at 01:59:16PM -0400, Kevin Pilch-Bisson wrote:
> > On Thu, Jul 11, 2002 at 10:58:54AM -0700, Justin Erenkrantz wrote:
> > > Would anyone have a problem if I move the svn_wc_adm_access_t
> > > declaration and functions to the top of svn_wc.h? In order to use
> > > the baton in svn_wc calls, it has to be declared first. -- justin
> >
> > I'd rather changed the functions that need it to take a void * (or maybe an
> > incomplete type, and keep the details out of the public header personally.
>
> There's a problem with that as the svn_wc_entry is going to need
> to know the internals of this structure (as it needs to populate
> the cache).
>
> So, I don't think we'll be able to get away with just confining the
> real structure to lock.c as entries.c will need it too. -- justin
>
Sure, but couldn't we put the definition in subversion/libsvn_wc/wc.h instead
of include/svn_wc.h. That way the whole wc library could use it, but it stays
out of the public API. If it was in svn_wc.h then someone writing a client
and trying to use libsvn_wc might think they had to do something with it other
than just pass it around.
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kevin Pilch-Bisson http://www.pilch-bisson.net
"Historically speaking, the presences of wheels in Unix
has never precluded their reinvention." - Larry Wall
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- application/pgp-signature attachment: stored
Received on Thu Jul 11 20:21:13 2002