[interfacekit] Re: linking app_server to libbe
- From: Ingo Weinhold <bonefish@xxxxxxxxxxxxxxx>
- To: interfacekit@xxxxxxxxxxxxx
- Date: Wed, 12 Mar 2003 18:28:53 +0100 (MET)
On Wed, 12 Mar 2003, Isaac Yonemoto wrote:
> > I'm curious: Do you know any reason for using flattened BMessages? AFAIK
> > (unlike ours) the R5 app server is not linked against libbe and therefore
> > has to do similar hackery as the kernel to produce those.
>
> So as I understand it, app_server is not linked against libbe because
> libbe is designed so that on loading, the global be_app variable is
> initialized. Which process would probably lock up if the app_server
> weren't running yet.
>
> Is it unreasonable for, instead, the global be_app variable to be
> initialized the first time by the BApplication constructor, which would
> fail if the be_app global variable has a non-null value? This way,
> app_server could link and use all the goodies in libbe, so long as it
> doesn't try to make a BApplication object.
In fact, be_app is initialized in the BApplication constructor, as is
be_roster. So, I really have no idea, what would prevent the app server
from being linked against libbe.
BTW, our libopenbeos initializes be_roster at library initialization time,
which fails for both the app server and the registrar, but does no harm,
if you never use that variable or any (part of a) class accessing it.
CU, Ingo
- References:
- [interfacekit] linking app_server to libbe
- From: Isaac Yonemoto
Other related posts:
- » [interfacekit] linking app_server to libbe
- » [interfacekit] Re: linking app_server to libbe
- [interfacekit] linking app_server to libbe
- From: Isaac Yonemoto