also sprach Pierre Habouzit <[EMAIL PROTECTED]> [2008.01.09.2229 +0100]:
> > The alternative is to have the control socket use the D-Bus system,
> > meaning only one package is needed, but messages
> > would need to pass through the dbus-daemon in order to be recieved.
>
> Well that wasn't what
* "Joe Smith"
| The alternative is to have the control socket use the D-Bus system,
| meaning only one package is needed, but messages
| would need to pass through the dbus-daemon in order to be recieved.
You can use point-to-point DBus connections without a daemon if you
for some reason desire
"Pierre Habouzit" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
Well that wasn't what I understood, but I'm really not a D-Bus expert
at all :) Though it doesn't makes sense to let the D-Bus connector be a
separated component as you then only pull the library which is of a
reasona
On Wed, Jan 09, 2008 at 06:33:31PM +, Joe Smith wrote:
> "Pierre Habouzit" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]
> >I assume that it's
> >possible to write tools that directly hit your netconf server withouth
> >going through the dbus daemon, making it lightweight and a
"Pierre Habouzit" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
That's wrong, and XML-RPC *SUCKS*, as does most of the text-only
interfaces, when you want real-time events. DBus isn't such a bad way to
do things, it doesn't requires the daemon up and running to interact
with netc
On mar, jan 08, 2008 at 11:32:29 +, martin f krafft wrote:
> While working on Ikiwiki, it dawned on me that I really ought to be
> using XML RPC for this [2]. Why? Because it's already
> there to do exactly the kind of thing I am doing, and standardised.
> Furthermore, dbus *is* XML RPC.
Tha
On 1/8/08, martin f krafft <[EMAIL PROTECTED]> wrote:
> Furthermore, dbus *is* XML RPC.
Umm, no -- dbus is a custom binary protocol.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Le mardi 08 janvier 2008 à 12:32 +0100, martin f krafft a écrit :
> Many people have suggested using dbus for this, but I always refused
> because I did not want the dependency.
OK, now just have a look at what has been proposed, and please
reconsider this decision. The dbus library is 300 KB of
Hi,
martin f krafft wrote:
However, that would be
independendent of the control socket, for which I still need
a protocol.
CORBA?
There is excellent support for it from C++, and since the interface
definitions are compiled into stub code rather than including a full
parser covering all eve
On Tue, 2008-01-08 at 12:32 +0100, martin f krafft wrote:
> Hi there,
>
> As you may know, netconf[0] sports a control socket with which you
> can control a running instance of the daemon. The idea is that tools
> like /sbin/ifup then are nothing more than clients of this socket,
> issuing the rig
Hi there,
As you may know, netconf[0] sports a control socket with which you
can control a running instance of the daemon. The idea is that tools
like /sbin/ifup then are nothing more than clients of this socket,
issuing the right commands to the daemon process.
0. http://netconf.alioth.debian.or
11 matches
Mail list logo
