double marshalling ...

[Havoc Pennington]

Hi,

It isn't immediately obvious to me what you're suggesting here, maybe
give me more concrete pointers to what old and new code would be like?

Havoc

On Wed, 2004-04-14 at 09:30, Michael Meeks wrote:
> Hi there,
> 
> 	I'm currently looking at chopping a swathe of the cut/paste code out of
> dbus-message.c and making the code more generic; better suited to a
> recursive type system (if that's ok).
> 
> 	It's interesting that the message marshalling stuff is all double cased
> - adding lots of branching in the inner marshalling loop. [ although I
> can see that this may be useful for the dbus-message-builder for testing
> (something nice we're missing in ORBit2) ]. Assuming that the client
> transmits the endianness of the message with the message itself - may I
> strip those special cases out of the code, and push up a slower
> swap_bytes based thing into the dbus-message-builder where it doesn't
> cause so much bloat in the common case ?
> 
> 	Regards,
> 
> 		Michael.