On Wednesday 15 April 2009 04:58:29 pm Paolo Abeni wrote:
> hello,
>
> On Wed, Apr 15, 2009 at 6:51 AM, Brad Hards <[email protected]> wrote:
> > I partly agree your logic, but that really implies that 0xC010 can never
> > contain a SearchFID, which means that it should not have appeared in that
> > section of MS-OXCDATA.
>
> So the official MS documentation is either wrong or inconsistent. Why
> this does not surprise me that much ?!? ;-)
Actually, wrong _and_ inconsistent :-)

To keep myself in a better frame of mind, I think of the documentation as like 
an open source project. Its an opportunity to contribute bug reports and 
patches...

> If that can help, I have a pcap trace and libmapi dump  containing
> the relevant notifications (and they match the patched IDL) obtained
> with a MS Exchange server 2007 SP 1. I think a libmapi test can't be
> easily setup due to current libmapi notification API.
I'm fairly sure the patched version will work with that specific release of 
Exchange 2007 SP1 - I trust your investigation. My concern (i.e. the reason 
why I haven't committed the fix) is "what might break" in either newer or 
older versions of exchange. 

I think testing notifications will require some form of special test server. 
I've got about 1% of the design done.

> BTW I have a few more patches in my local tree regarding notification
> (that should also simplify writing mapi tests for such situations),
> but, since they are quite intrusive I would like to wait for this one
> being processed/merged/dropped/rewritten before going further.
Sorry to be so slow on this. I had some real-life to deal with.

Brad
_______________________________________________
devel mailing list
[email protected]
http://mailman.openchange.org/listinfo/devel

Reply via email to