Re: [certi-dev] Problem with timeAdvanceGrant

[Eric Noulard]

2011/6/21 Christoph Laesche <address@hidden>:
> Hi,
>
> I tested my example using the cvs version and it seems to work there.
> I've now switched to using the cvs version, but I had to fix some issues
> for which I attached some patches.

Hi Christoph,

Thank you very much for you patches.
I'll try to process them soon in order to make those changes upstream.

In the meantime could you post them into the patches tracker:
https://savannah.nongnu.org/patch/?group=certi

it will be easier to track them from there
and if you register a (free) Savannah account and login before posting
them you'll get automatic follow-up message as soon as we process them.

> vector_cvs.patch: my previous patch ported to the cvs-version
>
> enabletimeregulation.patch: (ugly) fix for a bug in the
> enableTimeRegulation function of the rti1516ambasassador (this problem
> has also existed in the 3.4.0 version)
>
> fedtime_windows.patch: fix of CMakeList fixing a linker error on windows
>
> libvardata.patch: I moved VarData1516 to an extra library because I
> found no other way to compile it under windows (linker errors, fixing
> resulted into circular dependencies) - perhaps there is a better solution.

The design of the RTI library mainly interdependency between libRTI
and libFedTime
is a plain mistake.... precisely because circular ref makes it difficult
(if not impossible) to link on Windows or if you add visibility
handling in shared lib.
More information here: https://savannah.nongnu.org/bugs/?30003

I have some unfinished patch for that (it is attached to the bug tracker)
I'll have a look at your solution too.

-- 
Erk
Membre de l'April - « promouvoir et défendre le logiciel libre » -
http://www.april.org