Re: [certi-dev] jCerti should be compatible with CVS head

[Eric Noulard]

2010/3/29 Andrej Pancik <address@hidden>:
> Hello everyone,
>
>  I have tried to commit new update to jCerti but I did not succeed.
> CVS cries about me not having rsh (that is correct - I have windows 7
> and I am unable to get it to work). I am sorry for inconvenience but I
> have invested too much time into resolving this issue and I have
> decided to upload it into tracker for now:
> https://savannah.nongnu.org/task/download.php?file_id=20058

No problem I'll commit the content of the zip.
Concerning CVS access I shoudl have given you more detail.
Write access required SSH authentication.

You should:

1) Have generated an SSH key pair.
    On windows this can be done with the puttygen tool from PuTTY
    http://www.chiark.greenend.org.uk/~sgtatham/putty

2) Put your public key on you personal Savannah page:
    https://savannah.nongnu.org/my/admin/editsshkeys.php

3) Use an SSH client using the private key for doing all CVS actions
    Eclipse/Netbeans may use their own java SSH client or
    the can use "external" ssh client like PuTTY.

I'm just telling this as a reminder for any developper,
don't waste more time on this if you don't want to.

I'll put your modification into CVS without trouble.

>  This update should be compatible with current head of Certi however;
> I did not check it class by class so there may be some differences in
> behavior. This version of jCerti was built with old generator and
> manual modification of some classes. I am working on new generator
> that uses native types same way as C++ version does.

Ok very good.

>  Test federates such as uav-send/receive work flawlessly but I was not
> able to get the VirtualAir to work with Certi head (to be exact:
> monitor-asn.py with patched FlightGear effectively shutdowns RTIG
> process - that is definitely not normal). I will investigate this
> issue and post the bug report if appropriate.

I did fix a bug yesterday in the afternoon concerning some exception
that was thrown and killed the RTIG (and the RTIA),
depending on the time of your test may be an update would solve this issue.

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