[ros-users] Release of new stack client_rosjava

Nicholas Butko nbutko at ucsd.edu
Thu Dec 16 15:18:23 UTC 2010


It was in publisher and subscriber, not nodehandle.

Previously I emulated ros::ok() with something like

while(pub.isValid()) {
  pub.publish("Still awake");
}

If there is now a proper ros::ok(), this behavior should probably be
disabled for consistency.

--Nick

2010/12/16 Lorenz Mösenlechner <moesenle at in.tum.de>

> As far as I can see, NodeHandle::isValid doesn't call out to roscpp
> but tries to simulate the behavior of NodeHandle::ok by using a flag
> that is set on initialization and cleared on shutdown. The roscpp ok
> method also checks for ros::ok.
>
> I have added JNI wrappers for ros::ok and NodeHandle::ok now and
> pushed them into master.
>
> Lorenz
>
> > At some point I wrapped ros::ok in isValid. The rosjava bindings are
> pretty out of date.
> >
> > --Nick
> >
> >
> > On Dec 15, 2010, at 3:29 AM, Lorenz Mösenlechner <moesenle at in.tum.de>
> wrote:
> >
> > > Hi,
> > >
> > > it seems that the ok method is not wrapped yet. I will add that as
> > > soon as I find some spare time.
> > >
> > > I filed a ticket here:
> > > http://code.in.tum.de/indefero/index.php//p/client-rosjava/issues/1/
> > >
> > > Lorenz
> > >
> > >> Hi,
> > >>
> > >> is there an equivalent for ros::ok() available in rosjava? I have the
> > >> problem that my Java nodes don't get notified about the SIGINT, so I
> > >> don't really know how to stop a Java node (except for implementing a
> > >> user command for that, which is not too elegant)
> > >>
> > >> thanks in advance
> > >> cheers, Máté
> > >>
> > >> On 2010-12-09 17:13, Lorenz Mösenlechner wrote:
> > >>> Hi,
> > >>>
> > >>> rosjava has finally been released. We moved the packages `rosjava'
> and
> > >>> `test_rosjava' to the new stack `client_rosjava'. It can be checked
> > >>> out with
> > >>>
> > >>> git clone git://code.in.tum.de/git/client-rosjava.git client_rosjava
> > >>>
> > >>> Debian packages should be available soon.
> > >>>
> > >>> The release includes a new message generator that doesn't depend on
> > >>> genmsg_cpp anymore. The message format got slightly changed:
> > >>>
> > >>>  * The `bool' data type is now `boolean' instead of int in the Java
> > >>>    message file.
> > >>>
> > >>>  * Unsigned data types are now represented by the next bigger
> (signed)
> > >>>    Java data type where possible.
> > >>>
> > >>>  * Variable size arrays over complex types are now of type
> > >>>    `ArrayList<type>'
> > >>>
> > >>>  * Variable size arrays over primitive types still are normal Java
> > >>>    arrays
> > >>>
> > >>>  * Messages now provide the methods `equals' and `hashCode'
> > >>>
> > >>> Since Java message files cannot be included in the Debian packages,
> > >>> the way messages are generated had to be changed, too: if a package
> > >>> compiles java code, the message files of all dependencies are now
> > >>> generated inside the package that depends on them. That means rosjava
> > >>> can be used on shared installations now without requiring a pre-built
> > >>> jar file that contains the missing messages.
> > >>>
> > >>> Lorenz
> > >>>
> > >>>
> > >>
> > >>
> > >
> > _______________________________________________
> > ros-users mailing list
> > ros-users at code.ros.org
> > https://code.ros.org/mailman/listinfo/ros-users
>
> --
> Lorenz Mösenlechner            | moesenle at in.tum.de
> Technische Universität München | Boltzmannstr. 3
> 85748 Garching bei München     | Germany
> http://ias.cs.tum.edu/         | Tel: +49 (89) 289-26910
>
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20101216/6a91241b/attachment-0003.html>


More information about the ros-users mailing list