java-player-users

[Java-player-users] setPosition message format

[John O. <joh...@gm...>]

I have been having some trouble with getting setPosition to work on a
Position2DInterface.  (Player/stage 2.0.1 and Javaclient 2.0.1)

Taking a quick(well perhaps not that quick) look at the code in player the
pack function (player_position2d_cmd_pack) it appears to look for both a
position command and a velocity in the message:

...
  if(xdr_player_pose_t(&xdrs,&msg->pos) != 1)
    return(-1);
  if(xdr_player_pose_t(&xdrs,&msg->vel) != 1)
    return(-1);
  if(xdr_u_char(&xdrs,&msg->state) != 1)
    return(-1);
...

while the java version only sends the
position(Position2DInterface.setPosition), and from what I can tell this is
leading to my "warning : decoding failed on message to 4:1 with type 2"
messages.

Have I made a mistake here or is this something to be patched?  And is it a
player problem or JavaClient?  My instinct is that player should not be
looking for a velocity in a position command however I do not entirely
understand the P/S message formats and wanted confirmation that this was the
case(or not).

thanks

John
-- 
View this message in context: http://www.nabble.com/setPosition-message-format-tf2034617.html#a5597853
Sent from the java-player-users forum at Nabble.com.

Re: [Java-player-users] setPosition message format

[Radu B. R. <ru...@cs...>]

John, that sounds about right, but before I start digging and debugging, would it be 
possible to try it again using HEAD (CVS) Player/Stage and SVN Javaclient ?

There were quite a few bugs that were fixed in the repository since the last official 
version, and it might be that yours was fixed too.

If it still doesn't work, I'd be more than happy to fix it. Now if I could only make the 
day 48h instead of 24h, maybe I could write some documentation for it as well. ;)

Cheers,
Radu.

John Oliver wrote:
> I have been having some trouble with getting setPosition to work on a
> Position2DInterface.  (Player/stage 2.0.1 and Javaclient 2.0.1)
> 
> Taking a quick(well perhaps not that quick) look at the code in player the
> pack function (player_position2d_cmd_pack) it appears to look for both a
> position command and a velocity in the message:
> 
> ...
>   if(xdr_player_pose_t(&xdrs,&msg->pos) != 1)
>     return(-1);
>   if(xdr_player_pose_t(&xdrs,&msg->vel) != 1)
>     return(-1);
>   if(xdr_u_char(&xdrs,&msg->state) != 1)
>     return(-1);
> ...
> 
> while the java version only sends the
> position(Position2DInterface.setPosition), and from what I can tell this is
> leading to my "warning : decoding failed on message to 4:1 with type 2"
> messages.
> 
> Have I made a mistake here or is this something to be patched?  And is it a
> player problem or JavaClient?  My instinct is that player should not be
> looking for a velocity in a position command however I do not entirely
> understand the P/S message formats and wanted confirmation that this was the
> case(or not).
> 
> thanks
> 
> John

-- 
| Radu Bogdan Rusu | http://rbrusu.com/
| http://www9.cs.tum.edu/people/rusu/
| Intelligent Autonomous Systems
| Technische Universitaet Muenchen

Re: [Java-player-users] setPosition message format

[John O. <joh...@gm...>]

Hopefuly I will get around to doing this at some point.  But I think it
should still exist as the Javaclient still only sends the position in the
latest svn version.  However looking at the P/S cvs repo I cant see the pack
functions that exist in the tar'd distrobutions(in the libplayerxdr
directory) the functiontable.c still refers to it but I cant find
it(player_position2d_cmd_pack) anywhere to verify that it is still looking
for a velocity command.

Anyway this may become clear once I get a chance to test it.

thanks again

John
-- 
View this message in context: http://www.nabble.com/setPosition-message-format-tf2034617.html#a5619627
Sent from the java-player-users forum at Nabble.com.

Re: [Java-player-users] setPosition message format

[Radu B. R. <ru...@cs...>]

Hey John,

Just wanted to let you know that you were right, and that I fixed the bugs in SVN. It 
seems that I missed the definition change of player_positionXd_cmd_pos (velocity was 
added), therefore I did not upgrade the Javaclient classes. The bug was affecting all 
three position{1,2,3}d interface.

Thanks for the head-up!

Cheers,
Radu.

John Oliver wrote:
> Hopefuly I will get around to doing this at some point.  But I think it
> should still exist as the Javaclient still only sends the position in the
> latest svn version.  However looking at the P/S cvs repo I cant see the pack
> functions that exist in the tar'd distrobutions(in the libplayerxdr
> directory) the functiontable.c still refers to it but I cant find
> it(player_position2d_cmd_pack) anywhere to verify that it is still looking
> for a velocity command.
> 
> Anyway this may become clear once I get a chance to test it.
> 
> thanks again
> 
> John

-- 
| Radu Bogdan Rusu | http://rbrusu.com/
| http://www9.cs.tum.edu/people/rusu/
| Intelligent Autonomous Systems
| Technische Universitaet Muenchen