From: Radu B. R. <ru...@cs...> - 2007-11-14 21:16:58
|
Dear Leo, Yup. I tend to think that almost all interfaces are affected by the same bug. Decoding XDR packages is pretty straightforward, and the only thing that you need to do is look at the interface definition in Player. Then you see that for a float you need 4 bytes, an int 4 bytes... etc.. I used http://www.faqs.org/rfcs/rfc1014.html for Javaclient 2.0. Now, the problem is... one should use Player CVS in order to modify Javaclient SVN, otherwise it all becomes a big mess. I am happy to know that you're interested. Have a look and see whether you can modify the interfaces which fail for you (you can look at the latest commits, like the one from yesterday on SonarInterface which fixed a bug thanks to Alejandro Pustowka's report), and then contact me later privately and I'll set you up with access rights on the repository. Cheers, Radu. Leo Nomdedeu wrote: > Dear Radu > > As I said, I'm willing to help, and, if needed, mantain the javaclient. > Right now I face some problems working with jc, as, for example, the planner > and the localize interfaces, for example, have both the same "bug", that is, > player sends an extra integer, and as it is not contemplated in the XDR > decoding (in the handle response routine), the PlayerPose information you > get is shifted one 4 bytes field, and so, you get Px = bugos data, Py = > realPx, Pa=RealPy, and RealPa is lost in the space ;) > > The same thing, but just at the inverse order, happends in the > Graphics2dInterface. When coding, it adds twice the point_count, and so all > the data is shifted when Player receives it (and obviously, the polyline is > far from being what you thought it should be. > > As I said, I have no source of information about the correcness of the XDR > messages, and I can only try to intercept with ethereal, or digg into the C > client, and, as you will know, it's not the very straigth forward way to > work ... ¿could you be so kind to provide me some sort of documentation > about the XDR messages? > > Fortunatelly, I need jc for my PhD, so, I will be very active, I think ;) > > Bests > Leo > > 2007/11/14, Radu Bogdan Rusu <ru...@cs...>: >> >> Dear Leo, >> >> Thank you for your interest in Javaclient and well, the whole Player >> project ! =) >> >> You are right, the Javaclient project is very inactive at the moment, and >> a few interfaces need to have their XDR >> decoding routines fixed. It's a matter of hours (patching + testing), but >> unfortunately, my time is very limited these >> days, and, as I am not using Javaclient in my research work for now, I am >> unable to make those corrections myself at >> this point. Most of my efforts have gone towards the Player project itself >> as a whole, and well, my PhD thesis. =) >> >> Therefore, any help would be highly appreciated, not just for me or for >> the project, but for the entire community who >> uses it and would like to use it in the future. >> >> Let me know if you want to help, and I can guide you through some of its >> internals (not that big of a deal anyway as >> I've tried to make everything pretty straightforward). >> >> Cheers, >> Radu. >> >> Leo Nomdedeu wrote: >>> Hi all >>> >>> My name is Leo Nomdedeu, from Spain. I'm new to this list. >>> I'm working on an european project where we decided to use P/S some time >>> ago. Now I face the need to use java-client to control a set of mobile >>> robots using P/S. I'm havin some trubles with some interfaces, e.g. >> Planner >>> giving wrong x,y data (a,x,y) instead of (x,y,a), graphics not drawing >>> (decoding failed in stage), position turning in place instead of going >> to >>> the position designed, etc. >>> I wish to aid as much as possible with this player-client, and I have >>> allready made some simple modifications from my own but I cannot be sure >> if >>> this modifications are good or not due to I do not have access (aka, I >> have >>> not found) to any information on the current XDR messages Player is >> waiting >>> for. ¿Is there any place I can look at to get such info? >>> >>> An other question I have. ¿Is the Java-Client project alive? I mean, I >> have >>> not seen much activity in the mailing lists from 2006 year to now ... >>> >>> I wish to continue my development in Java as I have several years of >>> expertise with this language, so I'm ready to work in the java-client >>> project if needed. It's better than having to work in C ... >>> >>> Wainting for your answers... >>> Leo >> -- >> | Radu Bogdan Rusu | http://rbrusu.com/ >> | http://www9.cs.tum.edu/people/rusu/ >> | Intelligent Autonomous Systems >> | Technische Universitaet Muenchen >> > > > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Splunk Inc. > Still grepping through log files to find problems? Stop. > Now Search log events and configuration files using AJAX and a browser. > Download your FREE copy of Splunk now >> http://get.splunk.com/ > > > ------------------------------------------------------------------------ > > _______________________________________________ > Java-player-developers mailing list > Jav...@li... > https://lists.sourceforge.net/lists/listinfo/java-player-developers -- | Radu Bogdan Rusu | http://rbrusu.com/ | http://www9.cs.tum.edu/people/rusu/ | Intelligent Autonomous Systems | Technische Universitaet Muenchen |