From: Radu B. R. <ve...@in...> - 2005-05-25 21:59:11
|
Hello Paul, Thanks for the kind words. Really appreciated :) 1) done. changed the getTimeForDataSent() from protected to public. I was going to inherit them to the actual interfaces, but it seems that I forgot to actually do it. I'm committing the changes to the SF CVS right now. (I still have a few things to fix before releasing 1.6.3.2) 2) also fixed. Thanks a lot for the head-up! You can get the latest updates using the SF CVS. I could really use some feedback on them, since Maxim is still busy. Also, let me know if you'd like 1.6.3.2 to come out, in case the CVS doesn't suit your needs. Cheers, Radu. On Wed, May 25, 2005 at 02:16:17PM -0400, R. Paul Wiegand wrote: > Greetings, > > First, let me say "thanks" to Radu Bogdan Rusu for picking up the > Java client for Player/Stage. It's work, and it's appreciated. > > In the 1.6.x versions of the Java client, there has been some code > restructuring. In that restructuring, several features seem to have > been deprecated or removed. For at least two of these, we've had to > put them back in our local copies in order to avoid making fairly > radical code changes to our existing software. I wanted to write to > the list in order to: 1.) Inform others of these changes so they can > avoid spending time stumbling to find them, and 2.) Try to convince > the developers to return the functionality. > > The two items critical to our work are as follows: > > The getTimeForDataSent() method in PlayerDevice was changed from > public to protected. We actually use this for a variety of purposes > (including debugging P/S timing problems, etc.) ... so we've had to > return this method to the public scope. Is there a reason it is > protected? > > The FiducialInterface class used to have an exposed accessor method > to get the array of what are now called > "PlayerFiducialItems" (formerly "Fiducials"). This method has been > entirely removed. I believe I understand why (it exposes internal > information and weakens the integrity of the class); however, we > actually use this array for our own purposes. > > > There are more protective ways to provide this kind of information, > if data integrity is the problem. If that is not the issue, why have > these features been removed? And can the developers be convinced to > add them back in the general distribution? > > Thanks, > Paul. > > > ============================= > R. Paul Wiegand > > Naval Research Laboratory > 4555 Overlook Ave. SW > Washington, DC > ============================= Yours sincerely, Radu Bogdan Rusu -- | Radu Bogdan 'veedee' Rusu | http://www.rbrusu.com | PhD student/teaching assistant | Robotics Research Group, Technical University of Cluj-Napoca[.ro] |