-
Stage 3 breaks the localization_origin for model_position. It seems to just be a simple typo mistake. This patch corrects the typo and updates the documentation.
2009-03-18 23:26:36 UTC in The Player Project
-
This patch adds support for Player to see the simulated camera from Stage. It also fixes a bug where the stage camera was using the parent's frame of reference rather than the camera's.
Patch is against SVN trunk.
2009-02-06 07:12:23 UTC in The Player Project
-
This patch fixes the bug in PlayerV which causes 100% CPU usage due to a tight loop (nanosleep of 0 seconds, 0 nanoseconds).
Now the sleep time is fixed at 50ms, for a 20Hz GUI update rate. (Bug referenced on mailing list "Some issues upgrading to 2.1");
This does mean that if you specify a rate of > 20Hz, you won't get it. Also it means that if you specify a rate of < 20Hz there might be a...
2008-10-12 17:24:27 UTC in The Player Project
-
playerc_client_read_peek.diff should be applied.
2008-10-07 16:29:03 UTC in The Player Project
-
Actually now that I look at this, why is this function implemented the way it is? I think this could be better implemented using the client_peek call, which already does a blocking poll, rather than a test/sleep loop. This also would make it trivial to implement a playerc_client_read_timeout. I'll see about changing this and post back. Please let me know if there is a reason playerc_client_read...
2008-10-03 20:55:39 UTC in The Player Project
-
The documentation states that playerc_client_read will return NULL on error, however sometimes it returns -1. This is due to someone trying to add a check so it will return NULL when the socket is invalid (client->sock < 0) however it was added to the wrong if statement. This patch fixes it.
This also fixes libplayerc++ as the error checking is done against libplayerc. So programs like...
2008-10-03 20:19:12 UTC in The Player Project
-
Currently the garminnmea driver will output "short time field; ignoring" everytime it gets a message with a short time field, however some GPS devices, such as the NovaTel Propak's, will output a messages at 20Hz regardless of if they have a signal or not. When indoors this causes the "short time field; ignoring" message to be printed 20 times a second...
This patch creates a static boolean...
2008-10-03 19:09:17 UTC in The Player Project
-
The latest patch adds an include for , which I don't have (2-1-patches).
Also it changes SickLMS200_Register to sicklms200_Register.
I imagine both of these come from 2.2 ?
Changing those two things worked and the driver is usable (Tested on Mac OS X/LMS291).
Thanks for your work on this driver. My SICK's flash memory thanks you, and the high_avail is great too.
(It seems I...
2008-10-03 19:01:40 UTC in The Player Project
-
Player 2.1.0 Release does not build on Mac OS X. This is due to new timeout code in messages.cc that uses the Linux/POSIX function clock_gettime. Mac does not support clock_gettime.
I've taken a replacement code for clock_gettime from a patch to the Orocos Project and put it into Player's replace system. I've emailed the patch author to confirm that the function is under the GPL, but I think...
2008-06-14 22:45:10 UTC in The Player Project
-
It seems the laser was never updated for the 2.1 API. This fixes it, and also sets the max_range to 8, and sets the value for all of them to max_range.
2008-05-16 18:37:25 UTC in The Player Project
