dont print warning about dropped first point #366
Conversation
…iour The warning about dropped points isn's useful, if the first point just contains the expected start point. Dropping it is expected behaviour, thus degrading this to a debug output message. A subsequent feature could be to check the current position against the parametrized tolerance of this point. Closes ros-controls#291
mathias-luedtke
left a comment
mathias-luedtke
left a comment
There was a problem hiding this comment.
Basically, I am okay with this patch (as discussed offline..).
However, I am a little bit worried about the skipping of the first point (which is unrelated to this PR),
Does this really prevent from showing the message in your tests?
| return Trajectory(); | ||
| } | ||
| else if ( // If the first point is at time zero and no start time is set in the header, skip it silently | ||
| msg.points.begin()->time_from_start == ros::Duration(0.) && |
There was a problem hiding this comment.
Why not .isZero()?
There was a problem hiding this comment.
Please chnage this, there is no need to call the constructor and the comparison operator instead of using the function that was implemented for this purpose.
| } | ||
| else if ( // If the first point is at time zero and no start time is set in the header, skip it silently | ||
| msg.points.begin()->time_from_start == ros::Duration(0.) && | ||
| msg.header.stamp == ros::Time(0) && |
joint_trajectory_controller/include/joint_trajectory_controller/init_joint_trajectory.h
Show resolved
Hide resolved
|
I hate that message, glad its going away! |
|
I am waiting for that pr to merge also #367 |
|
I went ahead and just updated this PR to use |
|
Thank you @ipa-mdl for pushing it directly. |
5 participants
Since this is expected behaviour, if no start time is given in header of the message.
The warning about dropped points isn's useful, if the first point just contains
the expected start point. Dropping it is expected behaviour, thus degrading this
to a debug output message.
A subsequent feature could be to check the current position against the
parametrized tolerance of this point.
Closes #291