LineStripMarker: ignore invalid points

[rhaschke]

Fix #1439.

[rhaschke]

@wjwwood: What is your opinion about including such validFloats() checks in rviz?
Obviously, you didn't include them in the past - maybe for performance reasons, expecting the data publishers to provide proper messages.
On the other hand, I would argue, that such wrong messages shouldn't crash rviz.

[simonschmeisser]

In my opinion it should not be possible to crash rviz with external data! Ever! Well at least not easily ;)

[rhaschke]

@simonschmeisser, thanks for reviewing.

[wjwwood]

Sorry for the late reply.

Obviously, you didn't include them in the past - maybe for performance reasons, expecting the data publishers to provide proper messages.

Though I can imagine it might be a performance issue, the safety is probably be worthwhile. I didn't write this code originally though, so I cannot say one way or the other.