planning_scene updates: expose success state to caller #297
Conversation
| @@ -1185,14 +1187,16 @@ void planning_scene::PlanningScene::setPlanningSceneDiffMsg(const moveit_msgs::P | |||
|
|
|||
| // process collision object updates | |||
| for (std::size_t i = 0 ; i < scene_msg.world.collision_objects.size() ; ++i) | |||
| processCollisionObjectMsg(scene_msg.world.collision_objects[i]); | |||
| ret &= processCollisionObjectMsg(scene_msg.world.collision_objects[i]); | |||
There was a problem hiding this comment.
I'm wondering if this is out of style for MoveIt!'s code base - we have almost no &= in the code base (except in the rviz plugin, twice) and I found it hard to understand at first. Wouldn't += accomplish the same thing but be easier to quickly understand? Its probably just me being ignorant but I'm not used to seeing it
This is required to get the information back for the ApplyPlanningSceneService.
v4hn
force-pushed
the
apply-planning-scene
branch
from
02c45fc
to
0322f52
Compare
|
Looks like this is already cherry-picked into |
|
@v4hn Yeah I think so,,,although I know it's cumbersome. |
|
#301 (comment)
for reference. Let's continue this over there.
|
|
I still think if you just had a PR merged and you want to reflect it in a newer branch, its no big deal to cherry-pick and commit it without a commit. I say its fine because its what I've seen others do here and it makes sense to me. I created those PRs for syncing in the past day because they were old and messy and potential error-prone. I think we should just do a better job of always keeping the branches in sync |
This is required to get the information back for the
ApplyPlanningSceneService.
This is required for moveit/moveit_ros#686
As it turns out, according to abi-compliance-checker these changes
are also ABI compliant... I didn't expect that.