Fixes and Improvements #43

tonynajjar · 2024-03-05T09:35:37Z

Fixes:

Fixes for BT upgrade
fix polygons not being cleared in ComputeCoveragePathAction

Improvements:

Make visualization topics Transient Local
for demo, cancel goal when exiting with CTRL+C

SteveMacenski

@tonynajjar just add in the readme the transient local-ness of those topics

opennav_coverage/src/visualizer.cpp

README.md

SteveMacenski · 2024-03-05T20:47:57Z

Linting failed an an odd error basically complaining about conflicts with every BT.CPPv4 node?

opennav_coverage/CMakeLists.txt

tonynajjar · 2024-03-05T21:41:03Z

Linting failed an an odd error basically complaining about conflicts with every BT.CPPv4 node?

I found a couple of build issue that I fixed. The logs were super verbose though, maybe there is something else, let's see

tonynajjar · 2024-03-05T21:44:05Z

It's still failing, looking at the logs, they contain both /opt/ros/iron/include/behaviortree_cpp_v3 and /opt/ros/iron/include/behaviortree_cpp, maybe that's a problem? Locally it builds...

On rolling there is only the newer version:

root@tony-xmg-22:/opt/overlay_ws# ls /opt/ros/rolling/include/ | grep behavior
behaviortree_cpp

This reverts commit 3669e7d.

SteveMacenski · 2024-03-05T22:50:02Z

That is odd - this CI shouldn't have any caching in it (@emersonknapp any caching happen in the ros action ci?) for it to know about BT.CPPv4 if its gone here. Does this cause you issues locally?

tonynajjar · 2024-03-06T10:48:10Z

Does this cause you issues locally?

Locally on rolling no, I didn't test on iron though, I'm on humble and the devcontainer is only for rolling I believe - setting up iron will be some effort

tonynajjar · 2024-03-06T13:10:20Z

That is odd - this CI shouldn't have any caching in it (@emersonknapp any caching happen in the ros action ci?) for it to know about BT.CPPv4 if its gone here

@SteveMacenski I'm guessing it knows about BT.CPPv4 because it's using iron. I tried using rolling in this commit but that fails as well because it can't find nav2_util -> are there no binaries for nav2 on rolling?

SteveMacenski · 2024-03-06T16:17:34Z

There is no Nav2 binaries for rolling, but you could add it in the https://github.com/open-navigation/opennav_coverage/blob/main/.github/deps.repos for use in CI for testing. I'm not sure if you can isolate it to only nav2_utils and its necessary dependencies, but I suppose we could change the Action's colcon build to include packages-up-to for the opennav coverage packages, which would include only the parts of Nav2 strictly required. It would though need to build the Nav2 BT libraries which is non-trivial.

SteveMacenski · 2024-03-06T16:45:15Z

-_- action ci tools don't call rosdep, I literally can't believe that

tonynajjar · 2024-03-06T16:50:40Z

-_- action ci tools don't call rosdep, I literally can't believe that

hmm I think it does

but for some reason it can't find "standard" packages: https://github.com/open-navigation/opennav_coverage/actions/runs/8175554856/job/22352920043?pr=43

tonynajjar · 2024-03-06T17:25:23Z

I'm wondering if https://discourse.ros.org/t/preparing-ros-2-rolling-for-the-transition-to-ubuntu-24-04/35673/8 could have anything to do with it. It still says Jammy in the CI though No definition of [geometry_msgs] for OS version [jammy]

tonynajjar · 2024-03-06T17:27:54Z

I'm wondering if https://discourse.ros.org/t/preparing-ros-2-rolling-for-the-transition-to-ubuntu-24-04/35673/8 could have anything to do with it. It still says Jammy in the CI though No definition of [geometry_msgs] for OS version [jammy]

Now I have a stronger suspicion:

root@tony-xmg-22:/opt/overlay_ws# rosdep resolve geometry_msgs
#apt
ros-rolling-geometry-msgs
root@tony-xmg-22:/opt/overlay_ws# rosdep update
reading in sources list data from /etc/ros/rosdep/sources.list.d
Warning: running 'rosdep update' as root is not recommended.
  You should run 'sudo rosdep fix-permissions' and invoke 'rosdep update' again without sudo.
Hit https://raw.githubusercontent.com/ros/rosdistro/master/rosdep/osx-homebrew.yaml
Hit https://raw.githubusercontent.com/ros/rosdistro/master/rosdep/base.yaml
Hit https://raw.githubusercontent.com/ros/rosdistro/master/rosdep/python.yaml
Hit https://raw.githubusercontent.com/ros/rosdistro/master/rosdep/ruby.yaml
Hit https://raw.githubusercontent.com/ros/rosdistro/master/releases/fuerte.yaml
Query rosdistro index https://raw.githubusercontent.com/ros/rosdistro/master/index-v4.yaml
Skip end-of-life distro "ardent"
Skip end-of-life distro "bouncy"
Skip end-of-life distro "crystal"
Skip end-of-life distro "dashing"
Skip end-of-life distro "eloquent"
Skip end-of-life distro "foxy"
Skip end-of-life distro "galactic"
Skip end-of-life distro "groovy"
Add distro "humble"
Skip end-of-life distro "hydro"
Skip end-of-life distro "indigo"
Add distro "iron"
Skip end-of-life distro "jade"
Skip end-of-life distro "kinetic"
Skip end-of-life distro "lunar"
Skip end-of-life distro "melodic"
Add distro "noetic"
Add distro "rolling"
updated cache in /root/.ros/rosdep/sources.cache
root@tony-xmg-22:/opt/overlay_ws# rosdep resolve geometry_msgs

ERROR: No definition of [geometry_msgs] for OS version [jammy]

No definition of [geometry_msgs] for OS version [jammy]
	rosdep key : geometry_msgs
	OS name    : ubuntu
	OS version : jammy
	Data:
_is_ros: true
		debian:
		  bookworm:
		    apt:
		      packages:
		      - ros-rolling-geometry-msgs
		osx:
		  homebrew:
		    packages:
		    - ros/rolling/common_interfaces
		rhel:
		  '9':
		    dnf:
		      packages:
		      - ros-rolling-geometry-msgs
		ubuntu:
		  noble:
		    apt:
		      packages:
		      - ros-rolling-geometry-msgs

What a timing 😄

SteveMacenski · 2024-03-06T19:01:01Z

Oh I see that too. So I guess transient during the Rolling migration that's underway, I see the flood of build farm emails with their work to 24.04. What's the path from here? I think this was just a test to see if it would resolve on rolling since Iron was problematic. Certainly we could move the CI to rolling too, I could roll with that 😉

SteveMacenski · 2024-03-06T23:20:11Z

By the way, this is not just a CI problem, I see this locally as well after doing an apt update + upgrade using the BTv4 branch. I actually see the 'all the BT nodes complaining' in opennav_docking as well now after the update which is interesting.

SteveMacenski · 2024-03-07T21:09:41Z

Experimenting now on the odd BT compiling issues. Did you happen to make any progress on it? I'm starting with purging v3 from my computer so its only v4. I suspect that has something to do with it.

tonynajjar · 2024-03-07T21:14:47Z

Mmm no progress as I didn't setup Iron. On my setup on rolling there wass only BTv4 so I couldn't reproduce locally. You were able to reproduce on iron right?

SteveMacenski · 2024-03-07T21:28:56Z

I don't recall specifically what my setup was when I saw it, I was in the middle of doing something else so I just removed opennav_coverage and went back to what I was trying to do at the time when I ran into it. Its possible I had nav2 main built on iron base (which could have had btv4 in it) or some other mixed policy. I was poking around alot of projects this week so I can't recall which specifically it was I found it with respect to.

What I know I did right now:

Rolling, with BTv4 installed only (apt removed v3)
Built in a workspace with Nav2 clean with no prior caches

Doesn't happen for one of the projects, I adapted that here as well now too in a separate branch with just the v4 changes and that works too. I'll open a separate PR in a moment with just the v4 updates to see what CI says about that and that I have something working locally.

I'm guessing it has to do with having the v3 and v4 installed next to each other

SteveMacenski · 2024-03-07T21:33:04Z

Ah, in CI

executing command [apt-get install -y ros-iron-nav2-behavior-tree]

That's going to pull in BT.CPPv3. Then we're using v4 that would explain it (in my PR)

tonynajjar · 2024-03-07T21:38:12Z

Yes and BTv3 is expected in Iron so the simplest solution I see is to port the CI to rolling when rosdep works again... Do you see another option?

SteveMacenski · 2024-03-07T21:53:02Z

Note that you missed adding BTCPP_format="4" to the BT XML + the inline test XML string which causes a warning on execution. Please add that and I'll close my PR (only difference)

I see is to port the CI to rolling when rosdep works again

I agree. If I'm reading the docs for action-ros-ci correctly, we can use colcon-extra-args: --packages-up-to <packages in repo> to automatically only build what is needed from Nav2. Else, we can specify exactly which packages to build with package-name:, but we'll need to figure out exactly which packages we need to build manually. Which, from a quick look, seems like ~46 packages which would be a pain in the ass to list.

https://github.com/ros-tooling/action-ros-ci/blob/634e9ac5fb970763bd6207e31e1875d0081be71e/action.yml#L31-L34

I also filed ros-tooling/setup-ros-docker#68 which is necessary to upgrade the CI to use 24.04 (in case 22.04 is totally messed up permanently, which would be really irritating). I asked for clarity in https://discourse.ros.org/t/preparing-ros-2-rolling-for-the-transition-to-ubuntu-24-04/35673/9?u=smac. I suppose either way we need to update CI to 24.04, so I suppose if that comes, we should just do that anyway.

I think that's the path forward and the contingencies, do you agree? I didn't want to make this your problem so I made sure to look at this today.

tonynajjar · 2024-03-07T22:25:14Z

Note that you missed adding BTCPP_format="4" to the BT XML + the inline test XML string which causes a warning on execution. Please add that and I'll close my PR (only difference)

Done

https://github.com/ros-tooling/action-ros-ci/blob/634e9ac5fb970763bd6207e31e1875d0081be71e/action.yml#L31-L34

ah nice find, there's nothing about it in the readme

I think that's the path forward and the contingencies, do you agree?

yep sounds like a plan. Let's see what they'll answer in the discourse

I'm looking forward to this moving in navigation2 (if that's gonna happen) to manage it more easily and avoid the double work.

SteveMacenski · 2024-03-07T22:43:38Z

Yeah the plan is to move it to Nav2 once the F2C updated features are in F2C & updated to be used here. I still haven't decided if I will rename these to use nav2_ prefix or keep opennav_, but that's pedantic

This reverts commit aa2cf20.

This reverts commit 668abc4.

tonynajjar · 2024-03-14T10:58:24Z

I tried again since rolling sync is out but still same issue with rosdep not finding some packages. Unfortunately I don't have the time currently to look further into it, could you maybe ask for a follow up on the issue?

SteveMacenski · 2024-03-14T16:46:31Z

It still failing everywhere, I wouldn't worry about it yet. I still see missing rosdep in Nav2 and tickets on discourse, I don't think anything changed

Fix BT.CPP upgrade

c81ba02

tonynajjar mentioned this pull request Mar 5, 2024

Coverage Demo fails #41

Open

Bug fixes

2688106

tonynajjar changed the title ~~Fix BT.CPP upgrade~~ Bug Fixes Mar 5, 2024

tonynajjar changed the title ~~Bug Fixes~~ Fixes and Improvements Mar 5, 2024

SteveMacenski approved these changes Mar 5, 2024

View reviewed changes

opennav_coverage/src/visualizer.cpp Outdated Show resolved Hide resolved

SteveMacenski and others added 2 commits March 5, 2024 10:23

Update opennav_coverage/src/visualizer.cpp

0841d1d

add readme

4cbdcc7

tonynajjar commented Mar 5, 2024

View reviewed changes

README.md Show resolved Hide resolved

SteveMacenski approved these changes Mar 5, 2024

View reviewed changes

tonynajjar added 2 commits March 5, 2024 21:03

fix flake8

06497c7

building fixes

62d628a

tonynajjar commented Mar 5, 2024

View reviewed changes

opennav_coverage/CMakeLists.txt Show resolved Hide resolved

tonynajjar added 2 commits March 5, 2024 21:52

testing with rolling

3669e7d

Revert "testing with rolling"

a3948ec

This reverts commit 3669e7d.

tonynajjar mentioned this pull request Mar 6, 2024

Test PR for CI [Do not merge] #44

Closed

test with rolling

2d48105

tonynajjar added 3 commits March 7, 2024 22:01

Merge remote-tracking branch 'origin/main'

b85696b

Add BTCPP_format="4" to trees

99eadcf

Use tree_->haltTree();

aa2cf20

tonynajjar added 2 commits March 14, 2024 08:52

Revert "Use tree_->haltTree();"

668abc4

This reverts commit aa2cf20.

Revert "Revert "Use tree_->haltTree();""

1934a78

This reverts commit 668abc4.

aosmw mentioned this pull request Apr 29, 2024

Feature/mw/fields2cover2 #53

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fixes and Improvements #43

Fixes and Improvements #43

tonynajjar commented Mar 5, 2024 •

edited

SteveMacenski left a comment •

edited

SteveMacenski commented Mar 5, 2024

tonynajjar commented Mar 5, 2024

tonynajjar commented Mar 5, 2024 •

edited

SteveMacenski commented Mar 5, 2024

tonynajjar commented Mar 6, 2024 •

edited

tonynajjar commented Mar 6, 2024

SteveMacenski commented Mar 6, 2024

SteveMacenski commented Mar 6, 2024

tonynajjar commented Mar 6, 2024 •

edited

tonynajjar commented Mar 6, 2024

tonynajjar commented Mar 6, 2024 •

edited

SteveMacenski commented Mar 6, 2024

SteveMacenski commented Mar 6, 2024 •

edited

SteveMacenski commented Mar 7, 2024

tonynajjar commented Mar 7, 2024

SteveMacenski commented Mar 7, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

tonynajjar commented Mar 7, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

tonynajjar commented Mar 7, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

tonynajjar commented Mar 14, 2024 •

edited

SteveMacenski commented Mar 14, 2024 •

edited

Fixes and Improvements #43

Are you sure you want to change the base?

Fixes and Improvements #43

Conversation

tonynajjar commented Mar 5, 2024 • edited

SteveMacenski left a comment • edited

Choose a reason for hiding this comment

SteveMacenski commented Mar 5, 2024

tonynajjar commented Mar 5, 2024

tonynajjar commented Mar 5, 2024 • edited

SteveMacenski commented Mar 5, 2024

tonynajjar commented Mar 6, 2024 • edited

tonynajjar commented Mar 6, 2024

SteveMacenski commented Mar 6, 2024

SteveMacenski commented Mar 6, 2024

tonynajjar commented Mar 6, 2024 • edited

tonynajjar commented Mar 6, 2024

tonynajjar commented Mar 6, 2024 • edited

SteveMacenski commented Mar 6, 2024

SteveMacenski commented Mar 6, 2024 • edited

SteveMacenski commented Mar 7, 2024

tonynajjar commented Mar 7, 2024

SteveMacenski commented Mar 7, 2024 • edited

SteveMacenski commented Mar 7, 2024 • edited

tonynajjar commented Mar 7, 2024 • edited

SteveMacenski commented Mar 7, 2024 • edited

tonynajjar commented Mar 7, 2024 • edited

SteveMacenski commented Mar 7, 2024 • edited

tonynajjar commented Mar 14, 2024 • edited

SteveMacenski commented Mar 14, 2024 • edited

tonynajjar commented Mar 5, 2024 •

edited

SteveMacenski left a comment •

edited

tonynajjar commented Mar 5, 2024 •

edited

tonynajjar commented Mar 6, 2024 •

edited

tonynajjar commented Mar 6, 2024 •

edited

tonynajjar commented Mar 6, 2024 •

edited

SteveMacenski commented Mar 6, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

tonynajjar commented Mar 7, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

tonynajjar commented Mar 7, 2024 •

edited

SteveMacenski commented Mar 7, 2024 •

edited

tonynajjar commented Mar 14, 2024 •

edited

SteveMacenski commented Mar 14, 2024 •

edited