Eloquent patch release 1

[ivanpauno]

Backports:

• Fix frontend topic remapping #111
• check for shutdown while waiting for a service response to avoid hang during shutdown #104
• Fix misleading deprecated warning when using launch arguments #106 (this breaks API, but of something completly unused and that was causing failures)
• Use imperative mood in constructor docstrings. #103 (just to avoid conflicts with other backported commits)
• Maintain order of parameters regarding name and from tag #99
• Fix push-ros-namespace in xml/yaml launch files #100
• Pass the node-name attribute through the substitution parser #101
• Add pid to launch_ros node name as suffix #98

Other stuff that could be included:

• Allow separate launch composition #77 (I didn't include it because it's an enhancement, not a fix).

[mjcarroll]

DCO is unhappy, at a minimum.

[ivanpauno]

DCO is unhappy, at a minimum.

The commit where the checker is failing is signed. It seems to be a false-positive, but I can add my signature if that makes the checker happier.

[ivanpauno]

DCO is unhappy, at a minimum.

The commit where the checker is failing is signed. It seems to be a false-positive, but I can add my signature if that makes the checker happier.

Sorry, there is actually one commit failing because of a missing DCO, and some others failing because of strange reasons. I can add my signature in all of them to avoid the checker to fail, but I don't know if that's correct. Also, if the commit is already in master, IDK if I should add an extra signature.

[dirk-thomas]

If you co-author a commit (e.g. a cherry pick or merge) you need to sign that commit too to pass DCO.

[ivanpauno]

If you co-author a commit (e.g. a cherry pick or merge) you need to sign that commit too to pass DCO.

That wasn't exactly the failure that the DCO checker was showing, but done.

[mjcarroll]

Can you fix that linter error?

[ivanpauno]

Can you fix that linter error?

Are you talking about the errors here ros2/launch#370 (comment)?

Does are from launch repository, and not from launch_ros.
They are there just because I didn't backport ros2/launch#362.
We usually don't backport those commits, but I can add it to the list if you want.

[ivanpauno]

CI is the same as in: ros2/launch#370 (comment)

[wjwwood]

I'm concerned about how #106 breaks behavior and API, but I think that probably no one is using it (or that probably no one would be affected by it) and it would make more sense to fix it. But we've been wrong about that in the past.

So, I'd tentatively vote to backport it and then make an announcement with the patch release letting people know. But I'll leave that up to @mjcarroll to decide how to proceed.

[wjwwood]

Aside from my concerns with #106 being backported, lgtm.

[mjcarroll]

I'll add it to the announcement. If it causes issues, we can revert, but I also believe it to be unlikely.

[ivanpauno]

@wjwwood I agree with your concerns. I proposed it just because the bug is to ugly, and there's no way of fixing it without breaking API.