search

ros-drivers / phidgets_drivers

ROS drivers for various Phidgets devices
Other
45 stars 61 forks source link

Update to libphidgets 20220203 #138

Closed clalancette closed 2 years ago

clalancette commented 2 years ago

Signed-off-by: Chris Lalancette clalancette@openrobotics.org

This should fix #133 , and also update us to the latest libphidget code from upstream.

I've targeted this at the galactic branch, but I actually think we should leave Galactic as it is and make a new, separate branch for Rolling. Then I'll retarget this at that new branch. @mintar what do you think?

clalancette commented 2 years ago

Also +1 on creating a new rolling branch. Or we could create a "humble" branch instead and release that to rolling and humble (same as we did with the galactic branch so far).

What we've been finding is that with the advent of Rolling, it is better to have a separate branch for Rolling, and then a single branch for each released distribution. It is a bit more work in terms of porting things back and forth, but it makes it much easier for everyone to understand what branch corresponds to what.

So my proposal is to create a rolling branch right now, and retarget this PR to it. Once we officially branch off Humble (next week sometime), we can create a humble branch. What do you think?

mintar commented 2 years ago

Yes, sounds good! This is more or less how I'm doing it with the ROS1 branches in my repos as well. Let's do it!

mintar commented 2 years ago

So my proposal is to create a rolling branch right now, and retarget this PR to it. Once we officially branch off Humble (next week sometime), we can create a humble branch.

Now that humble is released and there are Docker images for it, we should do this.

Agreed?

@clalancette : Since you're the ROS2 maintainer of this repo, I'll leave this to you. Unless you want me to do it.

clalancette commented 2 years ago

@clalancette : Since you're the ROS2 maintainer of this repo, I'll leave this to you. Unless you want me to do it.

I'll go ahead and make the branch, though I'm not entirely sure we should keep them synced and using the same matrix. That is, it would be simpler to just have the humble branch test against humble, and the rolling branch against rolling. It requires a bit more work to backport everything, but I think it is much simpler for users to understand.

Regardless, I'll create a humble branch (based off of rolling), then open a PR where we can discuss further.