Adrian Macneil
Adrian Macneil
Yes, I think both would be good (storing the schema, and also storing the ros distro).
cc @jhurliman @defunctzombie @james-rms
@MichaelOrlov I noticed your comment in https://github.com/ros2/rosbag2/issues/1143#issuecomment-1299570359: > It doesn't make sense to store ROS_DISTRO in addition to the DB schema versioning. At least I am not envision any cases...
Agreed, I think we can add ROS_DISTRO to Humble and Galactic without breaking backwards compatibility, it is it strictly an addition to the db3 file, and shouldn't require changing any...
@clalancette good point, only identifying Humble db3 files would be sufficient. We can assume the pre-Humble Marker message unless we explicitly detect `ROS_DISTRO=Humble` (technically it will not work for "early...
We took the ideas from this thread, https://github.com/ros2/rosbag2/issues/819, https://github.com/ros2/rosbag2/issues/855, and several others and incorporated them into the design of [MCAP](https://mcap.dev/), a highly efficient, self-contained recording format for ROS 2 and...
This is now supported in [MCAP](https://mcap.dev/) and the [rosbag2_storage_mcap](https://github.com/ros-tooling/rosbag2_storage_mcap) plugin.
So based on my read of this thread, are the following statements both true? - ROS 2 supports using _both_ `.msg` or `.idl` (CDR) as the "source of truth" message...
I see, I missed that distinction. So essentially my proclamation (b) is accurate - ROS 2 architecturally (at least at rmw layer) speaks full IDL, and `.msg` is only a...
+1 customizable nginx config file