Closed furushchev closed 9 years ago
Can one of the STRANDS admins verify this patch? Post a comment containing "ok to test" to enable Jenkins builds for this pull request.
The handling of machine tags is indeed nasty in ROS (or we didn't find the correct solution yet). Yes, you're right, in general, we should probably not have the machine tags in the usual launch files at all. @hawesie?
ok to test
I don't know which problem was @furushchev having, if he was using another launch file with machine tags it could have failed because he defined the same machine diferently twice, e.g, defining localhost with different users. This being said I like this solution, however, I don't think that removing machine tags in the launch files is viable, instead we can have strands launch files and general launch files without machine tags, does this sound logical? @marc-hanheide @hawesie
@Jailander yes, that's what I kind of meant... We could have a "wrapper" launch file for our purposes, but there should also be a "vanilla" one.
Ok just making sure ;-)
So should I make a wrapper launch file and send another pull request?
Hi, don't worry I'll do that
I have accepted this for now. Obviously the default should then in the future be that machine tags are not being used (use_machine=false
by default) and only in our own setting, a wrapper should set it to true... I still don't like this whole machine tag business. I wonder if we should initiate a discussion on this with the ROS people or on answers.ros.org to find the "right" way of doing this.
I'd like to choose to use machine tag or not. If this is included by another launch file and it also has machine tag ( 2 machine tag ), launch fails.