-
Notifications
You must be signed in to change notification settings - Fork 149
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Take advantage of REP-0140 tag changes #7
Comments
The necessary infrastructure has been rolled out by now. So this should not be blocked anymore. |
With format 2 in the |
Yes, I believe that's right. |
Looking at ros/rosdistro#3460, the only real change to In order to address this issue, I would start with updating the following script to contain the format 2 dependency specifications, right? |
@NikolausDemmel yes, you'd need to make it possible to access the more granular versions of the dependencies, if they're available. That file is probably the right place to start. Thanks. |
What should AFAICT, catkin simple currently only uses catkin_simple/cmake/catkin_simple-extras.cmake.em Lines 211 to 225 in 0e62848
I guess we would want to change that to consider only What about One more thing (that is not really specific to format 2): Shouldn't catkin_simple check if a build_export_depend is a catkin package, or not, and add it to Actually, catkin itself should probably also be updated to consider |
I believe
To fix the above error, catkin would need to consider both |
There is a big difference between the tags in the manifest file and the API of catkin_pkg. The manifest can be format 1 or format 2. But catkin_pkg always provides the exact same API. The data storage within But the API also provide the convenience of the format 1 style dependencies - so In the other direction if a format 1 manifest is being read its generic So whenever you CMake code is only interested in the set of run dependencies (and doesn't care what specific subtype in format 2 they are) it can simply use To summarize |
Sure. That doesn´t mean* it takes advantage of the new tags added with format 2 in all cases though. The following should still be true, as far as I understand it. Could you comment? If we agree, I can open a PR on catkin.
EDIT: *doesnt mean neccessarily |
I don't think this needs any change. |
So are you saying that as a general rule (w/ or w/o catkin simple) that a Let me give an example to make sure I have understood correctly. Let's assume that in a
No, it does not seem to be. In the definition of And I don't think it should be.
No it doesn't. As the
If I export a package with |
ros/catkin#790 addresses this |
This is obviously blocked on the ratification of REP-0140 and its implementation.
The text was updated successfully, but these errors were encountered: