Skip to content

Conversation

@guitargeek
Copy link
Contributor

Right now, the stltypes test only works by accident if the datatypes dictionary is loaded too, because that's where one can find the gbl.N variable that is used by the test. To avoid this dependency, define the N variable on the Python side instead.

Mimics the following ROOT commit:
root-project/root@1b41183

Right now, the stltypes test only works by accident if the datatypes
dictionary is loaded too, because that's where one can find the `gbl.N`
variable that is used by the test. To avoid this dependency, define the
`N` variable on the Python side instead.
@guitargeek
Copy link
Contributor Author

@wlav can this be merged please? It should not be controversial, and it's a bit awkward for the build system if the stltypes test is using something that is not defined in stltypes.h. It would be better if the tests don't "leak" into each other.

@guitargeek
Copy link
Contributor Author

Hi @wlav! Do you see any problem with this PR?

@wlav
Copy link
Owner

wlav commented Sep 21, 2025

In essence, yes: way in the beginning, it wasn't possible to redeclare a global variable such as N in Cling. Thus, I'd rather it be kept the same name (just "N") and pulled from Cling, so that if redeclaration fails somehow, it'll show up in the tests.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants