|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
cstratak commented 4 years ago | ||
churchyard commented 4 years ago Well this serves as good and running the test should not change the resulting RM in any way, so we are not really bootstrapping it. Other Python packages that are bootstrapped for test only usually have this kind of bcond. See for example https://github.com/hroncok/rpm-list-builder/blob/python37/python37.yaml | ||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
cstratak commented 4 years ago Another approach would be to exclude those tests during the pytest invocation, not sure which one is more beneficial, but I guess since the dependencies are not shipped, maybe it's better to remove them. | ||
churchyard commented 4 years ago I don't think it's better or worse. Let's keep it as it is? (Less work.) | ||
cstratak commented 4 years ago Yes, no need to ponder over that. | ||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
Would a bootstrap bcond make sense here?