You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The other thing worth documenting is that the output from %generate_buildrequires needs to provide stable output eventually, after doing the N installation iteration steps, no matter N value (may be zero for a cached buildroot).
The #1352 tries to achieve a good goal; WRT %generate_buildrequires -> it seems we could provide guidance that it is not a good idea to modify %prep output (source directory) by %generate_buildrequires. Or expect that such changes are not reverted before the %build starts.
The text was updated successfully, but these errors were encountered:
One of the formal facts of the current implementation/optimization: %prep output is never affected by the processing of %generate_buildrequires (thanks @kdudka for the point).
This is a pretty complicated feature, and we should document peculiarities, like the one in https://rpm-software-management.github.io/mock/Release-Notes-5.3
The other thing worth documenting is that the output from
%generate_buildrequires
needs to provide stable output eventually, after doing the N installation iteration steps, no matter N value (may be zero for a cached buildroot).The #1352 tries to achieve a good goal; WRT %generate_buildrequires -> it seems we could provide guidance that it is not a good idea to modify %prep output (source directory) by %generate_buildrequires. Or expect that such changes are not reverted before the %build starts.
The text was updated successfully, but these errors were encountered: