[Rpm-maint] [rpm-software-management/rpm] Fix regression of mktree not always rebuilding (PR #2726)

[Michal Domonkos]

Commit d4e808ebda3ec1d85fcdb176eae31901b74c6e91 added cmake caching for mktree so that we don't rebuild the tree on each "make check" invocation.  However, this means we can now easily miss some test dependencies being updated since we don't depend on everything, e.g. the data/ dir.

To fix this, instead of adding each and every file and directory as a dependency of the mktree.output command, just rebuild the tree every time.  The base image (made with DNF) is already cached anyway, and the added overhead is tiny and outweighs the (potential) engineering hours lost due to this.

The funny thing is, this was already fixed once in 2019: fc05045008d84429bf51154642e18fda70d3f256

Sometimes you've just gotta fix stuff twice before it sticks!

Fixes: 2725
You can view, comment on, or merge this pull request online at:

  https://github.com/rpm-software-management/rpm/pull/2726

-- Commit Summary --

  * Fix regression of mktree not always rebuilding

-- File Changes --

    M tests/CMakeLists.txt (13)
    M tests/mktree.fedora (3)

-- Patch Links --

https://github.com/rpm-software-management/rpm/pull/2726.patch
https://github.com/rpm-software-management/rpm/pull/2726.diff

-- 
Reply to this email directly or view it on GitHub:
https://github.com/rpm-software-management/rpm/pull/2726
You are receiving this because you are subscribed to this thread.

Message ID: <rpm-software-management/rpm/pull/2726 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rpm.org/pipermail/rpm-maint/attachments/20231017/f50d2191/attachment.html>