Pr #70 introduced an optimization that results in small files (smaller than the pathname of the delta source) being copied directly to the delta, skipping the delta process on those files. This helps speed things up, especially with bootc images.
However, this also broke codecov a bit. The new optimization results in old codepaths not being executed. This then makes codecov sad, as it thinks we're not testing as much code as we used to be.
This is likely due to the tests using small files, hitting the optimization, and skipping the codepaths they used to traverse. Thus, we need to update the tests to ensure that the old codepaths are hit again as they used to be. (I think this should largely be solved by increasing test file size, although I haven't totally confirmed myself yet.)
TLDR: tests need to be updated to return to traversing the codepaths they used to cover, likely by increasing the test file size.
Pr #70 introduced an optimization that results in small files (smaller than the pathname of the delta source) being copied directly to the delta, skipping the delta process on those files. This helps speed things up, especially with bootc images.
However, this also broke codecov a bit. The new optimization results in old codepaths not being executed. This then makes codecov sad, as it thinks we're not testing as much code as we used to be.
This is likely due to the tests using small files, hitting the optimization, and skipping the codepaths they used to traverse. Thus, we need to update the tests to ensure that the old codepaths are hit again as they used to be. (I think this should largely be solved by increasing test file size, although I haven't totally confirmed myself yet.)
TLDR: tests need to be updated to return to traversing the codepaths they used to cover, likely by increasing the test file size.