Blake, are you actually seeing this behavior, or just reading through the code? By far the best thing you could do here is write a new case for the test suite which demonstrates the bug. Do that, and you will be my hero, and I will fix your bug.
In any case, unless you think the patch causes this new issue, I think it sounds like probably a different bug altogether. The fine generator as designed always generates fines through "now", then potentially voids/adjusts for backdate. The number of fines necessary to reach "now" is calculated at line 554, and I am not sure how max_fines acting as an early abort might actually extend the number of fines generated.
Blake, are you actually seeing this behavior, or just reading through the code? By far the best thing you could do here is write a new case for the test suite which demonstrates the bug. Do that, and you will be my hero, and I will fix your bug.
In any case, unless you think the patch causes this new issue, I think it sounds like probably a different bug altogether. The fine generator as designed always generates fines through "now", then potentially voids/adjusts for backdate. The number of fines necessary to reach "now" is calculated at line 554, and I am not sure how max_fines acting as an early abort might actually extend the number of fines generated.