http://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=7688 --- Comment #104 from Galen Charlton <gmcharlt@gmail.com> --- (In reply to Paul Poulain from comment #102)
(In reply to Katrin Fischer from comment #101)
This is a huge patch. I am a bit worried of incorporating this without feedback from a library using serials and subscriptions - especially worried about the database update. Is it possible that we could get an anonymizsed database from a library with lots of subscriptions to test with?
Katrin, this improvement is used by 5 of our customers since more than one year. The feature has been signed off by frederic Demians, Kyle, and Bernardo. Not counting my (as QAer) tests, and Jared ones (as he pushed the patch before reverting it) So I think we should go forward, and push this patch (otherwise it will never be pushed).
Let me be clear: I will always exercise my own technical judgment. The fact that others have reviewed a patch is necessary but not sufficient for me to push it. That said, upon looking at this, I am seeing a lot that I like about this approach, but given that the main patch revamps the core of serials prediction without adding any new test cases, I will have to write some myself as part of validating the patch. This won't be instantaneous, but it won't take forever, either. -- You are receiving this mail because: You are watching all bug changes.