https://bz.apache.org/ooo/show_bug.cgi?id=126908
--- Comment #10 from orcmid <[email protected]> --- (In reply to orcmid from comment #9) > UNRESOLVED: Why there are jumps with some implementations is still > unresolved and we have not been able to reproduce it. If there are jumps > up, there should be places where a jump "down" is accomplished by the same > date twice in a row earlier. This smells like a rounding bug. It would happen when the serial-day value is broken down into years and day-within-year values. This may be a hardware difference and it may be a run-time library difference. My tests are on Microsoft Windows 10 Pro x64 and Intel i7-3770 CPU chips. I don't know whether may system has a later version of the Microsoft VC++ redistributable runtime than is installed with Apache OpenOffice otherwise. -- You are receiving this mail because: You are the assignee for the issue.
