https://bugs.documentfoundation.org/show_bug.cgi?id=136615
--- Comment #31 from Eike Rathke <er...@redhat.com> --- (In reply to Albrecht Müller from comment #29) > @Eike Rathke: > Rounding any day's 23:59:59.6 into the next day's 00:00:00 is related to > another thing: Adding exactly one minute to another exact minute should > always give two minutes – not sometimes one minute and sometimes two > minutes. I think you are mixing up things. Or at least the context and conclusion/example doesn't make sense to me. > You may want to have a look at the document “Trying to clarify > clock vs duration time, version 0.0.1” (see Bug 127170, attachment 165304 > [details]) to read about the mathematical background. Repeating my latest comment from that bug 127170: Note that the [there] attached sample document already latest since LO 6.2.8 (probably 6.2.5 for which bug 125099 was fixed) when recalculated produces the expected results even with the HH:MM:SS format and in column D has 0.0 differences. The upcoming 24.2 version even further refines the calculation of (date+)time differences aka duration to eliminate a possible slight accuracy error due to IEEE 754 double floating point values. (In reply to Regina Henschel from comment #30) > @Eike: The attachment lists the results from several application. These > results show, that a standardization is indeed needed. Since ODF 1.4 has the > 'round' variant, LibreOffice needs to provide a function according this > standard. As mentioned elsewhere, I consider that rounding specification *WRONG* and advise *AGAINST* it. Some major implementations (Excel, Gnumeric, Google) doing it wrong to be compatible with Excel doesn't make it right. > Eike, do you have an idea how we can do it? No. You can have either or, not both. Or implement the bad specification and introduce another set of functions like CLOCK.SECOND() and so on to obtain the unrounded matching wall clock values. > The problem I see is, > that existing documents would have a different result in ODF 1.4 than in ODF > 1.3, if we simply change the algorithms. Yes, that may happen. -- You are receiving this mail because: You are the assignee for the bug.
