Hi Winfried, On Tuesday, 2012-05-15 08:09:37 +0200, Winfried Donkers wrote:
> You did a lot of work! I expected some or a lot of comments on my > interpretations > of ODF1.2 as i don't have Excel to compare with, but you did all the work > yourself! > I'm a volunteer, I don't mind cleaning up (my own) mess :) Heh, sorry for having taken away the work from you :-) I'll keep that in mind for next time. I thought I'd get that algorithm solved because I quite know the date calculation area and the quirks that may be lurking there. > > Some detailed nitpicks in commit comments :-) > > If I'm correct, the main changes are that my interpretation of ignore > years/months were not right. Hm, yes. I wouldn't say it's your fault though, the ODFF definition is a bit weak to implement that from scratch if one is not familiar with the area. > There is one thing I am not very happy about: DATEDIF in calc now does > not accept date1 to be later (larger) than date2. IMHO this should be > possible as it is a very normal possibilty. It looks as if DATEDIF > does not accept this condition because Excel doesn't/can't. I thought > calc conforms to ODF and can export to xls(x) but with possible loss > of functionality. Now it is like LibreOffice is an open source copy of > MS Office. Usually for spreadsheet functions we strive for Excel interoperability. (Actually they implemented this function because Lotus 1-2-3 had it the same way, anyway...) If we allowed reversed date arguments, importing a document in Excel would give different results, errors in this case. The ODFF definition lacks a constraint that date1 must be <= date2, seems we didn't think of that back then, I'll submit a comment. Also Gnumeric implemented it the same. > I would very much like to adjust the code to make it possible again > for date1 to be later (larger) than date2, both to increase its use :) > and because I don't want to be a follower of MS Office :( I'd rather stick with the current implementation for the above reasons. > And a big thank you for all your corrections/inprovements! > I hope to be able to send in more formula based patches, I like this area:) Great! I'm looking forward. Do you have already an idea what you'll pick next? Eike -- LibreOffice Calc developer. Number formatter stricken i18n transpositionizer. GnuPG key 0x293C05FD : 997A 4C60 CE41 0149 0DB3 9E96 2F1A D073 293C 05FD
pgpJsKpn9oXVd.pgp
Description: PGP signature
_______________________________________________ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice