Hi Andrew,
My primary concerns
1. Change in behavior that may break existing code.
the intent of the patch is to restore the previous behaviour ( by
previous behaviour I mean before integration of the feature that
erroneously removed the 'special' processing for '+' for dates ).
Additionally the intent of the patch is to make the behaviour for '-'
similar for'+' e.g. if Date + 1 gives a Date then surely Date - 1 should
additionally give back at Date. Hopefully that makes sense :-)
2. Previous behavior was never flagged as an issue, and may arguably
be considered logical to some. for example, storing a "delta" in the
native type.
I don't think we should be afraid to change present behaviour if we
think it is wrong and especially if the impact is probably going to be
very minor ( as I think is the case here ). I would bet that the
behaviour we have at the moment is an incomplete attempt to behave in a
compatible way with vba. Where we can, I think if the cost is low we
should try to be compatible ( should actually look at what vba does
[*]). But... I take it that you are unhappy with changing the treatment
of '-' ? and wouldn't be in favour of the change right? ( funny somehow
I thought in the previous mail thread that you were in favour of even
letting the Date type be set for all those numeric operations, sorry I
misunderstood you )
3. The rules are more complicated now and require complex examples to
explain; for example, time_1 + time_2 + time_3 returns a date but
time_1 + time_2 does not. We cannot simply state that math done with
dates or including a date returns a date.
yeah that doesn't sound right :-/ should remain a Date I would think
[*] table of behaviour in vba, pre the bug being introduced and post patch
mso old behaviour present
--------------------------------
date + num Date Date Date
date - num Date Double Date
date * num Double Double Double
date / num Double Double Double
date + date Date Double Double [1]
date - date Double Double Double
date * date Double Double Double
[1] the date + date is clearly wrong (fixing that would address '3.'
above. )
personally I am all for fixing this to be compatible
Noel
Context
Re: [Libreoffice] [REVIEW][3-4] date change regression fdo#44385 · Andrew Douglas Pitonyak
Privacy Policy |
Impressum (Legal Info) |
Copyright information: Unless otherwise specified, all text and images
on this website are licensed under the
Creative Commons Attribution-Share Alike 3.0 License.
This does not include the source code of LibreOffice, which is
licensed under the Mozilla Public License (
MPLv2).
"LibreOffice" and "The Document Foundation" are
registered trademarks of their corresponding registered owners or are
in actual use as trademarks in one or more countries. Their respective
logos and icons are also subject to international copyright laws. Use
thereof is explained in our
trademark policy.