https://bugs.documentfoundation.org/show_bug.cgi?id=136615

--- Comment #29 from Albrecht Müller <albrecht.muel...@astrail.de> ---
@Regina Henschel:
You know the Open Document Format Specification better than I do: Does it
define the semantics of format strings, especially those related to date and
time? Does it specify how the interpretation of these strings is related to the
specification of the date and time functions you mention in Comment #27?

I fear that these things are not specified and that this is a big hole in the
specification: Compliant applications are free to implement different and
incompatible interpretations of format strings. The discussion shows that there
are many ways how to do this. In an environment where working time is paid for
it may be cheaper to pay license fees to Microsoft than to pay for the time
necessary to handle the undocumented subtle but nevertheless important
differences between these ways.

@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. 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) to read about the
mathematical background. I think there is simply a design decision between
having the change of day exactly at midnight and getting elementary arithmetic
right; you cannot have both. If I had a wall clock showing time units in the
order of magnitude of round off errors I would not be able to read the values
anyway. Therefore I would prefer correct elementary arithmetic on hours,
minutes and seconds. On the other hand I understand that there are good reasons
do put the day change exactly at midnight: There are a lot of functions other
than SECOND, MINUTE etc. that probably rely on a day change exactly at
midnight. So there is the ugly situation that LibreOffice uses two different
interpretations of time values and that this is an undocumented feature.

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to