[Libreoffice-bugs] [Bug 126481] An option for Rounding mode in ScInterpreter::ScRound

2019-08-01 Thread bugzilla-daemon
https://bugs.documentfoundation.org/show_bug.cgi?id=126481

Eike Rathke  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |WONTFIX

--- Comment #4 from Eike Rathke  ---
Problem is the actual underlying floating point value of =(105 * 1.75 + 50) /
1.1 is 212.47 and only for display rounded to 212.5 as it is within
the accuracy margin, you can check with =RAWSUBTRACT(212.5;(105 * 1.75 + 50) /
1.1) that produces 2.8421709430404E-14

There's not much we can do here (except always displaying the "real" floating
point value and get whining users). For 0 decimals (round to integer) ROUND()
effectively calls std::round() to avoid bad "correction" resulting in wrong
values, see bug 96821.

-- 
You are receiving this mail because:
You are the assignee for the bug.___
Libreoffice-bugs mailing list
Libreoffice-bugs@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

[Libreoffice-bugs] [Bug 126481] An option for Rounding mode in ScInterpreter::ScRound

2019-07-28 Thread bugzilla-daemon
https://bugs.documentfoundation.org/show_bug.cgi?id=126481

himajin100...@gmail.com changed:

   What|Removed |Added

   See Also||https://bugs.documentfounda
   ||tion.org/show_bug.cgi?id=61
   ||512

-- 
You are receiving this mail because:
You are the assignee for the bug.___
Libreoffice-bugs mailing list
Libreoffice-bugs@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

[Libreoffice-bugs] [Bug 126481] An option for Rounding mode in ScInterpreter::ScRound

2019-07-20 Thread bugzilla-daemon
https://bugs.documentfoundation.org/show_bug.cgi?id=126481

Regina Henschel  changed:

   What|Removed |Added

 CC||rb.hensc...@t-online.de

--- Comment #3 from Regina Henschel  ---
I think, that is nothing about rounding. That are the usual problems with
converting between decimal and binary numbers and with double arithmetic.
For example compare (show 15 decimal places)
=233.75/1.1-212
=233.75*(1/1.1)-212
or
=100/3-33
=100*(1/3)-33
=1000/30-33
=10/0.3-33
[The purpose of the parts -212 and -33 is to make differences in the last bits
visible.]

-- 
You are receiving this mail because:
You are the assignee for the bug.___
Libreoffice-bugs mailing list
Libreoffice-bugs@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

[Libreoffice-bugs] [Bug 126481] An option for Rounding mode in ScInterpreter::ScRound

2019-07-19 Thread bugzilla-daemon
https://bugs.documentfoundation.org/show_bug.cgi?id=126481

m.a.riosv  changed:

   What|Removed |Added

 CC||miguelangelrv@libreoffice.o
   ||rg
 Status|UNCONFIRMED |NEW
 Ever confirmed|0   |1

--- Comment #2 from m.a.riosv  ---
Looks like some kind of bug:

a1: 212,5000=(105 * 1,75 + 50) / 1,1
a2: 212,5000=ROUND(A1;16)
a3: 212,=ROUND(A1)

a5: 212,5000=VALUE("212,49")
a6: 213,000 =ROUND(A5)

-- 
You are receiving this mail because:
You are the assignee for the bug.___
Libreoffice-bugs mailing list
Libreoffice-bugs@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

[Libreoffice-bugs] [Bug 126481] An option for Rounding mode in ScInterpreter::ScRound

2019-07-19 Thread bugzilla-daemon
https://bugs.documentfoundation.org/show_bug.cgi?id=126481

--- Comment #1 from himajin100...@gmail.com ---
oops,typo:
=ROUND((105 * 1.75 + 50) / 1.1)

I just copied the formula from another reporter

-- 
You are receiving this mail because:
You are the assignee for the bug.___
Libreoffice-bugs mailing list
Libreoffice-bugs@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs