https://bugs.freedesktop.org/show_bug.cgi?id=80732
Priority: medium
Bug ID: 80732
Assignee: [email protected]
Summary: MOD function may provide false result when first
argument is calculated
Severity: normal
Classification: Unclassified
OS: Linux (All)
Reporter: [email protected]
Hardware: x86-64 (AMD64)
Status: UNCONFIRMED
Version: 4.0.6.2 release
Component: Spreadsheet
Product: LibreOffice
Created attachment 102032
--> https://bugs.freedesktop.org/attachment.cgi?id=102032&action=edit
some tests on MOD() function
I do not want reopen bug 50299 but testing that bug I made some error when
typing and tried the following:
MOD(4.1*100;10)
MOD(5.1*100;10)
Both give 10 as result instead of 0. We have the same result for the range 64.1
to 81.1. All other values I tested (0.1, 1.1, 2.1, ..., 15.1) give 0 or a very
small value which is OK for me (yes, I know what rounding errors are).
I do not know how MOD() function is computed but if the root cause is the same
as in bug 50299, something should be done to better control rounding errors in
this particular case.
NB: the function QUOTIENT() gives the correct result in all corresponding
cases.
Tested with LO 4.3.0.1.0+, 4.2.6.0+, 4.1.6 and 4.0.6 on Ubuntu 14.04 x86-64
Best regards. JBF
--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Libreoffice-bugs mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs
