diff options
author | Eike Rathke <erack@redhat.com> | 2020-12-01 19:08:26 +0100 |
---|---|---|
committer | Eike Rathke <erack@redhat.com> | 2020-12-02 00:27:30 +0100 |
commit | 49aff144c72a5258cf2ca392a0cfb7a31fb86819 (patch) | |
tree | 9f9db993150f964704f27ad51aac7a3b18b404af /sal | |
parent | 462c79c9e9c9163bd9b7ce48e19c9c2f5d49333c (diff) |
Related: tdf#138360 Rounding integers to decimals is futile
Change-Id: Ica25747a26d6c2637c46808d1b73aeeed6e1df37
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/107001
Reviewed-by: Eike Rathke <erack@redhat.com>
Tested-by: Jenkins
Diffstat (limited to 'sal')
-rw-r--r-- | sal/rtl/math.cxx | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/sal/rtl/math.cxx b/sal/rtl/math.cxx index 1d6cb88327f9..10417742b3a2 100644 --- a/sal/rtl/math.cxx +++ b/sal/rtl/math.cxx @@ -1146,7 +1146,10 @@ double SAL_CALL rtl_math_round(double fValue, int nDecPlaces, // Rounding to decimals between integer distance precision (gaps) does not // make sense, do not even try to multiply/divide and introduce inaccuracy. - if (nDecPlaces >= 0 && fValue >= (static_cast<sal_Int64>(1) << 52)) + // For same reasons, do not attempt to round integers to decimals. + if (nDecPlaces >= 0 + && (fValue >= (static_cast<sal_Int64>(1) << 52) + || isRepresentableInteger(fValue))) return bSign ? -fValue : fValue; double fFac = 0; |