int(1e20) yields -9.223D+18 (regression)
Reported by Samuel GOUGEON (@sgougeon)
BUG DESCRIPTION:
----------------
int(a) is wrong for finite values a >~ 1e18
Processing of %inf, -%inf and %nan is correct.
Scilab 5.5.2:
-->int(1e19)
ans =
1.000D+19
-->int(-1e19)
ans =
- 1.000D+19
Scilab 6.0:
--> int(1e19)
ans =
-9.223D+18 !!!!!
--> int(-1e19)
ans =
-9.223D+18 !!!!!
ERROR LOG:
----------
None. Wrong result
HOW TO REPRODUCE THE BUG:
-------------------------
int(1e19)
int(-1e19)
OTHER INFORMATION:
------------------
Source: https://github.com/rdbyk/balisc/issues/613