nojhan/eodev
Archived
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

eoRNG.h (eoRng::random): Add comment on truncation as standard way of

Browse source 
floating point to integer conversion.
This commit is contained in:
kuepper 2007-08-22 07:30:47 +00:00
commit 28e59ff064
2 changed files with 11 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

5
eo/src/utils/ChangeLog
View file

@ -1,3 +1,8 @@
2007-08-22 Jochen Küpper <jochen@fhi-berlin.mpg.de>
* eoRNG.h (eoRng::random): Add comment on truncation as standard way of
floating point to integer conversion.
2006-12-04 Jochen Küpper <jochen@fhi-berlin.mpg.de>
* eoRNG.h (eoRng::normal(stdev)): Revert erroneous change.

10
eo/src/utils/eoRNG.h
View file

@ -178,10 +178,12 @@ public :
*/
uint32_t random(uint32_t m)
{
// Make sure we always round towards zero, in order to get the
// half-open interval we want (as documented). The floor function
// does exactly this for the (always positive) values occuring here.
return uint32_t(floor(uniform() * double(m)));
// C++ Standard (4.9 Floatingintegral conversions [conv.fpint])
// defines floating point to integer conversion as truncation
// ("rounding towards zero"): "An rvalue of a floating point type
// can be converted to an rvalue of an integer type. The conversion
// truncates; that is, the fractional part is discarded"
return uint32_t(uniform() * double(m));
}
/** Biased coin toss