bitand(NaN,1) causes error

This is in R2011a, but possibly obscure enough to have persisted. bitand() appears to tolerate NaNs in the input unless the NaN appears as the last element of the input vector.

>> bitand([0 1 NaN],1)

??? Error using ==> bitand
Exceeded value of bitmax.

>> bitand([0 1 NaN 0 1],1)

ans =

     0     1   NaN     0     1

Has this been fixed in later versions?

Answer 1

[Dennis J]

Yes it has been fixed, now NaN is no longer accepted as input:

bitand([0 1 NaN 0 1],1)

will give

Error using bitand
Double inputs must have integer values in the range of ASSUMEDTYPE.