False == NA_Logical about rpy2 HOT 4 CLOSED

Original comment by Hari Krishna Dara (Bitbucket: haridsv, GitHub: haridsv).

"False == NA_Integer" also comes back as True.

from rpy2.

Original comment by Laurent Gautier (Bitbucket: lgautier, GitHub: lgautier).

This is originally a "feature" (serious).

The rationale is that in Python:

#!python

>>> bool(0) == True
False

>>> bool(<any number other than zero>) == True
True

and that the NA representations are numbers.

However, I just look at nan and notice that:

#!python

>>> float('nan') == True
False

Either way this is not perfect, as Python does not have the native support for missing values that R has.
The most obvious example is for NA (that the missing value for booleans), as it is neither True or False. Can you bring this to the mailing-list so this is debated ?

from rpy2.

Original comment by Laurent Gautier (Bitbucket: lgautier, GitHub: lgautier).

Removing version: 2.1.0 (automated comment)

from rpy2.

This was fixed.

>>> ri.NA_Logical is False
False

from rpy2.