octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Single/Double precision NA values


From: David Bateman
Subject: Re: Single/Double precision NA values
Date: Mon, 28 Jul 2008 13:49:27 +0200
User-agent: Thunderbird 2.0.0.12 (X11/20080306)

Olaf Till wrote:
> Just an issue: For some statistical functionality, I have used R from
> Octave over bidirectional pipes. Floating point data was sent to and
> from R over the pipes. NaN or NA values did not occur, but probably
> might be an issue in such cases.
>   
In such a case if you have popened the R process from Octave or
visa-versa, the numerical values are passed in an ascii form and so the
NA value will be passed as a string and so shouldn't be an issue. I
suppose you might fread and fwrite from/to the pipe and in that case yes
the binary value of the NA value will be important. So if this is the
case you are advocating the translation of the internal NA value of
Octave to an R compatible value when writing to a pipe..

All of this is really a great big kludge that could be fixed if the R
developers decided to change their NA value to something that handled
the single to double precision conversion and dealt with the backward
compatibility issues on their end, but their lack of response to my
e-mails kind of makes me think they don't care..

If the criteria for the Octave patch for proper casting of NA between
double and single precision is full R compatibility then I see no way of
implementing this without a kludge in Octave.

D.

-- 
David Bateman                                address@hidden
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary



reply via email to

[Prev in Thread] Current Thread [Next in Thread]