[R] Re: Matlab to R ?

[Paul Gilbert]

Martin Maechler wrote:

> Probably everyone agrees that a Matlab -> R translation cannot
> (and probably should not) be automated to full extent.  On the
> other hand, Paul Gilbert's "ex" script should be made into a
> perl or python {or "octave" !?} script and could be enhanced
> considerably, probably.
...

Perhaps I should have qualified the extent of the translation done by my
script somewhat more, but I'm not sure how much pay back there is from
doing a lot better job of the translation. (BTW, the script probably
pre-dates python and almost pre-dates perl.) I think you need to
distinguish at least three groups of interest in R/Matlab:

1/ Running Matlab code from R (Duncan Temple Lang's approach).

2/ Keeping a common code base to run in both R and Matlab. As far as I
know, no one is working on tools to do this, But Jim Ramsay's functional
data analysis library is based on this approach.

3/ One time conversion from Matlab to R.

My script was for the last purpose, but in addition I should add that it
is really meant for the case were one intends to keep  developing the R
version of the code once it is translated. If you just want a
translation that runs, then it is not very good (perhaps very bad). If
you want good R code with clean idioms, then even a running translation
is going to be pretty bad. So, you need to do a fair amount of work even
if you get a running translation. My script does a certain amount of the
boring part, but there is still a lot of work necessary to get good R
code.

Paul Gilbert
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._