[Rd] Strict seq: a recommendation

[Gabor Grothendieck]

This is a longstanding problem when programing with seq.
In the common case that one wants 1,...,n such that if n is zero
a zero length answer is wanted, one can write seq(length = n)
instead of 1:n.  However, the general case is still a problem,
as you point out.

On 8/23/06, Dominick Samperi <dsamperi at decisionsynergy.com> wrote:
> I should point out that this recommendation was not designed just
> to accommodate C/C++, it was motivated by problems
> with R code of the form:
> for(m in seq(beg,end))
> val <- x[m]
>
> Take beg=1, end=0, for example. This results in
> val <- x[0]
>
> and R issues no error about the subscript being out
> of bounds. Later when val is used you get not very
> informative error messages about variables of
> length zero (because val has length zero after the
> above assignment).
>
> ds
>
>
> Thomas Lumley wrote:
> > On Wed, 23 Aug 2006, Dominick Samperi wrote:
> >
> >> In for loops of the form:
> >> for(m in seq(beg,end))
> >> ...
> >>
> >> when beg > end, the default behavior is
> >> to decrement, and the action described
> >> by ... is executed at least once.
> >>
> >> On the other hand, if you view this
> >> construction as a translation of the C code:
> >> for(m = beg; m < end; m++)
> >> ...
> >>
> >> then the semantics of C/C++ is not
> >> respected, because the code in ... is
> >> not executed when beg > end in
> >> the case of C/C++.
> >
> > There is another important way in which the C loop is different from
> > the R loop.  If you modify m or end inside the loop the number of
> > iterations of the C loop changes but the number of iterations of the R
> > loop doesn't.  R's for() loop just isn't the same as C's.
> >
> >     -thomas
> >
> > Thomas Lumley            Assoc. Professor, Biostatistics
> > tlumley at u.washington.edu    University of Washington, Seattle
> >
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>