[Rd] is.vector could handle AsIs class better

[William Dunlap]

The use of the term 'vector' in R comes from S, where it was used, starting
in the latter part of the 1970s, to refer to the most primitive
(irreducible) parts of an object.  It has little to do with the
mathematical or physical concept of a vector and, in my opinion, should not
be used much by ordinary users.  In hindsight, it may have been better to
use some Joycean neologism instead of the word vector so people would not
have any notions of what it should do.

Bill Dunlap
TIBCO Software
wdunlap tibco.com


On Mon, Mar 30, 2020 at 2:26 AM Jan Gorecki <j.gorecki using wit.edu.pl> wrote:

> Thank you Gabriel,
> Agree, although I think that could be relaxed in this single case and
> AsIs class could be ignored.
> Best,
> Jan
>
> On Sun, Mar 29, 2020 at 7:09 PM Gabriel Becker <gabembecker using gmail.com>
> wrote:
> >
> > Jan,
> >
> > I believe it's because it has "a non-NULL attribute other than names" as
> per the documentation. In this case its class of "AsIs".
> >
> > Best,
> > ~G
> >
> > On Sun, Mar 29, 2020 at 6:29 AM Jan Gorecki <j.gorecki using wit.edu.pl>
> wrote:
> >>
> >> Dear R-devel,
> >>
> >> AsIs class seems to be well handled by `typeof` and `mode` function.
> >> Those two functions are being referred when explaining `is.vector`
> >> behaviour in manual. Yet `is.vector` does not seem to be handling AsIs
> >> class the same way.
> >>
> >> is.vector(1L)
> >> #[1] TRUE
> >> is.vector(I(1L))
> >> #[1] FALSE
> >>
> >> Is there any reason behind this behaviour?
> >> Could we have it supported so AsIs class is ignored when `is.vector`
> >> is doing its job?
> >>
> >> Best Regards,
> >> Jan Gorecki
> >>
> >> ______________________________________________
> >> R-devel using r-project.org mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-devel
>
> ______________________________________________
> R-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>

	[[alternative HTML version deleted]]