[Rd] dim(a <- ...) sets invisible flag erronously (PR#587)

[Prof Brian Ripley]

> Date: Fri, 30 Jun 2000 20:15:51 +0200 (MET DST)
> From: maechler@stat.math.ethz.ch
> To: r-devel@stat.math.ethz.ch
> Subject: [Rd] dim(a <- ...) sets invisible flag erronously (PR#587)
> CC: R-bugs@biostat.ku.dk
> X-Loop: R-bugs@biostat.ku.dk
> 
> {Nothing really harmful;  R 1.1.0}
> 
> dim(.) sets (or doesn't clear) the  ``invisible flag'' for printing in
> certain circumstances:
> 
>   > dim(a <- cbind(1))
>   >
> 
> does *not* print as it should :
> 
>   > dim(a)
>   [1] 1 1

This is nothing to do with dim: any function will do this.

> foo <- function(x) x
> foo(a <- cbind(1))

It happens because the way invisible is implemented is different in R and S:
in S there is a .Auto.print var in frame 1, and in R there is a global
variable R_visible, so anything during the evaluation that unsets the
flag will ensure that the result is invisible.

I don't see how to avoid this without checking the `depth' in the evaluation
pretty carefully (and R's semantics differ from those of S in that
braced expressions do not generate a frame).

Is this worth fixing?  print(.Last.value) works all the time, and perhaps
you might be persuaded not to write code that uses assignments as arguments?

> ---
> 
> Question to the real "hackers" :
> 
> How can I check this using R code alone (not using files),
> i.e. (how) can I ask an R expression if its ``invisible flag'' is set or not ?
> 
> {help(invisible) doesn't help here}.
> 
> Maybe one would want an
> 
>       is.invisible(.)
> 
> function?

There seems to be an unused internal function do_visibleflag in print.c.

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel 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-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._