problem with sub in graphs

[Martin Maechler]

>>>>> "Colin" == Colin Farrow <C.Farrow@geology.gla.ac.uk> writes:

   Colin>  with R-0.62.1 under Solaris 2.5.1.

    Colin> Use of the graphical parameter 'sub' in a high level graphics
    Colin> function is inconsistent, and incompatible with the blue book

    >> plot(1:10, main="test", sub="demo")

   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function
   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function
   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function
   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function
   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function
   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function

   Colin> ... and why so many warnings?

Because sub is part of the "..." which are passed down from  plot.default()
to several lower-level functions,
each of which creates (at least) one warning.

     >> hist(1:10, main="test", sub="demo")

   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function
   Colin> Warning: parameter "sub" couldn't be set in high-level plot() function

The same here:  hist.default() passes "..." to

	title(..) -- where it works
and to	axis(1,..) \
and	axis(2,..) / where it fails and produces the two warnings.

   Colin> ... despite the warnings the sub title is drawn !


   Colin> I think the problem existed in earlier releases of R but the
   Colin> warnings were not printed.

Exactly.  And because the problem existed, I liked the warnings to appear.
The idea is that the R code (not the low level C graphics) should be fixed;
This would mean that R functions must make sure to only path
valid "..." stuff to C.

Usefully, 
we would have a global option to turn off these warnings.
(as the other graphics warnings ! ==>   
 ``par(err = -1)'' would be the thing
)

Martin 
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._