[Rd] cat, print and documentation disagree (PR#8992)

[pdbailey at uchicago.edu]

The tone of your email makes it look like you think the exact
output of cat and print in these instances is irrelevant and
not worth thinking about and that the documentation is `close
enough,' or up to the (perhaps implied) R documentation
standard. This is a reasonable stance. If you hold it, please
just circular file my bug report as irrelevant and let's not
waste any more time on it.

But I'll focus on the claim you made explicitly. in response
to my pointing out that the documentation says that, ```cat'
converts numeric/complex vectors in  the same way as `print'
(and not in the same way as `as.character' which  is used by
the S equivalent), so 'options' '"digits"' and '"scipen"' are
relevant.'' You wrote that, ``[the cat() documentation] does
not say that it outputs the same characters [as print].''

You basic claim then is that they both convert numeric vectors
into characters in the same way, but that after the conversion
`cat()' mutates those characters (into subsets of the
converted characters), and that this second step in the
process of mutating the numbers into characters is not part of
the `conversion' of the number into characters. If you do
choose to stick by this odd claim, let me point out that the
claim implies that the documentation regards an internal
process that the user doesn't much care about because she can
never see the internal results without overriding `cat().'

I think what the documentation is saying is that in both
`print()' and `cat()' use the `options()' `digits' are
`sciphen' to the specificity they are described in the
`options()' documentation, but `cat()' and `print()' do not
follow the same convention to the extent that the `options()'
documentation does not specify the requirement.

Cheers,
Paul