[R] Help files mismatches and R CMD check

[Prof Brian Ripley]

On Thu, 3 Jul 2003, Frank E Harrell Jr wrote:

> I have a function t.test.cluster for which R CMD check results in a
> documentation warning because its argument names don't match those of t(
> ).  Is there a way around this error besides renaming the function?  
> Will this disqualify a package from admission to CRAN?  How did t.test
> avoid this warning?

t.test is a generic, so it avoids the warning. Sounds as if we would need 
either to improve the logic (so it does identify t.test as the appropriate 
generic) or add this to the stop list of reasonable exception (in package 
tools/R/utils.R).  In any case, it's our not your problem.

Brian

-- 
Brian D. Ripley,                  ripley at 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 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595