R devel - Nov 2001 - R CMD check

[Sent to R-help by mistake. Sorry]

I have been checking my package coda with the current R-devel
using "R CMD check".  As usual, this has uncovered a large number
of errors and inconsistencies in my documentation, for which I
am very grateful.

The only problem I have is with code/documentation mismatches when I
have written a method for a generic function, e.g.

* checking for code/documentation mismatches ... WARNING
Objects with usage in documentation object `plot.mcmc' but missing from
code:
[1] "plot"

Objects with usage in documentation object `summary.mcmc' but missing from
code:[1] "summary"

Objects with usage in documentation object `time.mcmc' but missing from
code:
[1] "end"   "start" "time" 

Objects with usage in documentation object `window.mcmc' but missing from
code:
[1] "window"

This occurs because, for example, I have documented the usage of 
summary.mcmc as

\usage{
summary(data, quantiles = c(0.025, 0.25, 0.5, 0.75, 0.975), ...)
}

I'm being encouraged to document it as

\usage{
summary.mcmc(data, quantiles = c(0.025, 0.25, 0.5, 0.75, 0.975), ...)
}

but I don't think this is right. Surely the correct usage of a method
is to call the generic function?

Martyn
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Martyn Plummer wrote:
> I'm being encouraged to document it as
> 
> \usage{
> summary.mcmc(data, quantiles = c(0.025, 0.25, 0.5, 0.75, 0.975), ...)
> }
> 
I think you are being encouraged to document it as

 \usage{
 \method{summary}{mcmc}(data, quantiles = c(0.025, 0.25, 0.5, 0.75, 0.975), ...)
 }

So it displays properly in the documentation and also gets checked properly. You
will then stop getting the "usage ...but missing from code" error and
start
getting the "generic/method mismatch error." These guys are real slave
drivers,
but as you said 
>this has uncovered a large number of errors and inconsistencies in my
>documentation, for which I am very grateful.
Me too.

Paul Gilbert
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._