?plot(x) when x is cd.fit doesn't give man page

[scottyaz]

R> ?plot(x)
Error in .helpForCall(topicExpr, parent.frame()) : 
  no documentation for function ‘plot’ and signature ‘x = "cd.fit", y = "ANY"’

[nickreich]

It's not clear to me what behavior you are looking for here. I never have used this ?fun() syntax, I just use ?fun.

But for example, if I run the following code:
y=rnorm(100)
x=rnorm(100)
f=lm(y~x)
?plot(f)
?plot

The two help calls bring up the same help file. So it does seem like it's a problem that your code is throwing an error, but if it's not breaking R CMD FIT, then maybe it's not a huge deal... Can we find other examples from other packages that give this same error?

Here's one:
(fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy))
?plot(fm1)
Error in .helpForCall(topicExpr, parent.frame()) :
no documentation for function ‘plot’ and signature ‘x = "ANY", y = "ANY"’

I'm not sure what the issue is here, although it does seem like something is buggy. But maybe at the R level and not at our package level.

[scottyaz]

I found somewhere in the S4/R documentation that ?function(object.myclass) should give the man page for the method for the class object.myclass. This works for ?logLik(fit) at the moment.