Skip to content

S4: inheritance of validity methods?

6 messages · Martin Morgan, Vitalie S.

#
Dear Developers,

In current implementation of validity method, objects are first coerced to  
superclass (slots are striped). Thus, it is not possible to write validity  
method which would perform some checks on children slots.

Say, I want to check if number of slots in a class is equal to "n":

setClass("A", representation(a="numeric", n="integer"),
          prototype=list(a=12, n=1L),
          validity=function(object){
              if(length(slotNames(object))!=object at n+1) paste("Number of  
slots must be ", object at n)
              else TRUE
          })


setClass("B", representation(b="numeric"), contains="A",
          prototype=list(a=12, b=14, n=2L))


new("B", a=11, b=33)
Error in validObject(.Object) :
   invalid class "B" object: Number of slots must be  2

Error, because an object of class "A" is passed to validObject with one  
slot "b" removed and n=2.

Is were a work around for this, or I am just doomed to write the same  
validity method for each children class?

Many thanks,
Vitalie.

--
#
"Vitalie S." <vitosmail at rambler.ru> writes:
I think you're doomed, but checking the number of slots in an instance
seems like a very unusual validity method -- isn't this the job of the
methods package, not your implementation of particular classes?

Martin

  
    
#
On Tue, 18 Aug 2009 15:42:48 +0200, Martin Morgan <mtmorgan at fhcrc.org>  
wrote:
That was a toy example. In fact that I need is to ensure that some  
specific slots have the same length. Names of these slots are kept in  
separate slot (@atributes).

I think it's not an insane idea to require a predefined relationship to  
hold between "future" children's slots.

To be honest, I find it hard to think of a compelling reason to convert  
the object before testing the validity. What can possibly go wrong if  
children slots are kept in place?


Vitalie.
--
#
Vitalie S. wrote:
Perhaps I was too quick in saying you were doomed. First, if A is
VIRTUAL then the class seen by its validity method is, as implied by ?as,

  the object will be from the closest actual class, in particular the
  original object, if that class extends the virtual class directly

so

  .validity <- function(object) {
      cat(class(object), "\n")
      TRUE
  }

  setClass("A",
           representation(a="numeric", n="integer", "VIRTUAL"),
           validity=.validity)

  setClass("B", representation(b="numeric"), contains="A")

leads to
B

This seems to require some careful thought about how derived classes are
to behave.

Another approach, probably obvious and which does not really bring
anything 'automatic', is to define a helper function that can be called
in each validity method, or itself used as the validity method.

  .validity <- function(object) {
     # TRUE or message
  }

  setClass("A", representation(a="numeric", n="integer"))
  setClass("B", representation(b="numeric"), contains="A",
           validity=.validity)
  setClass("C", representation(c="numeric"), contains="A",
           validity=function(object) .validity(object))

one might make .validity generic, so that one gets inheritance &
dispatch in the validity methods.

 setGeneric(".validity")
 setClass("A", representation(a="numeric", n="integer"))
 setClass("B", representation("A", b="numeric"), validity=.validity)
 setMethod(.validity, "A", function(object) {
    cat(class(object), "\n")
    TRUE
 })
I guess this is reflected in 'setIs' / 'setAs', which allow for derived
classes to have and interpret their representation in a way that is
completely different from parent classes.

Martin
#
On Tue, 18 Aug 2009 18:58:27 +0200, Martin Morgan <mtmorgan at fhcrc.org>  
wrote:
yes, indeed. At first glance seems as a nice trick, but:
[1] "C"
B

yah... same story at a different level:(.
This one has similar problem:


setGeneric(".validity", useAsDefault=function(object) {
     cat(class(object), "\n")
     TRUE
})

  setClass("A", representation(a="numeric", n="integer"),  
validity=.validity)
  setClass("B", representation("A", b="numeric"), validity=.validity)
  setClass("C", representation("B", c="numeric"), validity=.validity)
A
B
A
B
C

?validObject says:

      Validity testing takes place *bottom up*(...)for each of the classes  
that this
      class extends (the ?superclasses?), the explicit validity method
      of that class is called, if one exists.  Finally, the validity
      method of 'object''s class is called, if there is one.

This is ectly what happens above when A,B,C is printed.

It seams like there is no way to implement a general function which would  
be called for validation of each children class. Moreover, if I attempt to  
rewrite manually the validity method for each subclass taht would also not  
work, because validity is tested "bottom up". validObject for A is called  
anyway - and is obviously not valid:


  setClass("A", representation(a="numeric", n="integer"),
           validity=function(object) length(slotNames(object))==object at n+1
           )

  setClass("B", representation(b="numeric"), contains="A",
           validity=function(object) length(slotNames(object))==object at n+1
           )
Error in validObject(.Object) : invalid class "B" object: FALSE

So there is no way to validate at all!!!
--
1 day later
#
With the help of Martin Morgan here is the solution for the above question.

First create VIRTUAL class with all the test for children classes:

setClass("TESTS", validity=function(object) {
     cat("Validity here:", class(object), "\n")
     TRUE
})

.simple <- function(object) {
     cat("No validity -", class(object), "\n")
     TRUE
}
setClass("A", representation(a="numeric", n="integer"),
          validity=.simple)
setClass("B", representation("A", b="numeric"),
          validity=.simple)
setClass("C", representation("B", c="numeric"),
          validity=.simple)

Set each children to inherit from TESTS class:

setIs("B", "TESTS")
setIs("C", "TESTS")

Now we have exactly what we wanted, validity tests are executed only once:
Validity here: B
No validity - A
No validity - B
No validity - A
Validity here: C
No validity - B
No validity - C

The order in which validity is called above is somewhat counter intuitive.

For class "C" complete hierarchy is this:
[1] "B"     "TESTS" "A"     "TESTS" "TESTS"

so if validity would start bottomup then TEST should be the first to  
follow. And vector of unique superclasses would be
c("B", "A", "TEST").

Instead validity uses the vector of unique labels constructed *TOPDOWN* :
[1] "B"     "TESTS" "A"


Apparently this is an inconsistency in implementation, isn't it?

Vitalie.







On Mon, 17 Aug 2009 20:06:44 +0200, Vitalie S. <vitosmail at rambler.ru>  
wrote:
--