Extend exists to nested dictionary/list - r

Question

How can I extend the exists function to work with the following:
Any ideas how I would extend to this to looking at seeing whether a nested dictionary would also exist. I.e. for example: if(exists("mylists[[index]]['TSI']")), where the mylists object is a dictionary look up that also wants to contain a nested dictionary.
Now mylists will look like:
[[index]]["TSI"]=c(0="a",1="b")
How should I check this exists so that I may append it so I have:
[[index]]["TSI"]=c(0="a",1="b",2="c")
Here is more code that illustrates things better:
index is an ID
if(!is.null(listsar[[index]]["TSI"])) {
print("extending existing")
listsar[[index]][["TSI"]] <- c(listsar[[index]][["TSI"]], risktype=myTSI)
}else
{
print("creating new")
listsar[[index]][["TSI"]] <- c(risktype=myTSI)
}
However this does not seem to work. I get the "extending existing" and I never seem to get the "creating new". If I change the evaluation line to:
if(!is.null(listsar[[index]][["TSI"]]))
I get different statement:
"creating new"

Answer 1

You can test for NULL in most cases. Sample data (which is something you should have given us along with working code - wtf is c(0="a",1="b",2="c") supposed to be?)
> mylists=list()
> mylists[["foo"]]=list()
> mylists[["foo"]][["TSI"]]=c(a=0,b=1)
Does a "foo" exist at the top level?
> !is.null(mylists[["foo"]])
[1] TRUE
Yes.
Does a "fnord" exist at the top level?
> !is.null(mylists[["fnord"]])
[1] FALSE
No.
Does a "TSI" exist within "foo"?
> !is.null(mylists[["foo"]][["TSI"]])
[1] TRUE
Yes.
Does a "FNORD" exist within "foo"?
> !is.null(mylists[["foo"]][["FNORD"]])
[1] FALSE
No.
Does a "FNORD" exist within a top-level (and nonexistent) "fnord":
> !is.null(mylists[["fnord"]][["FNORD"]])
[1] FALSE
No.