improving speed of list which need some cleaning - r

Question

I'm trying to improve the speed of one of my script in R, one of the part which take a long time is the cleaning of a list under certain conditions.
Might not be necessary to perfectly understand what I want to do, and go directly to the code that I need to improve.
So here's the thing:
I have a list, each element of the list is a list of 2 elements :
- The first element is a vector of integer with a length between 1 and 4
- The second element is a vector of boolean of length 6
here's a piece of code to create such a list (of 1000 elements):
numberList<-1000
l.gen<-lapply(1:numberList,function(i){
return(list(var = floor(runif(floor(runif(1,1,5)),1,7)),vec = as.logical(floor(runif(6,0,1.99)))))
})
kind of look like that :
> l.gen
[[1]]
[[1]]$var
[1] 1 4 2
[[1]]$vec
[1] FALSE FALSE FALSE FALSE TRUE TRUE
[[2]]
[[2]]$var
[1] 3
[[2]]$vec
[1] FALSE FALSE FALSE TRUE TRUE FALSE
[[3]]
[[3]]$var
[1] 6
[[3]]$vec
[1] TRUE FALSE TRUE FALSE TRUE TRUE
[[4]]
[[4]]$var
[1] 6
[[4]]$vec
[1] TRUE TRUE TRUE FALSE FALSE FALSE
Now to cleaning part,
I want to remove from this list all elements "l" that meet two conditions:
the "$vec" of the element "l" has more than 3 times TRUE in common with another element of the list:
for example :
$vec
[1] TRUE TRUE FALSE TRUE TRUE FALSE
and
$vec
[1] TRUE FALSE FALSE TRUE TRUE FALSE
has 3 TRUE in common (the one's in 1st, 4th and 5th position) so it doesn't match the condition.
The second condition is tested in case we have the first one :
the $var of the element should have at least one element in common (doesn't matter their respective positions)
so
[[1]]$var
[1] 1 4 2
and
[[1]]$var
[1] 3 1
meet that condition (because the "1" is in both vectors)
In case two elements of the list meet both those conditions I delete the one with the shorter $var
for example in :
[[3]]
[[3]]$var
[1] 6
[[3]]$vec
[1] TRUE FALSE TRUE FALSE TRUE TRUE
[[3]]
[[3]]$var
[1] 6 3 5
[[3]]$vec
[1] TRUE TRUE TRUE FALSE TRUE TRUE
this element should be deleted :
[[3]]
[[3]]$var
[1] 6
[[3]]$vec
[1] TRUE FALSE TRUE FALSE TRUE TRUE
So here is the code I've tried that meet my requests :
res<-lapply(l.gen,function(l){
for (i in 1:length(l)){
if (length(l$var)<length(l.gen[[i]]$var)){
in.common<-sum(l$vec&l.gen[[i]]$vec)
if(in.common>limit){
var.in.common<-sum(l$var%in%l.gen[[i]]$var)
if(var.in.common>0){
return(NULL)
} else {
return("OK")
}
} else {
return("OK")
}
}
}
})
It works fine but it's kind of slow when the list is very Big.
I've tried to change the for loop with another lapply but it takes more time when the list is big as the return() in the for-loop works like a "break;" which can't be done in the lapply() which try every single element of the list.
I'm opened to every suggestion that might help.