Access list element based on attribute value

Question

Say I have a list whose elements have attributes, as below:

my_list <- list()

my_list[[1]] <- 1:10
my_list[[2]] <- 11:20
my_list[[3]] <- 21:30

attr(my_list[[1]], "att1") <- "a"
attr(my_list[[2]], "att1") <- "b"
attr(my_list[[3]], "att1") <- "c"

attr(my_list[[1]], "att2") <- "1"
attr(my_list[[2]], "att2") <- "2"
attr(my_list[[3]], "att2") <- "3"

Now, pretend this list is many hundreds of elements long, and I don't know a priori which element of the list has the attributes I want. But I know that I want the element, with say, att1 == "b" and att2 == "2" (but where I don't know that it happens to correspond to list element 2).

Is there a way in R to look up which element(s) in a list has a particular combination of attributes?

Answer 1

You can filter a list with Filter:

Filter(function(x) attr(x, "att1") == "b" & attr(x, "att2") == "2", my_list)

If you expect the element to be unique and want to select it, add [[1]] on the end.

---

Personally, I'd put the data into a table:

library(data.table)
myDT = data.table(
  att1 = sapply(my_list, attr, "att1"), 
  att2 = sapply(my_list, attr, "att2"),
  data = my_list
)

#    att1 att2               data
# 1:    a    1       1,2,3,4,5,6,
# 2:    b    2 11,12,13,14,15,16,
# 3:    c    3 21,22,23,24,25,26,

Then you can verify that att1 + att2 uniquely pin down an element

nrow(myDT) == uniqueN(myDT, by=c("att1", "att2"))
# [1] TRUE

and write a helper function for subsetting

setkey(myDT, att1, att2)
get_element = function(a1, a2) myDT[.(a1, a2), data[[1]]]

get_element("b", "2")
#  [1] 11 12 13 14 15 16 17 18 19 20
# attr(,"att1")
# [1] "b"
# attr(,"att2")
# [1] "2"

You might also want to look at the purrr and broom packages, which offer different, "tidyverse" syntax for tables with a list column.