Sum named vector values where the names are reversed in R

Question

I have a list of named vectors. I am trying to sum their values. But some of the names within a vector have reversed equivalents. For example, if I have some data that looks like this:

myList <- list(`1` = c('x1:x2' = 2, 'x2:x1' = 1, 'x3:x4' = 1),
               `2` = c('x1:x2' = 3, 'x6:x1' = 2, 'x1:x1' = 1, 'x4:x3' = 1),
               `3` = c('x3:x4' = 2, 'x1:x2' = 1, 'x4:x3' = 4),
               `4` = c('x5:x2' = 1, 'x2:x5' = 1)
               )
> myList
$`1`
x1:x2 x2:x1 x3:x4
    2     1     1

$`2`
x1:x2 x6:x1 x1:x1 x4:x3
    3     2     1     1

$`3`
x3:x4 x1:x2 x4:x3
    2     1     4

$`4`
x5:x2 x2:x5
    1     1

Here, we can see that in myList[[1]] we have x1:x2 = 2 and x2:x1 = 1. As these are the reverse of each other, they are equivalent, so, essentially, x1:x2 = 3.

I am trying to sum the values for each named element (including the reverse) over each list element.

My desired output would look something like this:

    var count listNo
1 x1:x2     3      1
2 x3:x4     1      1
3 x1:x2     3      2
4 x6:x1     2      2
5 x1:x1     1      2
6 x4:x3     1      2
7 x3:x4     6      3
8 x1:x2     1      3
9 x5:x2     2      4

Answer 1

This is tricky. I'd be interested to see a more elegant solution

`row.names<-`(do.call(rbind, Map(function(vec, name) {  
    x <- names(vec)
    l <- sapply(strsplit(x, ":"), function(y) {
      paste0("x", sort(as.numeric(sub("\\D", "", y))), collapse = ":")
      })
    df <- setNames(as.data.frame(table(rep(l, vec))), c("var", "count"))
    df$listNo <- name
    df
  }, vec = myList, name = names(myList))), NULL)

#>     var count listNo
#> 1 x1:x2     3      1
#> 2 x3:x4     1      1
#> 3 x1:x1     1      2
#> 4 x1:x2     3      2
#> 5 x1:x6     2      2
#> 6 x3:x4     1      2
#> 7 x1:x2     1      3
#> 8 x3:x4     6      3
#> 9 x2:x5     2      4

^{Created on 2022-03-06 by the reprex package (v2.0.1)}

Answer 2

Another tidyverse option could be:

map_dfr(myList, enframe, .id = "listNo") %>%
    mutate(var = map_chr(str_split(name, ":"), ~ str_c(sort(.), collapse = ":"))) %>%
    group_by(listNo, var) %>%
    summarise(count = sum(value))

  listNo var   count
  <chr>  <chr> <dbl>
1 1      x1:x2     3
2 1      x3:x4     1
3 2      x1:x1     1
4 2      x1:x2     3
5 2      x1:x6     2
6 2      x3:x4     1
7 3      x1:x2     1
8 3      x3:x4     6
9 4      x2:x5     2