1

I've generated empty lists corresponding to every letter of the alphabet. The names of the lists take the form l_A, l_B, ... I also have a one-column data frame populated with strings (all lowercase, no symbols). I want to sort these strings by their first letter into the corresponding list, e.g., the word "aardvark" would go in list l_A.

I'm having problems calling the correct list when iterating through the data frame. This is my attempt to sort using for loops. For the example, I'm using a random list of words (all lowercase) built using the library "OpenRepGrid" to build the starting data frame.

install.packages("OpenRepGrid")
library(OpenRepGrid)

# random words data frame
df <- data.frame(tolower(randomWords(100)))

# lists for each letter
for (i in 1:26){
    nam <- paste0("l_", LETTERS[i])
    assign(nam, list()) 
    }

# sorting words in data frame into lists using for loops
for (i in 1:nrow(df)){
  for (j in 1:26){
    if (substr(df[i, 1], 1, 1) == letters[j]){
      append(eval(as.name(paste0("l_", LETTERS[j]))), df[i, 1])
    }
  }
}

The empty lists generate correctly and I don't get any errors, but all of my lists end up still being empty. This is leading me to think the issue is trying to call the lists using eval(as.name(paste0("l_", LETTERS[j]))). I know I could write if statements for each list rather than using a nested for loop, but this is cumbersome with 26 lists!

Even if you have a much more elegant method, I'd still like to know what I need to do for my current method to work as well as your own recommended solutions.

TIA!

Dee G
  • 133
  • 9

1 Answers1

1

OP is creating NULL list in a loop. It can be done in more easier ways as well

> l_A
list()

In the second loop, we need to get the value of the list created for appending and assign back to the same object

for (i in 1:nrow(df)){
  for (j in 1:26){
    if (substr(df[i, 1], 1, 1) == letters[j]){
      tmp <- paste0("l_", LETTERS[j])
      assign(tmp, append(get(tmp), df[i, 1]))
    }
  }
}

-checking

> l_A
[[1]]
[1] "as"

[[2]]
[1] "airplane"

[[3]]
[1] "as"

[[4]]
[1] "away"

[[5]]
[1] "airplane"

[[6]]
[1] "airplane"
...
akrun
  • 874,273
  • 37
  • 540
  • 662
  • That worked, thank you. How would I loop through the lists to assign the length of the longest list as a value to a variable "max"? E.g. if the longest list has 12 words, "max" would equal 12 – Dee G Dec 26 '21 at 19:05
  • @DeeG Not clear. Did you meant `max(lengths(mget(paste0("l_", LETTERS))))# [1] 18` – akrun Dec 26 '21 at 19:08
  • If instead of lists, I have character vectors (called ```v_A```, ```v_B```, etc.), how do I set the length of these vectors to all be the same as ```max(lengths(mget(paste0("v_", LETTERS))))``` ? – Dee G Dec 26 '21 at 19:44
  • @DeeG the same code should work i.e. `v_A <- 1:5; v_B <- 6:11; max(lengths(mget(paste0("v_", LETTERS[1:2]))))# [1] 6` – akrun Dec 26 '21 at 19:47
  • the code works, but I don't know how to then use that code and set the length of all the vectors to be the max length without doing it manually for each vector – Dee G Dec 26 '21 at 19:48
  • @DeeG lets say you get the `list` as `lst1 <- mget(paste0("v_", LETTERS[1:2])); mx <- max(lengths(lst1)); lst2 <- lapply(lst1, "length<-", mx);list2env(lst2, .GlobalEnv)` – akrun Dec 26 '21 at 19:50
  • Now, check the lengths again `lengths(mget(paste0("v_", LETTERS[1:2])))# v_A v_B 6 6` The `length<-` pads NA to the right wherever the length is short – akrun Dec 26 '21 at 19:51
  • Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/240458/discussion-between-dee-g-and-akrun). – Dee G Dec 26 '21 at 20:11