Looping through several levels

Question

I have a naïve understanding of For loops in R. I would appreciate your help to learn it better. I am working on a project with several tables of data in it. imagine the following structure: dictionary$table$variables$type

In the last level (i.e., type), the values are FALSE and/or TRUE. I need to show for which variables across all the tables the value of type equals only TRUE or only FALSE or a combination of both.

I found this function to suit my purpose: for instance for a table called HARIX, all values equal FALSE.

unique(dictionary$HARIX$variables$type)
[1] FALSE

Now I need to turn it into a loop so that it does the job for all the tables included in the dictionary. I don't know how to write this command. I wrote the one below but it doesn't reach the write level I need (i.e., type).

value <- list()
for(table_name in project$name){
  value[[table_name]] <- unique(dict[[table_name]]$variables$type)
  print(val[[table_name]])
}

I really appreciate your help. If you also have any advice on how to use the lapply function for this purpose, I would love to learn that too.

Answer 1

Get rid of the for loop and use the map function from the purrr package:

Data sample

Remember you have to provide a data sample in order to explain your problem better...

v1 <- c(F, F, F)
v2 <- c(F, T, F)
v3 <- c(T, T, T)

dictionary <- list(
  HARIX = list(var1 = list(type = v1), 
               var2 = list(type = v2),
               var3 = list(type = v3)),
  OTHER = list(var4 = list(type = v3), 
               var5 = list(type = v1),
               var6 = list(type = v2)))

Solution

require(purrr)

map(dictionary, ~ map(., ~ map(., unique)))

Output

$HARIX
$HARIX$var1
$HARIX$var1$type
[1] FALSE


$HARIX$var2
$HARIX$var2$type
[1] FALSE  TRUE


$HARIX$var3
$HARIX$var3$type
[1] TRUE



$OTHER
$OTHER$var4
$OTHER$var4$type
[1] TRUE


$OTHER$var5
$OTHER$var5$type
[1] FALSE


$OTHER$var6
$OTHER$var6$type
[1] FALSE  TRUE

Here is the map function documentation

lapply bonus

You have requested a lapply solution:

  lapply(dictionary,
         function(word) lapply(word,
                               function(var) lapply(var, unique)))

Explanation

You have to go down the list till you reach the right level. So you have to use nested map or nested lapply.