CODEWITHSUNDEEP
rdplyr
daisy_speed_713
daisy_speed_713

Reputation: 43

Apply function to each row in dataset when conditions are met

I have a boolean variable (parent_boolean). If parent_boolean = True then I want to apply a function to table_name, if parent_boolean = False then I want to apply a different function to table_name in the data set. What is the easiest way to do this in DPLYR? 

table_name    parent_table  parent_boolean column_count
 <chr>                                             <chr>                              <lgl>             <dbl>
1 table_1       table_1      TRUE                     1
2 table_2       table_1     FALSE                     3                                      

dp_child <- function(table_name, parent_table) { function runs SQL statement against database) 

dp_parent <- function(table_name) {function runs SQL statement against database)

I want to run the dp_parent function on each row that the table_name is the parent_table and I want to run the dp_child function on each row that the table_name is not the parent_table.

Upvotes: 0

Views: 955

Answers (2)

ThomasIsCoding
ThomasIsCoding

Reputation: 101247

Here is a base R solution using ifelse(), where a toy example is give as below

# two distinct function
f1 <- function(x) x**2
f2 <- function(x) -sqrt(x)

# apply function to column B according to value in column A
df <- data.frame(A = c(T,T,F,T),B = 1:4)
df <- within(df, C <- ifelse(A,f1(B),f2(B)))

such that

> df
      A B         C
1  TRUE 1  1.000000
2  TRUE 2  4.000000
3 FALSE 3 -1.732051
4  TRUE 4 16.000000

EDIT with the data frame in your post, maybe you can try the code below

result <- apply(df, 1, function(v) {
  if (v["parent_boolean"]) {
    dp_parent(v["table_name"])
  } else {
    dp_child(v["table_name"],v["parent_table"])
  }
})
df <- cbind(df,result)
## or 
# add_column(df, result)

Upvotes: 1

dc37
dc37

Reputation: 16178

You can use ifelse in while creating a new column using mutate:

A <- c(T,F,F,T,T)
B <- LETTERS[1:5]
df <- data.frame(A,B)

library(dplyr)
df %>% mutate(C = ifelse(A, paste0(B,1),paste0(B,"A")))

      A B  C
1  TRUE A A1
2 FALSE B BA
3 FALSE C CA
4  TRUE D D1
5  TRUE E E1

Does it answer your question ?

Upvotes: 1

Related Questions