Updating dataframe based on user input

Question

I have the following dataframe

v    w    e   
dog  cat  -> 
frog cat  -> 
frog dog  -> 
frog fish ->
fish cat  ->

This is the function I wrote. It will output an updated dataframe based on the user input. The inputs V, W, and E can be a single string or a list. But V, W, and E must have the same length.

convert_df <- function(df, V, W, E) {
  df %>%
    dplyr::mutate(flag = (v %in% V & w %in% W | w %in% V & v %in% W),
                  txt = ifelse(flag, paste(V,E,W), paste(v,e,w)))
}

For example, this would be my expected output for the following. As you can see I want to change the txt column dog <-> frog and cat <-> dog :

> V <- c('dog', 'cat')
> W <- c('frog', 'dog')
> E <- c('<->', '<->')
> convert_df(df,V,W,E)

v    w    e   
dog  cat  <-> 
frog cat  -> 
frog dog  <-> 
frog fish ->
fish cat  ->

However my function will output

dog  cat  <-> 
frog cat  <-> 
frog dog  <-> 
frog fish ->
fish cat  ->

How can i fix my function? The indexes for V,W, and E are aligned to what i want to change in the dataframe. And my current function just checks if the v and w are in V and W regardless of the index alignment.

UPDATE:

Here I switch the order of V and W for the first element. I want to make my function dynamic so that the following would give the expected result as well:

> V <- c('frog', 'cat')
> W <- c('dog', 'dog')
> E <- c('<->', '<->')
> convert_df(df,V,W,E)

v    w    e   
dog  cat  <-> 
frog cat  -> 
frog dog  <-> 
frog fish ->
fish cat  ->

DaveArmstrong · Accepted Answer · 2021-11-02 16:18:57Z

1

I would probably do this with a join:

df <- tibble::tribble(
  ~v,   ~w,   ~e,   
"dog","cat", "->",
"frog","cat", "->", 
"frog","dog", "->",
"frog","fish","->",
"fish","cat", "->") 


v <- c('dog', 'frog')
w <- c('cat', 'dog')
E <- c('<->', '<->')

newdat <- do.call(data.frame, list(v=v,w=w,E=E))

df <- left_join(df, newdat) %>% 
  mutate(e = ifelse(is.na(E), e, E)) %>% 
  select(-E)


# # A tibble: 5 × 3
#    v     w     e    
# <chr> <chr> <chr>
# 1 dog   cat   <->  
# 2 frog  cat   ->   
# 3 frog  dog   <->  
# 4 frog  fish  ->   
# 5 fish  cat   ->

answered Nov 2, 2021 at 16:18

DaveArmstrong

22.5k2 gold badges16 silver badges30 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

r2evans Over a year ago

Is there a reason do do.call(data.frame, list(v=v,w=w,E=E)) when you can do just data.frame(v=v,w=w,E=E)?

r2evans Over a year ago

Also, you can replace your ifelse with coalesce(E, e), a handy NA-handling function in dplyr.

Eisen Over a year ago

I just updated my question - I also want the function to work if the inputs of v and w are swapped by the user.

Martin Gal Over a year ago

@Eisen Just do a double left_join:

df %>% left_join(df2, by = c("v" = "V", "w" = "W")) %>% left_join(df2, by = c("v" = "W", "w" = "V")) %>% mutate(E = coalesce(E.y, E.x, e))

.

Collectives™ on Stack Overflow

Updating dataframe based on user input

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related