Add also the empty rows in a data frame construction - r

Question

I want to construct a data frame based on two data frames
Here it is an example
#toy example
name <- c("Li", "Pedro", "Dave")
age <- c(20, 30, 40)
d1 <- cbind.data.frame(name, age)
name <- c("Pedro", "Dave", "Grace")
fav_col <- c("red", "blue", "yellow")
lastname <- c("Sanchez", "Stone", "Flint")
fav_food <- c("pizza", "hamburguers", "salad")
d2 <- cbind.data.frame(name, fav_col, lastname, fav_food)
d1$name <- as.character(d1$name)
d2$name <- as.character(d2$name)
cols <- c()
for(i in 1:nrow(d1)) {
some <- dplyr::filter(d2, name==d1$name[i])
cols <- rbind.data.frame(cols, data.frame(some$name, some$fav_col, some$fav_food))
}
Doing this I am obtaining a data frame called "cols" and looks like this:
some.name some.fav_col some.fav_food
Pedro red pizza
Dave blue hamburguers
But what I want is
some.name some.fav_col some.fav_food
NA(or empty) NA(or empty) Na(or empty)
Pedro red pizza
Dave blue hamburguers
The first iteration when i = 1 must produce an empty exit because there is no Li in the second data frame, and I want this empty space in my data frame. Do you know how I could get this?
At the end I want to add the second and third columns of "cols" to "d1" to get:
name age fav_col fav_food
Li 20 NA (or empty) NA (or empty)
Pedro 30 red pizza
Dave 40 blue hamburguers
Also I don't want the empty spaces that the second data frame could produce like this:
name age fav_col fav_food
Li 20 NA NA
Pedro 30 red pizza
Dave 40 blue hamburguers
Grace NA yellow salad
I just want to merge the tables keeping only the names of the first data frame and add the two extra columns. I would appreciate any help

Answer 1

You can use union_all from dplyr.
library(tidyr)
library(dplyr)
df <- union_all(d1, d2) %>%
mutate_if(is.factor, as.character) %>% # only required when your text columns are
group_by(name) %>% # identified as factor and not character.
summarise_all(max, na.rm = TRUE) %>% # Because max only works on numeric or char
ungroup
df
# # A tibble: 3 x 4
# name age fav_col lastname
# <chr> <dbl> <chr> <chr>
# 1 Dave 40. blue Stone
# 2 Li 20. NA NA
# 3 Pedro 30. red Sanchez
To get your desired output, you can add drop_na to the chain.
df %>% select(name, fav_col) %>% drop_na
# # A tibble: 2 x 2
# name fav_col
# <chr> <chr>
# 1 Dave blue
# 2 Pedro red

Answer 2

I used this, and I added paste and collapse just in case that someone needs to add results from different cells in just one.
f_add_col <- function(vec) {
add_col <- dplyr::filter(d2, name==vec[1])
return (paste(add_col$fav_col, collapse = "|"))
}
cbind.data.frame(d1, fav_col=apply(d1, 1, f_add_col))
Then I did the same for the column fav_food.