结合data.frame和data.frames列表，在R中没有公共变量

Question

I have a data frame (D) and a list of data frames (L) that I want to combine into a new data frame. There is one row in D for every data frame in L, and I want to join these data together so that each row in D is matched with the corresponding data frame in L and replicated across each row. The data frames in L have varying row numbers, but they all have the same columns and could easily be combined into a single data frame (eg, using plyr::rbind.fill ). There are no common variables between D and the data frames in L - the only way I know which rows go together is by the order in which they appear in D and L.

Here is toy data with the same structure as my data:

# the data frame
D <- data.frame(name = c("john","sally","ben"), age = c(23, 31, 27))

# the list of data frames
john <- data.frame(attempt = 1:3, result = c("fail","fail","fail"))
sally <- data.frame(attempt = 1, result = c("success"))
ben <- data.frame(attempt = 1:5, result = c("fail","fail","success","fail","success"))
L <- list(john, sally, ben)

The dumb way I have tried to do this is with a for loop:

# loop to combine data frame and list
new_D <- data.frame()
for (i in 1:nrow(D)) {
    add <- cbind(D[i,], L[[i]])
    new_D <- rbind(new_D, add)
}

It works, but it is very slow and my files are quite large, so it is not practical. What is a cleaner and more efficient way to do this in R?

Answer 1

Name the list elements, convert the list to a single data.table with an index column ("name"), join with the original data on the "name" column:

names(L) <- D$name
D2 <- data.table::rbindlist(L, use.names = TRUE, idcol = "name")  
D2[D, on = "name"]
#     name attempt  result age
# 1:  john       1    fail  23
# 2:  john       2    fail  23
# 3:  john       3    fail  23
# 4: sally       1 success  31
# 5:   ben       1    fail  27
# 6:   ben       2    fail  27
# 7:   ben       3 success  27
# 8:   ben       4    fail  27
# 9:   ben       5 success  27

Answer 2

We can do a split by sequence of row and then with Map cbind the datasets

do.call(rbind, Map(cbind, split(D, seq_len(nrow(D))), L))

---

Or set the names of 'L' with the paste ed rows of 'D', bind the rows and separate into two columns

library(tidyverse)
do.call(paste, c(D, sep = ",")) %>%
     set_names(L, .) %>%
     bind_rows(.id = 'grp') %>% 
     separate(grp, into = c('name', 'age'))