Joining on multiple variables within a function

I am working in R.

I have two data sets - data1 and data2.

data1 <- data.frame(region_name = c("North", "North", "West"),  
                    type = c("big", "small", "big"), 
                    gamma_rate=7:9)


data2<- data.frame(region_name = c("West", "West", "East"),  
                   type= c("small", "big", "big"), 
                   beta_rate=7:9)

Both of these data sets have columns called "region_name" and "type" in them.

I want to left_join data 2 onto data 1, by "region" and "type", but within a function. If I was doing it without a function, this would be my output:

data_final <- data1 %>% 
  left_join(data2, by = c("region_name" = "region_name", "type" = "type"))

This my function:

my_function(group1, group2) {
  data_final <- data1 %>%
                left_join(data2, by = c({{group1}} = {{group1}}, {{group2}} = {{group2}}))
}

output <- my_function(region_name, type)

I know the bit in the "by...." argument is incorrect in the function. Can anyone help with out how to correct it?

But it looks like it is just for one join variable?

CodePudding user response：

If you truely want it as a function, I would pass your data in as well, then a function becomes universal and you can use it no matter how your two tables are called.

data1 <- data.frame(region_name = c("North", "North", "West"),  
                    type = c("big", "small", "big"), 
                    gamma_rate=7:9)


data2<- data.frame(region_name = c("West", "West", "East"),  
                   type= c("small", "big", "big"), 
                   beta_rate=7:9)


my_function <- function(join, df1, df2) {
  df1 %>% left_join(df2, by = join)
}

my_function(data1, data2, join = c("region_name", "type"))

  region_name  type gamma_rate beta_rate
1       North   big          7        NA
2       North small          8        NA
3        West   big          9         8

CodePudding user response：

library(dplyr)

data1 <- data.frame(region = c("A", "B", "C"), 
                    type = c("A", "B", "C"), 
                    value = c(1, 2, 3))


data2 <- data.frame(region = c("A", "B", "C"), 
                    type = c("A", "D", "C"), 
                    value = c(4, 5, 6))

my_join_function <- function(group1, group2){
  data1 %>% 
    left_join(data2, 
              by = c(group1, group2))
  
} 

my_join_function('region', 'type')

region	type	value.x	value.y
A	A	1	4
B	B	2	NA
C	C	3	6

CodePudding user response：

A data.table solution

library(data.table)

Dummy data

data1 <- data.table(region_name = c('a', 'b')
                    , type = 1:2
                    ); data1
region_name type
1:           a    1
2:           b    2

data2 <- data.table(region_name = c('b', 'c')
                    , type = 2:3
                    ); data2
region_name type
1:           b    2
2:           c    3

function

my_function <- function(x)
{
  return(data2[data1, on=(x)])
}

run function

my_function(c('region_name', 'type'))
region_name type
1:           a    1
2:           b    2