Set a new header to the table dynamically from the labels of banner

I have a function which is creating table one by one for the list of banner. but now want to show them as binding as cbind.

I am able to bind the table by columns but not able to make banners as headers for table. also I want to remove names like n.x, n.y . just wanted to keep them N and i needed a flextable output

currently table is coming like below

library(knitr)
library(tidyverse)
library(flextable)

banner <- c("All","Other")
t1 <- structure(list(` ` = c("CA", "USA","UK", "GER", "Italy","France", "China"),
                                               Local = c("38%", "58%","71%", "78%", "91%", "91%", "84%"),
                                               Outsider = c("43%", "28%", "29%"," 9%", " 7%", " 4%", " 5%"), 
                                               Mixed = c("19%","13%", " 0%", "13%", " 1%", " 5%", "11%"),
                                               N = c("9999", "9999","9999", "9999", "9999", "9999", "9999")), row.names = c(NA, -7L), class = "data.frame")

t2 <- structure(list(` ` = c("CA", "USA","UK", "GER", "Italy","France", "China"),
                     Local = c("71%", "93%","96%", "96%", "96%", "96%", NA),
                     Outsider = c(" 7%", " 4%", " 4%"," 0%", " 0%", " 0%", NA),
                     Mixed = c("21%"," 4%", " 0%", " 4%", " 4%", " 4%", NA),
                     N = c("2800", "2800","2800", "2800", "2800", "2800", NA)), row.names = c(NA, -7L), class = "data.frame")
tl <- list(t1,t2)


fun1 <- function(tl,banner){
  t_list1<-list()
  for (i in 1:length(banner)) 
  {
    t_list <- list()
    t_list1[[i]] <- tl[[i]]
  }
  t1<-Reduce(function(...)merge(...,all=TRUE,by=" "),t_list1)
  
  t1 <- t1 %>% flextable()
  t1
}

fun1(tl=tl,banner = banner)

the output should be look llike below

Solution

One option to fix both your issues would be to first rename the column names of your sublists and add the banner as a prefix. Additionally I convert your lists to dataframe at this step. After doing so you column names are unique and no .x or .y will be added when using merge. And as a second benefit we can now utilize separate_header() to split the column names to achieve your desired final result:

library(flextable)

tl <- list(t1, t2)
names(tl) <- banner

tl <- Map(
  function(x, y) {
    names(x)[-1] <- paste(y, names(x)[-1], sep = "_")
    data.frame(x, check.names = FALSE)
  },
  tl, names(tl)
)

fun1 <- function(tl) {
  t1 <- Reduce(function(...) merge(..., all = TRUE, by = " "), tl)

  t1 |> 
    flextable() |> 
    separate_header()
}

fun1(tl = tl)

EDIT Of course could you move Map inside your function to do the renaming.

library(flextable)

tl <- list(t1, t2)

fun1 <- function(tl, banner) {
  tl <- Map(
    function(x, y) {
      names(x)[-1] <- paste(y, names(x)[-1], sep = "_")
      data.frame(x, check.names = FALSE)
    },
    tl, banner
  )
  
  t1 <- Reduce(function(...) merge(..., all = TRUE, by = " "), tl)

  t1 |> 
    flextable() |> 
    separate_header()
}

fun1(tl = tl, banner = banner)