Use repeated colnames

[MarcoPortmann]

I think in combination with group headers it is common to have repeated colnames. What I want to achieve is "a" header a with subheader sub1 & sub2 and a header "b" with sub1 & sub2. Howver, I don't see a way to feed them into tinytable. dplyr::rename does not allow duplicates and, as far as I know, tt doesn't have an argument to set colnames.

 tibble(a_sub1 = 1, a_sub2 =2, b_sub1 = 3, b_sub2 = 4) %>% 
   tt() %>% 
   group_tt(j = list(a = 1:2, b = 3:4))

[vincentarelbundock]

Sorry, I don't understand what you mean. Could explain differently and ideally show me an example of a table like the one you want?

[MarcoPortmann]

I'd like to achieve an output like this:

tibble(a_sub1 = 1, a_sub2 =2, b_sub1 = 3, b_sub2 = 4) %>% 
  rename(sub1 = "a_sub1", sub2 = "a_sub2", "sub1 " = "b_sub1", "sub2 " = "b_sub2")  %>% 
  tt() %>% 
  group_tt(j = list(a = 1:2, b = 3:4))

Howver, it would be nice to achieve it without the workaround of extra spaces in "sub1 " and "sub2 " which are required since a tibble cannot have two columns with identical colnames.

[vincentarelbundock]

tinytable already supports duplicated column names without any problem:

library(tinytable)
data.frame(sub1 = 1, sub2 = 2, sub1 = 3, sub2 = 4, check.names = FALSE) |>
    tt() |>
    group_tt(j = list(a = 1:2, b = 3:4))

a		b
sub1	sub2	sub1	sub2
1	2	3	4

I’m closing this because this is not a limitation of tinytable, but rather a feature request for tibble() to allow duplicated column names.

[MarcoPortmann]

Well, I'd argue, it makes sense that a tibble doesn't allow duplicated column names because a tibble is employed for calculus, data storage etc. and it should be machine-readible. tt aims at prodcuing human-friendly output and offers group_tt to contain a part of the column name. In this context it makes sense, in my opinion to have dpulicated headers. Btw., I encountered the issues as I was translating code from kable to tt. kable offers the argument col.names to set the column names.
But I guess, I could also live with the workaround that I used above.

[vincentarelbundock]

Actually, it looks like tibble does in fact support repeated names, and has a specific argument for it:

tibble(sub1 = 1, sub2 = 2, sub1 = 3, sub2 = 4, .name_repair = "minimal") |>
    tt() |>
    group_tt(j = list(a = 1:2, b = 3:4))

[MarcoPortmann]

You're right. Interesting. It seems that .name_repair = "minimal" can be provided to tibblebut not to other functions like rename. Renaming via colnames(tab) <- works as well.

I guess the issue can be closed.