Re-position multiple data frame columns at once.
Extended wrapper around dplyr::relocate() to support relocation of multiple columns to multiple respective positions at once.
Arguments
- df
A data.frame or tibble with columns that shall be relocated.
- move
The columns that shall be moved. Can be column names, indices, <
tidy-select> or <data-masking> expressions.- before
The columns, where the respective
movecolumn shall be placed before. Can be column names, indices, <tidy-select> or <data-masking> expressions.- after
The columns, where the respective
movecolumn shall be placed after. Can be column names, indices, <tidy-select> or <data-masking> expressions.
Details
All column (selection) indices are converted to names, to allow multiple relocation operations for the same column without index confusion.
Examples
df <- data.frame("a" = 1:2, "b" = 3:4, "c" = 5:6, "d" = 7:8)
df
#> a b c d
#> 1 1 3 5 7
#> 2 2 4 6 8
# Move column "a" after "b" and column "c" after "d"
relocate_multi(df, move = c("a", "c"), after = c("b", "d"))
#> b a d c
#> 1 3 1 7 5
#> 2 4 2 8 6
# The same operation with indices
relocate_multi(df, move = c(1, 3), after = c(2, 4))
#> b a d c
#> 1 3 1 7 5
#> 2 4 2 8 6
# The same operation with tidy-select expression
df |> relocate_multi(move = tidyselect::all_of(c(1, 3)), after = tidyselect::all_of(c(2, 4)))
#> b a d c
#> 1 3 1 7 5
#> 2 4 2 8 6
# Use data masking to move "b" before "a"
df |> relocate_multi(move = rlang::expr(b), before = rlang::expr(a))
#> b a c d
#> 1 3 1 5 7
#> 2 4 2 6 8
# Relocation operations are sequential:
# Move "c" after "d" -> abdc
# Move "a" after "c" -> bdca
# Move "a" after "d" -> bdac
relocate_multi(df, move = c("c", "a", "a"), after = c("d", "c", "d"))
#> b d a c
#> 1 3 7 1 5
#> 2 4 8 2 6