Issue when joining data.tables with greater than and lesser than signs in column names

Question

When column names have > (greater than) or < (lesser than) sign and I try to join on these columns, the column with the sign gets trimmed and I get an error like this one:

Error in colnamesInt(x, names(on), check_dups = FALSE) : argument specifying columns specify non existing column(s): cols[1]='COLUMN WITH'

Code to reproduce the issue

library(data.table)

table_a <- data.table(
  `COLUMN WITH >FAILS` = character(),
  BLA = integer(),
  BLO = integer()
  
)

table_b <- data.table(
  `COLUMN WITH >FAILS` = character(),
  BLA = integer(),
  BLI = integer(),
  BLU = integer()
  
)

on = setdiff(names(table_a), c("BLA", "BLU", "BLO"))

table_a[table_b, on = on]

My scrappy workaround was to replace the special character before the join and put the original column name after the join, it works but doesn't look very elegant. I don't know if that is a known issue, I haven't found any documentation about this. Can you think of a better way of solving this?

Answer 1

Backticks inside quotes allow to use reserved symbols in the join :

table_a[table_b, on = "`COLUMN WITH >FAILS`"]

Empty data.table (0 rows and 6 cols): COLUMN WITH >FAILS,BLA,BLO,i.BLA,BLI,BLU

Or more generally:

table_a[table_b, on = paste0("`",on,"`")]

Empty data.table (0 rows and 6 cols): COLUMN WITH >FAILS,BLA,BLO,i.BLA,BLI,BLU