使用全局变量作为列名时使用 data.table 更新列

Question

I am attempting to update a column in a data.table based on if the values in the column match an external variable. The issue I am running into is using a global variable as a column name in data.table syntax.

Specifying the column name explicitly works as expected and updates the column.

dt[,`1:4880` := ifelse(`1:4880`==allele, 2, 1)]

However, I am unable to update the column when referring to it by a variable:

colname="1:4880"
dt[,..colname := ifelse(..colname==allele, 2, 1)]

Instead of updating the values in the column described by the variable name, the output is written to df$..colname instead. Strangely, the output is correct so the ifelse function is working as intended, it's just writing the output to the wrong column name.

Is there a way to make the:= operator recognize a column name specified as a variable?

Answer 1

On the lhs , we can wrap with () to evaluate the value and either use get or specify it in .SDcols

dt[, (colname) := ifelse(.SD[[1]] ==allele, 2, 1), .SDcols = colname]

It is not clear whether the allele is another column or an object created with some value