R data.table fread bug

February 28, 2022

I’m trying to understand data.table’s fread function. I am hoping to read in a csv file, starting at the row where ‘!’ first appears. The number after "!" changes from table to table, but the "!" is a constant feature.

Here’s a version of the data I’m using. When I copy and paste this into a csv file, it works.

113
$LastIndex,AGE_DFN_MORT
!2,PROD_CODE,0,1,2,3,4,5
*,"CDADOA",100,100,100,100,100,100

Here’s my attempt:
fread("<file.location>.csv", skip = "!")

What I would expect to happen is that the very first column name would be !2, followed by PROD_CODE and so on, however it instead populates all column names with V1, V2, V3 etc.

Have I misunderstood something core about the ‘skip’ argument to the fread function?

EDIT: Thanks – the issue is the absence of the ‘header’ argument

>Solution :

I tried this, it works:

myData <- 
'113
$LastIndex,AGE_DFN_MORT
!2,PROD_CODE,0,1,2,3,4,5
*,"CDADOA",100,100,100,100,100,100'


myData <- fread(myData, skip = '!2', header = TRUE)

myData
#    !2 PROD_CODE   0   1   2   3   4   5
# 1:  *    CDADOA 100 100 100 100 100 100