Clever Geek Handbook

Setting read_excel to guess some column types

I am using read_excel

from a batch readxl

to read a file where there are 3 columns that I want to coerce as text, and the rest I'm happy to let the read_excel

type guess. Can I do it?

I tried to use col_type

to set the columns I want to be text

and the rest is like blank

, but this results in the columns being skipped blank

. I tried using NA

instead blank

, but doing so all columns will be bound to text

, when read_excel

otherwise I would read as "number".

My code was

col_filter <- rep('blank', 14)
col_filter[1, 3, 7] <- 'text'
read_excel(file, sheet, col_type=col_filter)

14 is the number of columns in the excel file, 1 3 7 are the columns I want to read as text. Bonus if I can do this without knowing the number of columns beforehand (without reading the file first to check the number of columns).

+3


source to share


1 answer


You can use this:



col_filter <- readxl:::xlsx_col_types(path = file, n = 100,nskip = 1)
col_filter[1, 3, 7] <- 'text'
read_excel(file, sheet, col_type=col_filter)
+3


source






More articles:


All Articles