Dynamically read/process csv thousand separator

Asked May 16 '17 at 13:24

Active May 16 '17 at 13:43

Viewed 29 times

I got data (many columns), some containing numeric data, others are string. The numeric data is stored with a thousand separator, hence the data looks like:

col1; col2; col3; col4; col5
100410; "abc"; "def"; 100.277; 30.997
100411; "abc"; "def"; 130.277; 30.927

The following will read the data correctly, but with col4 and 5 as character type.

df <- read.table(pathToFile, stringsAsFactor=F, sep = ";")
str(df)

 $ col1  : int  100410 100410 ...
 $ col2  : chr  "abc" "abc"
 $ col3  : chr  "def" "def"
 $ col4  : chr  "100.277" "130.277"
 $ col5  : chr  "30.997" "30.927"

Now: how can I dynamically convert the character columns which do contain numeric data (col4 and col5), but keep the string columns the way they are (col2 and col3)?

edited May 16 '17 at 13:43

asked May 16 '17 at 13:24

Dendrobates

3,294
6
37
56

1

You have a bunch of commas in your example, are they supposed to be `;`? – Axeman May 16 '17 at 13:38
1

The duplicate has "," as a thousands separator but the same techniques would work for "." – MrFlick May 16 '17 at 13:38
1

Use `readr::read_csv2` and it works automatically. – Axeman May 16 '17 at 13:41
You're right @Axeman, edited the question. – Dendrobates May 16 '17 at 13:44
1

@Axeman, readr was Indeed exactly what I was looking for! Works as expected/hoped for! Thx for your help ;) – Dendrobates May 16 '17 at 14:05

Dynamically read/process csv thousand separator

0 Answers0