removing ^M from a csv file

Question

I have a problem when I try to remove ^M from a csv

When I type vim or vi file.csv, I get

A, TK,2015-04-06,14.4^M,14.7,10.0,0.0,54.0^M,13.3^M,135.0^M,33.8
B, NV,2015-04-06,14.4^M,14.7,5.4,0.0,47.0^M,14.8^M,97.0^M,31.3

I have tried with

tr -d '^M' < file.csv > file2.csv

But it doesn't remove, also with sed.

Answer 1

You could use dos2unix command which is provided to do that.

Using GNU/sed just for fun :

sed -i -e "s/\r//g" file

Using tr :

tr -d '\r' <file1 >file2

Answer 2

Where did you get that file from? It looks like an old System X file from a Mac. The old pre-OSX OS used <CR> as line endings. Unix uses <LF> , and Windows/DOS uses <CRLF> .

Do you have dos2unix . This program can convert line endings from Unix/Linux, DOS/Windows, or System X Macs to any of the formats you want. In your file, I take it you need to convert the ^M which are Control Ms and not a Caret-M to NL characters.

Answer 3

You can try:

perl -pE 's/(\^M|\r)//g' < file >file2

should remove

• the literal ^M - sequence of two characters ^ and M
• and the ^M as \\r character.