How to use "regular expression" to delete "^M" character

[Jiangshanshan]

I use "vi" to open a .txt file from NT, and there are a lot of "^M" at the end of each line, how can i use regular expression to delete them.

[Denver Osborn · Accepted solution]

The best thing for this situation would be to use the dos2ux command.

# dos2ux nt_file.txt > new_file.txt

new_file.txt will not contain the ^M

-hope this helps.
denver

[Dan Hetzel]

Hi,

It's easy ! First of all hit ':' to go in 'ex' mode.
Then type the following:
1,$s///

That's it !

Dan

PS: allows you to type any kind of non-printable character in ex mode

[Steven Sim Kok Leong]

Hi,

Alternatively,

$ cat with_ctrl_M.txt | col -b > without_ctrl_M.txt

Hope this helps. Regards.

Steven Sim
Brainbench MVP for Unix Admin
http://www.brainbench.com

[Suhas_2]

My two cents ...
In addition to all the above ways...
You can do the following...

# strings nt_file.txt >new_file.txt

Hope this helps..;-)
Suhas Oka.

[Joseph C. Denman]

This is what I would do!!!!

FTP the file as ASCII from the NT box to the UNIX box. The ^M is a . Unix uses newlines. If you FTP the file as ASCII, conversion is automatic.

Hope this helps....

[Andreas Voss]

Hi,

within vi you can remove ^M with:

1G
!G!dos2ux

1G will go to the first line of the document (entering is not visible)
!G will select the document form top to bottom (entering is not visible, after the G a ! occours at the bottom line)
dos2ux will remove the ^M from the selected lines.

Regards

[Stefan Schulz]

The easiest thing is to use the dos2ux command as Denver Osborn mentioned. Of course you should use the ux2dos command if you copy this file to a Windowsbox again.

If for some reason you can't use dos2ux i would prefer the vi with substitiution like Dan posted. In addition to the post from Dan i would suggest using the following substitution:

1,$s///g

With an additional "g" at the end. Which means "global". If the String you want to replace can be found several times in a line this global option will replace them all. Without this option only the first apperance in this line will be replaced.

Hope this helps and my english was good enough .

Stefan

[IWANIEC, EDOUARD]

And what about this (in vi):

%s/.$//

Laszlo

[Andre Lattier]

you don't need a regular expression, you can use tr :
cat myfile | tr -d '\015'