How to remove ^M Junk values while converting sas dataset to flat file

[PRAVIN_JAIN]

I need to convert a SAS dataset into text file. After converting into flat file, I am getting ^M  in one of the field (only single row). We are reading this file in python, and python treats it as a new line.

How this junk value can be handled at dataset level before converting into text file ?

In Dataset it simply looks like missing value.

In VI editor value look like this -

}0}0}^M}

[Oligolas]

Hi,

you might have leading not printable characters in your dataset but it's not clear to me.

Can you provide some raw data?

[Tom · Accepted solution]

What is ^M?  Do you mean the two byte string '^M'? Or perhaps the single character '0D'x (also know as a carriage return)?

Either way remove the values from the field in your dataset before writing the values to the text file.

Either change it to some other character.

var1 = translate(var1,' ','0d'x);

Or remove it.

var1 = compress(var1,'0d'x);

If something else is going on you need to provide more information.  For example the lines from the SAS log of the step that created the text file.

[jklaverstijn]

This looks like you crated a file wirth DOS line terminators (cariage return+linefeed) and looking at it in a Unix/Linux environment. Linux and UNIX consider the linefeed as the terminator, regarding and showing the carriage return (Ctrl-M) as part of your data.

We do not now how you create the file but possibly a filename statement is involved. In that case you can use the TERMSTR option to influence the choice of line terminator.

So determining on your situation the ^M may be a problem or not. Depending on what your plan is with the file if may be ok or it needs to go. SAS leaves it to you.

Hope this helps,

- Jan./