Merge multiple lines into one record

[A1ex777]

Hello.

I have dataset with one column that contains a lot of records.

For example the first 3:

abc

 def

    ghi

How can i merge it into 1 record? -

abc def ghi

The important thing also is that all blanks must be saved.

[Tom]

Which spaces need to be preserved?  Just the leading spaces?

data want;
  set have end=eof;
  length new_var $300 ;
  new_var = trimn(new_var) || var;
  if eof;
run;

Or the trailing spaces that are added to pad the string value to the full length of the variable?

data want;
  set have end=eof;
   length new_var $300 ;
   substr(new_var,(_n_-1)*vlength(var)+1,vlength(var))=var;
  if eof;
run;

Do you really only want one observation?  Why ? Or is there some grouping variable?

data want;
  set have end=eof;
  by id;
  length new_var $300 ;
  if first.id then new_var=' ';
  new_var = trimn(new_var) || var;
  if last.id;
run;

[A1ex777]

@Tom Tom thanks a lot for quick answer.
Yes i need both type of spaces.

There's no grouping variable. I just have dataset with column that contains separate records with html text. So i need to store it in one record.

[Tom · Accepted solution]

That seems silly.  Just leave the HTML text as is and then when writing it to an actual HTML file just don't instead any line breaks.

So if you have a dataset named HAVE with a variable named TEXT that is defined to hold 100 bytes and you want to use it to create a one line HTML file named myfile.html use something like:

data _null_;
  file 'myfile.html' recfm=n ;
  set have;
  put text $char100.;
run;