Compress empty space of all variables

[Yegen]

Is there any way to compress blank spaces of variable names of all variables without listing the variables individually?

[Reeza · Accepted solution]

You can use _character_ to reference all character variables in an array and compress the spaces. Note that the spaces are determined by the length of the variable, so those spaces will still exist.

array _test(*) $ _character_;

do i=1 to dim(_test);
    _test(i) = compress(_test(i));
end;

---

@Yegen wrote:

Is there any way to compress blank spaces of variable names of all variables without listing the variables individually?

---

[ballardw]

---

@Yegen wrote:

Is there any way to compress blank spaces of variable names of all variables without listing the variables individually?

---

Spaces in variable names or values of variables?

[Yegen]

Perfect thank you. 

[Yegen]

@ballardw spaces in variable names. 

The solution proposed by @Reeza works well with character variables, but I am still facing issues with date variables. In particular, I have imported excel files (using a macro) and for some reason blank spaces were added to the variable names. I cannot identify the number of spaces even if I use a code as follows: 

proc contents
     data = work.temp_data
          noprint
          out = data_info
               (keep = name varnum);
run;

[Yegen]

@Reeza and @ballardw I have attached a few figures that might provide some hints with the error I am getting:

• I am sorting the data by using the filter. 

•  I am trying to edit the filter 

• Now I am just re-running the same filter that SAS Studio gave me, but I am getting an error.  

[Yegen]

Here is the data description: 

[ballardw]

---

@Yegen wrote:

@ballardw spaces in variable names. 

The solution proposed by @Reeza works well with character variables, but I am still facing issues with date variables. In particular, I have imported excel files (using a macro) and for some reason blank spaces were added to the variable names. I cannot identify the number of spaces even if I use a code as follows: 

proc contents
     data = work.temp_data
          noprint
          out = data_info
               (keep = name varnum);
run;

---

It sounds like you may have imported the data with the option validvarname set to "any".

Run this code to see if that is the case:

proc options option=validvarname;
run;

If the LOG shows a result similar to

 VALIDVARNAME=ANY  Specifies the rules for valid SAS variable names that can be created and
                   processed during a SAS session.

then you can likely "fix" the problem by setting

option validvarname=V7;

and reimporting.

[Yegen]

@ballardw, this is the macro I was using to import the excel files:

%macro merging;
%do i=1 %to 8;


PROC IMPORT DATAFILE="/home/institution/user/SDC_M_A/sdc_oct5_&i..xlsx"
	DBMS=XLSX
	OUT=WORK.merging&i;
	GETNAMES=yes;
RUN;

%end;
%mend;
%merging;

If I use your code, this is the log file I am getting: 

90         
 91         proc options option=validvarname;
 92         run;
 
     SAS (r) Proprietary Software Release 9.4  TS1M5
 
  VALIDVARNAME=ANY  Specifies the rules for valid SAS variable names that can be 
                    created and processed during a SAS session.
 NOTE: PROCEDURE OPTIONS used (Total process time):
       real time           0.00 seconds
       cpu time            0.00 seconds
       

[Yegen]

You are 100% right, @ballardw. When I added the

option validvarname=V7;

 statement in the macro the issue was fixed. 

I assume the import function was just adding random blank spaces when you don't specify the validvarname? 

[Reeza]

It’s usually not random, it’s likely in the excel file in some form. 

[Yegen]

That makes sense, thanks @Reeza