SAS Rename With Capitals

[bradklassen]

Hi,

I am having a problem renaming columns with capitals. SAS will only capitalize the variables that are different on the left and right of the equal signs. I would like to rename a variable exactly the same just with capitals. Ex. I would like run_number to become Run_Number. However, SAS will not recognize the command and leave the variable name as run_number.

I have provided the code below.

data temp.GPS_Edit5 (rename =(lightbeam=_Lightbeam run_number=Run_Number last_name=Last_Name first_name=First_Name
       discipline=Discipline distance_m=_Distance_m speed_kmh=_Speed_km_h roll_angle=_Roll_angle lightbeam_time_s=_LB_time_s           corr_z=Corr_z));
set temp.GPS_Edit4;
run;

Thanks!

[Kurt_Bremser]

What for? For code, run_number and Run_number are 100 % equal.

If you want to have something nice to read, put in in a label.

Allowing upper- and lowercase without making the whole language case-sensitive is one of the blunders SAS has committed over the years, IMO.

C, C++, C#, Java, Python, Go, Ruby are all case sensitive.

[bradklassen]

I would like it in that format for the purpose of exporting it to Tableau. I already have some tableau calculated fields created that reference Run_Number rather than run_number. Tableau will not identify that they are supposed to be the same columns.

[Kurt_Bremser]

---

@bradklassen wrote:

I would like it in that format for the purpose of exporting it to Tableau. I already have some tableau calculated fields created that reference Run_Number rather than run_number. Tableau will not identify that they are supposed to be the same columns.

---

Oh yes. See the second part of my previous post.

OTOH, SQL and PL/1 (on which the data step language is modeled) are also case insensitive.

[Tom]

Did you try renaming on the way in instead of on the way out?

data have;
  run_number=1;
  length name $32 ;
  name=vnamex('run_number');
  put (_all_) (=);
run;
data want ;
  set have(rename=(run_number=Run_Number));
  name=vnamex('run_number');
  put (_all_) (=);
run;

296  data have;
297    run_number=1;
298    length name $32 ;
299    name=vnamex('run_number');
300    put (_all_) (=);
301  run;

run_number=1 name=run_number
NOTE: The data set WORK.HAVE has 1 observations and 2 variables.


302  data want ;
303    set have(rename=(run_number=Run_Number));
304    name=vnamex('run_number');
305    put (_all_) (=);
306  run;

Run_Number=1 name=Run_Number
NOTE: There were 1 observations read from the data set WORK.HAVE.
NOTE: The data set WORK.WANT has 1 observations and 2 variables.

[VDD]

I agree with @Kurt_Bremser use labels that is why SAS has labels, so you can see the names without the underscores and uppercase.

 those labels also help those who don't know the data putting the Caps in here and there will not assist in the final outcome.

Use the right tool for the job.

[FreelanceReinh]

Hi @bradklassen,

If you only rename variables, you don't need to read the data with a SET statement, which can be time-consuming for a large dataset. It's more efficient to use PROC DATASETS.

Example:

data GPS_Edit4;
run_number=1;
corr_z=2;
speed_kmh=3;
run;

proc datasets lib=work nolist; /* in your case: lib=temp */
modify GPS_Edit4;
rename run_number=Run_Number corr_z=Corr_z speed_kmh=_Speed_km_h;
quit;

If you need the SET statement anyway, you can use the RETAIN statement (before the SET statement) and thus shorten the RENAME= dataset option:

data GPS_Edit5;
retain Run_Number Corr_z;
set GPS_Edit4(rename=(speed_kmh=_Speed_km_h));
* ... more statements ...;
run;