How to identified special characters variables in dataset

[BrahmanandaRao]

Hi Experts,

Good Morning 

How to identify special characters variables in a dataset and how to delete those special characters from those variables 

[PeterClemmensen]

Can you provide some sample data? Makes it much easier to provide a usable code answer. 

You mean special characters in variable values, correct?

And do you want to do this in all character values in the data set?

[BrahmanandaRao]

data special;
input var1$ 1-12 var2$ 14-22  var3$ 24-35    var4$ 37-52  var5$ 54-67 var6$ 69-76 var7$  78-84;
datalines;
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!54	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj78	yyaa778
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!55	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj79	yyaa779
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!56	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj80	yyaa780
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!57	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj81	yyaa781
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!58	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj82	yyaa782
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!59	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj83	yyaa783
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!60	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj84	yyaa784
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!61	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj85	yyaa785
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!62	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj86	yyaa786
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!63	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj87	yyaa787
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!64	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj88	yyaa788
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!65	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj89	yyaa789

;
run;

In the above dataset  seven variables are there in sample dataset var6 and var7 only does not have special characters   first i want to idetify which variables have special characters and then delete those special charaters  keep alpha and numeric in those variables

[Kurt_Bremser]

Use the COMPRESS function on all character variables (define an ARRAY with _CHARACTER_) with the "kad" modifier.

[BrahmanandaRao]

data special;
input var1$ 1-12 var2$ 14-22  var3$ 24-35    var4$ 37-52  var5$ 54-67 var6$ 69-76 var7$  78-84;
datalines;
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!54	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj78	yyaa778
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!55	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj79	yyaa779
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!56	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj80	yyaa780
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!57	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj81	yyaa781
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!58	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj82	yyaa782
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!59	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj83	yyaa783
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!60	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj84	yyaa784
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!61	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj85	yyaa785
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!62	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj86	yyaa786
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!63	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj87	yyaa787
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!64	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj88	yyaa788
*/=464aa_!@#	12AAP@!$%	RAL%&*()!!65	8711UURR##!!*())	ZKQL+_))&%#%##	lsjkfj89	yyaa789

;
run;


data rspl;
set special;
array v[*] _character_;
do i= 1 to dim(v);
v[i]=compress(v[i],,'kad');
end;
drop i;
proc print noobs;
run;

[Kurt_Bremser · Accepted solution]

And did it do what you expected?

[Kurt_Bremser]

And don't forget to define what constitutes a "special" character (ASCII 0-31 or above 127, UTF, letters like the German "umlauts", ...?).