topic Re: Variable Name as value in SAS Data Management

Variable Name as value

SASPhile — Fri, 23 Jun 2017 12:53:52 GMT

Sample data is attached:

How to create a dataset that contains the variable as value ,its origibal value and truncated values as shown below:

In the first record, business_name is truncated for company name "Byrum". The lengths of business_name and new_business_names are different and their values are also truncated.
In the recond row, the address got truncated for "ICORE".

I would like to show them in a dataset as following:

company_name	variable	Original_value	truncated_value
Byrum	business_description	GH Utilization&IV Solutions	GH Utilization&IV
ICORE	business_Address	P.O.BOX 5523,Kryton,TN 38172-2468	P.O.BOX 5523,Kryton,TN 38172

The dataset can have 100's of variables with original and truncated values. I want to compare only those variables there is a truncation in their values.

Re: Variable Name as value

Tom — Fri, 23 Jun 2017 12:59:25 GMT

What is the question that you want answered?

Do you want to find the observations where the truncated_value is different?

data want ;
  set have;
  where truncated_value ne original_value;
run;

Re: Variable Name as value

RW9 — Fri, 23 Jun 2017 13:14:27 GMT

Its been mentioned before, please don't post Office files as they are security risk. To get a good answer, post test data in the form of a datastep, what the output should look like, and explain any logic between the two.

Re: Variable Name as value

SASPhile — Fri, 23 Jun 2017 13:39:39 GMT

It is ficticious data!

Re: Variable Name as value

RW9 — Fri, 23 Jun 2017 13:45:41 GMT

It does not matter, Office files are a security risk, some even block these files. Not to mention the extra work responders would need to do to import that file, guess what the structure is etc. Please follow the folowing post on how to provide test data as a datastep in the body of the post (using the {i} button):

https://communities.sas.com/t5/SAS-Communities-Library/How-to-create-a-data-step-version-of-your-data-AKA-generate/ta-p/258712

Re: Variable Name as value

SASPhile — Fri, 23 Jun 2017 14:14:15 GMT

Sure!

Re: Variable Name as value

Patrick — Sat, 24 Jun 2017 03:19:19 GMT

@SASPhile

A bit better prepared sample data and desired output would be appreciated - but here you go:

DATA WORK.have(drop=_:);
  INFILE DATALINES4
    DLM='7F'x
    MISSOVER
    DSD;
  INPUT
    company_name     : $10.
    business_description : $30.
    _business_description2 : $30.
    business_Address : $40.
    _business_Address2 : $40.;
  output;
  business_description=_business_description2;
  business_Address=_business_Address2;
  output;
  DATALINES4;
ByrumGH Utilization&IV SolutionsGH Utilization&IVP.O.BOX 5523,Kryton,TN 38172-2468P.O.BOX 5523,Kryton,TN 38172-2468
ICORENovo DistributionNovo DistributionP.O.BOX 5523,Kryton,TN 38172-2468P.O.BOX 5523,Kryton,TN 38172
;;;;
run;

proc sort data=have;
  by company_name;
run;

data want(keep=obs_num company_name variable_name max_string trunc_string);

  length obs_num 8;
  if 0 then set have;
  length Variable_Name $32 max_string trunc_string $100.;

  array test {2} business_description business_Address;
  array _max_len [2] 8 _temporary_;
  array _max_str [2] $100 _temporary_;

  /* determine longest string per company and variable */
  do while(1);
    set have;
    by company_name;
    do _i=1 to dim(test);
      if first.company_name or lengthn(test[_i])>_max_len[_i] then
        do;
          _max_len[_i]=lengthn(test[_i]);
          _max_str[_i]=test[_i];
        end;
    end;
    if last.company_name then leave;
  end;

  /* DQ: output cases with truncated strings */
  do while(1);
    set have end=last;
    by company_name;
    obs_num+1;
    do _i= 1 to dim(test);
      if lengthn(test[_i])<_max_len[_i] then
        do;
          variable_name=vname(test[_i]);
          max_string=_max_str[_i];
          trunc_string=test[_i];
        end;
    end;
    if last.company_name then leave;
  end;

run;

Re: Variable Name as value

SASPhile — Tue, 27 Jun 2017 14:15:39 GMT

I'm seeing empty columns for variable_name max_string and trunc_string

##- Please type your reply above this line. Simple formatting, no
attachments. -##