topic Re: Hash Logic Example in SAS Programming

Hash Logic Example

AshokD — Tue, 28 Aug 2018 14:23:53 GMT

Hi All,

I need some guidelines to proceed with my requirement. Here are the details:-

#1) I have a input txt file data as below:-

Row1 -->123456 Transact1 Transact2 Transact3

Row2--> 456789 Transac1 Transact2

#2) I have queried from database and created another dataset 'Details' with below values:-

TransactionType Service

Transact1 A1

Transact2 A2

Transact3 A1

I have to compare #1 and #2 datasets to determine the Service for each transactions for the employee

Once we determined the service ,I have to output the data to different fles:-

If service is A1 then output to File1

If service is A2 then output to File2

Could you please someone guide me on the SAS Hash Logic functionality for this requirement ?

Thanks in advance for your help .

Re: Hash Logic Example

Reeza — Tue, 28 Aug 2018 14:56:40 GMT

Can you show what you've tried so far?

And provide sample data?

There's a tutorial here on Hash and you can find many more on lexjansen.com

https://www.lexjansen.com/pharmasug/2011/TT/PharmaSUG-2011-TT15.pdf

Re: Hash Logic Example

PeterClemmensen — Tue, 28 Aug 2018 14:57:06 GMT

What defines an employee? Do you need help importing the .txt file into SAS?

Please be more specific about what you want your desired result to look like.

Re: Hash Logic Example

Kurt_Bremser — Tue, 28 Aug 2018 16:18:16 GMT

Use the format approach I gave you in https://communities.sas.com/t5/SAS-Programming/Hash-or-Array-solution/m-p/489950

It only needs one pass through the large dataset, so it's best performance- and resourcewise. And easier to understand and maintain than a hash.

Re: Hash Logic Example

AshokD — Tue, 28 Aug 2018 16:21:40 GMT

Hi @Reeza

Here are the details:-

Dataset A:-

EmployeeID Transactiontype

123456 Transact1

345678 Transact2

678903 Transact3

Dataset B:-

EmployeeID TransactionType Service

123456 Transact1 A1

345678 Transact2 A2

678903 Transact3 A1

I have to compare A and B datasets to determine the Service for each transactions for the employeeIDs.

Once we determined the service ,I have to output the data to different fles:-

If service is A1 then output to File1

If service is A2 then output to File2

The Logic I have tried it:-

data FILE1 FILE2;

if _N_ = 1 then do;

declare hash h(dataset: "B")

h.defineKey('EmployeeID');
h.defineData('TransactionType','Service ');
h.defineDone();

end;

set A;
rc = h.find();
if (rc = 0) then do;
if Service ='A1' then output FILE1;
else if Service ='A2' then output FILE2;
end;
run;

Please have look , correct me if I am wrong anywhere.

Re: Hash Logic Example

AshokD — Tue, 28 Aug 2018 16:22:45 GMT

Hi @PeterClemmensen,

Here are the details:-

Dataset A:-

EmployeeID Transactiontype

123456 Transact1

345678 Transact2

678903 Transact3

Dataset B:-

EmployeeID TransactionType Service

123456 Transact1 A1

345678 Transact2 A2

678903 Transact3 A1

I have to compare A and B datasets to determine the Service for each transactions for the employeeIDs.

Once we determined the service ,I have to output the data to different fles:-

If service is A1 then output to File1

If service is A2 then output to File2

The Logic I have tried it:-

data FILE1 FILE2;

if _N_ = 1 then do;

declare hash h(dataset: "B")

h.defineKey('EmployeeID');
h.defineData('TransactionType','Service ');
h.defineDone();

end;

set A;
rc = h.find();
if (rc = 0) then do;
if Service ='A1' then output FILE1;
else if Service ='A2' then output FILE2;
end;
run;

Please have a look , provide your suggestion