I tried the UNIVARIATE approach on a wide table of 7500 columns (35 rows), and UNIVARIATE ran in 300s.  All other techniques I tried on this table ran much much slower.  I think the Univariate approach is very powerful.  Here is my code just as an example.     ods listing close; ods output frequencies=_afreq; proc univariate data=sashelp.cars NEXTRVAL=20 freq; var _numeric_; run; ods listing; ... View more

Hello, My 2 proc report for 9 columns works well, but sometimes, the columns are truncated (I have 6 columns, and another 3 columns in the end). When I re-run the programme it is not truncated. Maybe you know some option to fix it? Thank you! ... View more

hi, The reason this is happening is because when you select *, you are telling sas you want to include everything from both the tables, and as year and season exists on both tables you can't have two variables with the same name. In order to get rid of it, you would need to tell SAS which of the two tables you want to pull year and season from. ... View more

Thank you so much! I also just figured that I can merge all the datasets and it will give me my desired dataset. ... View more

@Kastchei Thanks for all the testing and debugging. As you already figured out: o  compile once I   case insensitive   As I understand it the o wouldn't be necessary when using a constant string. It's only of importance if you pass in the RegEx via a variable.   RegEx are resource intensive and though "slow". But they are really powerful when it comes to pattern matching.   As for the \w: Good catch!    Here some docu links. Especially the 2nd one (metacharacters) is sometimes a bit hard to find. http://documentation.sas.com/?docsetId=lefunctionsref&docsetTarget=n0r8h2fa8djqf1n1cnenrvm573br.htm&docsetVersion=9.4&locale=en  http://support.sas.com/documentation/cdl/en/lefunctionsref/63354/HTML/default/viewer.htm#p0s9ilagexmjl8n1u7e1t1jfnzlk.htm    Cheers, Patrick ... View more

Thank you very much sir,    Is there possible to add a logic check only sent email if the sql return is not null?  ... View more

Piggybacking on @Ksharp's code here an option which modifies the master data set in place. data master; set sashelp.class; run; data trans; length name $ 8 sex $ 1; /* updates */ name='Robert'; sex='N'; output; name='Thomas'; sex='A'; output; /* inserts */ name='Ksharp'; sex='M'; output; run; data master(drop=_:); modify master end=last; if _n_=1 then do; dcl hash h1(dataset:'trans'); dcl hiter hh1('h1'); _rc=h1.defineKey('Name'); _rc=h1.defineData(all:'y'); _rc=h1.defineDone(); end; /* updates */ if h1.find()=0 then do; replace; _rc=h1.remove(); end; /* inserts */ if last then do; _rc = hh1.first(); do while (_rc = 0); output; _rc = hh1.next(); end; end; run; Please Note that for the Insert case all values for variables which only exist in the Master dataset but not in the Transaction dataset get retained (which is of course wrong). If you don't have all variables in your transaction dataset then you need to use a "call missing()" for the Insert case for all variables which only exist in the master dataset. ... View more

LIBNAME mylib CLEAR;  This statement only deassign  In Log you can see NOTE: Libref mylib has been deassigned.  ... View more

It is possible to use PROC SQL within a DATA step.  Take a look at my South Central SAS User's Group paper from last year:  http://www.scsug.org/SCSUGProceedings/2011/poling1/SQL%20Function.pdf.  While the code presented in the paper only executes a SELECT statement, I think that you could modify the code to execute a CREATE TABLE statement. ... View more

Is the server you are connecting to through EG also Windows? ... View more