Yes,  compress(x, , 'kn'); this should still work for stripping out the characters, but I don't think there is an easy way of translating the characters.  You could create a replace dataset, something like: TEXT_FROM          TEXT_TO Å                              A ø                              0 ... Then have a code generator: data _null_;      set replace end=last;      if _n_=1 then call execute('data want; set have; var1=tranwrd("'||strip(text_from)||'","'||strip(text_to)||'";');      else call execute('tranwrd("'||strip(text_from)||'","'||strip(text_to)||'");');      if last then call execute('run;'); run; You would only need to add any new records to that replace table (which could be in Excel...) rather than updating your code each time.  Note the code above will create a datastep with each tranwrd statement from the dataset replace. ... View more

Well, just run that step under SAS 9.3 Win, and it seemed to store it fine. Personally I would strip these out, something like:  compress(x, , 'kn'); However you say you want to replace them with other characters.  Not sure how viable that is, what will you change the ø to for instance, some of these special characters would have different resolutions so you may have to do a set of translations. ... View more

As a previous post had mentioned why not perform your tasks on the SQL server where your data resides, store the results into a temporary table and then extract the data into SAS?  SQL databases should be fully optimized and geared up to performing code on very large datasets, although once you reach a certain point then it becomes less a database and more of a data warehouse with different storage requirements and processes. ... View more

Yep, that's pretty much what my code is doing, however your writing the generated code out to a text file, then including that text file.  That just seems like extra work to write the fil, include it, and then of course you need to ensure it is removed.  Just change the: if _n_=1 then put "data a&&pays&nopays;set a&&pays&nopays (drop=";   put var /;    if lastobs then put ");run;"; To: if _n_=1 then call execute("data a&&pays&nopays;set a&&pays&nopays (drop='");   call execute(var); if lastobs then call execute(");run;"); ... View more

Sorry, where?  In my second example I create a dataset which has a list of the conditions  to be performed: condition intitule like 'RT1%' ... So an example of a code generator: data test;   attrib teststring format=$20.;   teststring="Hello"; output;   teststring="World"; output; run; data _null_;   set test end=last;   if _n_=1 then do; /* This is the first row, has the text Hello */     call execute('data newtest;                     attrib onestring format=$200.;                     onestring="'||strip(teststring)||'"');   end;   /* this code executes for all other rows */   else do;     call execute('||"'||strip(teststring)||'"');   end;   /* This code is executed at last observation */   if last then do;     call execute(';run;');   end; run; /* Generated code from the data _null_ is */ data newtest;   attrib onestring format=$200.;   onestring="Hello"||"World"; run; ... View more

To add, you could create a dataset with your conditions and then generate from that also: data conditions:      attrib condition format=$200.;      condition="intitule like 'RT1%'"; output;      condition="intitule like '%RT12%'"; output; ... run; data _null_;      set conditions end=last;      if _n_=1 then call execute(' proc sql; select catt(nomvar,"_t&&l") into : droplist_u separated by ' ' from dico&&pays&nopays where ('||strip(condition)||');      call execute(' or '||strip(condition));      if last then call execute(');quit;'); run; ... View more

Well, several ways really. Way 1 - generate the final step: data dropcolumns;      set sashelp.vcolumn (where=(libname="..." and memname="..."));      if column name in ("ISPP...","..."); run; data _null_;      set dropcolumns end=last;      if _n_=1 then call execute('data WANT; set HAVE; drop=(');      call execute(' '||strip(name));      if last then call execute(';run;'); run; Way two: Normalise the data, e.g. transpose the data so you have the variables going down the table rather than across, then drop the ones you don't want and transpose the table back up again. ... View more

CSV file format does not have the concept of "Sheets".  It is a flat text format file.  Excel has the concept of "Sheet".  If you want to view the data in separate sheets in Excel then you need to change your export target to be Excel not CSV.  However I recommend if you are exporting to Excel to use the newer ExceXP tagset to output your data as you have far more control and options: Base SAS: Demo: ExcelXP Tagset and Microsoft Excel ... View more