Seems like you two main issues resolve around the VBscript part. So probably should ask on a VBscript forum, especially since you do not actually include any of the SAS portion.   You might describe your SAS environment. Depending on specifics of what the "automation script" is supposed to do there may be tools in SAS that does what is needed. ... View more

To schedule EG projects in SMC you first need to export your projects to SAS programs. These can then be scheduled in SMC via Schedule Manager. I suggest you reach out to your SAS administrator regarding the scheduling setup for your SAS installation.  ... View more

Just update your first step that is getting the filenames to also try to call the FINFO() function.   There is an example in the documentation.  *This example stores information items about an external file in a SAS data set; data info; length infoname infoval $60; drop rc fid infonum i close; rc=filename('abc', 'physical-filename'); fid=fopen('abc'); infonum=foptnum(fid); do i=1 to infonum; infoname=foptname(fid, i); infoval=finfo(fid, infoname); output; end; close=fclose(fid); run; ... View more

@Seb_A_Sanders wrote: I have a list of flags that I would like to define in a %LET statement   %LET vcodes = LCI LJL DIB MEH CRB LEB BAS AGE STC DMA; I then need to filter a couple of tables based on these, I've done this with a string of values, but it doesn't seem to work for characters:   PROC SQL; CREATE TABLE codes AS SELECT * FROM Source_table WHERE warning_code IN (&vcodes); RUN; I have also tried: PROC SQL; CREATE TABLE codes AS SELECT * FROM Source_table WHERE warning_code IN ("&vcodes"); RUN; With no luck, is there a trick to this? Why does it work for numeric values, but not character?   Thanks in advance! It does not work for character values because the result is   WHERE warning_code IN ("LCI LJL DIB MEH CRB LEB BAS AGE STC DMA"); when what you would need is WHERE warning_code IN ("LCI" "LJL" "DIB" "MEH" "CRB" "LEB" "BAS" "AGE" "STC" "DMA"); Each literal value needs its own set of quotes. Numeric values, not needing quotes don't have this issue with macro substitution.   Here is one of many ways to create a quoted list of space delimited items. %macro quotevals(vcodes=); %let temp=; %do i= 1 %to %sysfunc(countw(&vcodes.)); %let temp= &temp. %sysfunc(quote(%scan(&vcodes,&i))); %end; &temp %mend; %let qcodes = %quotevals(vcodes = LCI LJL DIB MEH CRB LEB BAS AGE STC DMA); %put quoted codes=: &qcodes; ... View more

Hi @Seb_A_Sanders    I think your only way out of the problem is to use SQL passthru. This way you can handle +32 long names of both tables and columns. Se the following code, which is a subset of working production code cut down to 3 variables:   options Validvarname = V7; proc sql; connect using xkmdne as sconn; create table &_OUTPUT (label="archive_NEXUS2_medicine_medication_additional_information") as select medication_id length=8 , administratively_deleted_datetim length=8 format=nldatm20. label="administratively_deleted_datetime" , last_past_fmk_dosage_period_end_ length=8 format=ddmmyyd10. label="last_past_fmk_dosage_period_end_date" from connection to sconn ( select medication_id, administratively_deleted_datetime as administratively_deleted_datetim, last_past_fmk_dosage_period_end_date as last_past_fmk_dosage_period_end_ from archive_NEXUS2_medicine_medication_additional_information ); disconnect from sconn; quit; Explanation:   Make a connection to the relevant SQL database using a previously assigned libname to the database.   Create a new SAS table with name abbreviated to 32 chars. Use original SQL table name as dataset label. select from connection to database, use original SQL column names as variable labels.    Connection to database gives the result of the second select, where long column names are abbreviated to sonething useful. This select is sent to execution in the database, so all long names are kept on the SQL Server side, and SAS sees only your abbreviations. ... View more

You need to do more planning, before any programming can take place.   Why would 6 accounts be able to evenly split 40p?   What would be a fair result?   Those aren't programming questions, they're planning questions. ... View more

I will leave the discussion on why to transpose to others. This is my suggestion assuming you made all the right choices   In EG you can create a query with a left outer (or in my case right) join between a table of all months and your data such that all missing monts are added to the data with all other columns missing. Then do the transpose. You should then get a var for every month. EG would create such a join as follows:   PROC SQL; CREATE TABLE WORK.QUERY_FOR_TEST AS SELECT t2.month, t1.dept, t1.value FROM WORK.TEST t1 RIGHT JOIN WORK.MONTHS t2 ON (t1.month = t2.month); QUIT; Hope this helps,   - Jan. ... View more