Hi,   Here is my solution to your question   data have; length Observation 3 Code $20; input Observation Code; datalines; 1                  abccccccccccc 2                  aaaaaaaaaaaa 3                  bbbcccccccccc 4                  aaaaacccccccc 5                  bbbbbbbbbbbb 6                  bbbbbbbaaaaa 7                  cccccbbbbbbbc ; run;   data want;     set have(where=(prxmatch('/[^bc]/',strip(code))=0));  /* Only read observation with nothing but b/c/both */     if (findc(code,'c') AND findc(code,'b')); /* Only keep observations with both b&c */ run; ... View more

Hi,   The approach I have used to achieve something like that with Web Report Studio & OLAP Cubes was,  - Calculate/Create Country aggregate for all numeric measures/variables and them to Town level data table  - Calculate/Create City aggregate for all numeric measure/variables and join them with Town level data table Basically you end up with three sets of all the numeric measures/variables   Use differenet stats for these sets of variables - Country & City level aggregates ---> Use Mean/Average in order to show a "constant" Parent value - Town ---> Use Sum   In terms of selective display of variables based on Filter selection, you'll have to see if any of the techniques listed in this blog Use parameters to pick your metric in Visual Analytics Reports can help.   Good luck, Ahmed ... View more

Try this,   ODS ESCAPECHAR='^';   PROC FORMAT ; VALUE dy 1 = '1 ' 29='29' OTHER =' '; value trtn 1= "TRT A 5x10^{super 10}" 2="TRT B 1x10^{super 11}" 3='Placebo'; RUN;   Hope it helps, Ahmed ... View more

Hi,   In addition to the answers you have got so far, I would recommend checking out the following link Sample 24804: %SQUEEZE-ing Before Compressing Data, Redux. Running your data sets through the %squeeze macro, could be a general good practice to control the overall disk space consumed by your SAS data sets, regardless whether you are uploading them to LASR or not. This approach does not have the CPU overhead that occurs when reading compressed data set.   Hope it helps, Ahmed     ... View more

Hi Bob,   I don't have a direct answer to your question, but I wanted to highlight the issue from a different prospective, maybe that would give you an alternative way to achieve your goal.   Without knowing your SAP-Hana Libname settings, there are two Libname statement options I would recommend you revisit - schema= - Readbuff= - PRESERVE_TAB_NAMES=   If I'm not mistaken, typicall EG query builder, in fact, any GUI query builder, would try to extract list of Tables & Columns from the under lying dicitonary tables, and if you are not specifying a specific schema ownning the tables, you could be retrieving a list of all the tables within your SAP-Hana Instance!   Check out this SGF paper "EMPOWERING SAS® USERS ON THE SAP HANA PLATFORM", it may provide some hints and tips that can help with your issue.   Good luck, Ahmed ... View more

Ricky,   Have a look at this SGF paper http://support.sas.com/resources/papers/proceedings16/3660-2016.pdf   Hope it helps, Ahmed ... View more