About ChrisBrooks

ChrisBrooks · ‎08-03-2017

This is a good example - except sign on as different users https://support.sas.com/rnd/scalability/tricks/connect.html

ChrisBrooks · ‎08-03-2017

Your PC session can be connected to more than one remote session at the same time although when issuing your signon, signoff and rsubmit you have to name the sessions. You also have to coordinate their timings with the waitfor statement. Example 4 at this link shows you how it's done. So at the start of your run you log on twice - once as your normal user and once as your super user. Your program should rsubmit to your normal user session until you hit the part where you want to run as the super user. You then rsubmit that to your super user session. I've run multiple sessions at the same time using this technique a lot - it's easy once you get the hang of it.

ChrisBrooks · ‎08-02-2017

This is the simplest way data a; input staff house car; cards; 1234 1 0 2234 5 1 3234 2 2 ; run; data b(drop=house car); length staff 8 remarks $17; set a; remarks=cat(house,' house and ',car,' car'); run;

ChrisBrooks · ‎08-01-2017

If you can isolate the issue to Firefox it might be worth raising it with SAS Support - I've always found them very responsive when I've raised issues with them.

ChrisBrooks · ‎08-01-2017

What browser are you using? I say this because there have been problems reported in the past with strange behaviour when using Chrome (I've experienced them myself although the latest version of UE seems fine). Try using a different one and see if you get the same issues.

ChrisBrooks · ‎07-31-2017

Can you post the rest of the SQL statement as it'll be much easier for us to give an informed response to your issue?

ChrisBrooks · ‎07-31-2017

I've used Ultraedit to write my SAS code for a number of years - that has the concept of Projects you're looking for plus a lot of other reallly nice features (bookmarking lines of code, fast searching in code files etc). Of course you have to drop back into display manager or EG to run it for testing but you quickly get into the habit of that.

ChrisBrooks · ‎07-31-2017

Another problem is you're storing a date as a text value which means when you sort it 20Dec2016 is coming before 30Sep2016. Also as mentioned holding duplicate key values is a very bad idea which will cause no end of problems further down the processing track....

ChrisBrooks · ‎07-30-2017

Quite simply a and b are the same table which you are joining with itself - b is just an alias for a which is set up by the "from a b" portion of the query.

ChrisBrooks · ‎07-28-2017

Interesting - I must try these libnames out!

ChrisBrooks · ‎07-28-2017

There's a FedSQL function called radians - the following creates a table called rads in work which holds the result of the calculation proc fedsql; select radians(45) into rads; quit;

ChrisBrooks · ‎07-28-2017

Generally speaking it's much easier to program with a long data set than a wide one but if you must do it then Proc Transpose is the best way. Can you post a data step with some sample data (not an excel file) and let us know how you want it transposed.

ChrisBrooks · ‎07-28-2017

Google is your friend..... http://support.sas.com/techsup/notes/v8/19/411.html

ChrisBrooks · ‎07-27-2017

Is the variable character or numeric? If it's numeric I would expect a missing value to automatically be inserted during the importation; if it's character I would expect it to automatically be an empty string.

ChrisBrooks · ‎07-27-2017

There's a SAS Interface to Metadata that you might be able to use - the metadata_newobj function can create metadata objects. I'd advise extreme caution in developing and testing it though as programatically altering metadata could turn into a nightmare. The link to the documentation is here

Online Status	Offline
Date Last Visited	‎10-08-2024 11:59 PM