Always keep in mind that in SAS, time and datetime values are counts of seconds. A fractional difference must therefore come from fractions of seconds, which are not displayed by the formats assigned to your variables. ... View more

"How can I know what is the libaname  code that connect between sas and tera here??? I dont need to run the libaname in order to run it but I think that in background it is running but I dont know what  libname is running ." Do you want to know specifically what Terdata instructions the libname statement triggers? Try this before your libname: options sastrace="d,,,ds" sastraceloc=saslog nostsuffix msglevel=i; If you are interested what SQL is sent down, you can settle for: options sastrace=",,,d" sastraceloc=saslog nostsuffix msglevel=i;   ... View more

Hi Tom, thanks a lot for reply! I think this is very brilliant, I hope I can write code like this after a lot of practice. In fact, the key word set, retain, and if.first/last (and some other techniques that do not produce a visible output or a product that directly show how they work) are the techniques that I use least because I have not fully understand the logic (how they work) of them. It is good learning experience for me which let me know how to improve my code and thinking. ... View more

Hi Tom, I just tried both of your suggestions in sas, and I got the correct answer . You solved my problem, thanks so very much! ... View more

Everyone has their preferences.  SQL syntax is very good for somethings.  But in general it is extremely verbose and much more cumbersome that DATA step code.   Take your example: create table temp_new as SELECT byvar, ind, x1, y1, MAX(x1) OVER(ORDER BY byvar ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) AS x1_running_max FROM temp group by byvar; In a data step that is just: data want2; set temp; by byvar; if first.byvar then x1_running_max=x1; else x1_running_max=max(x1_running_max,x1); retain x1_running_max; run; So instead of having the say "create table...as...from" you just need to say "data...set...".  Instead of "group by" you just need "by".  Plus you don't have to put COMMAS between the names in the list if more than one BY variable is need.  Instead you can just use spaces.   And SQL does not support variable lists.   ... View more

Since you are only pulling over some of the columns it should not be that hard to use elimination to determine which one SAS considers as using a data type it does not support.   But I wonder what you mean by creating time of a VIEW?  And why you want to know it? The last time a view is modified would be the last time the view's definition was changed.  Is that what you need to know?  If you want to know the last time the DATA that the view might surface was changed you probably will need to check the modification times of the TABLES that the view queries. ... View more

Thanks! ... View more

Nope it's a question from sas base practice exam ... View more

What kind of assets are trying to reach, and how do you intend to use them (from SAS)? ... View more

Hello @Mahis  We are also on SAS 9.4M6  and do use cleanwork utility to cleanup the SASTemp which has users work artifacts.(Work Library). In my experience, it has only deleted only those folders whose process does not exist. Never did I face a scenario where the artifacts of the running process have been deleted. When you say "The error appears only at the point where it needs to write an output dataset into WORK:" please examine if there are disk quota limits set at the OS level. May be the quota is being exceeded. (I understand that disk quotas may be set in terms of space or Inode numbers).  Also consider the suggestion by @MargaretC  ... View more

Thanks ... View more

You should only use the ancient CALL SYMPUT() routine when it is important that leading and/or trailing spaces be put into the macro variable.   Instead use its replacement the CALL SYMPUTX() routine.   If it worked for you it is only because the result of the SCAN() function call completely filled the variable STATION which you defined as length $6.  If STATION had been defined longer or the value pulled out of TEXT had been shorter you would have put one or more trailing spaces into the macro variable STATIONID.   In which case your data step would have made two output datasets since you would have told SAS to run something like: data U0000D _output ;     ... View more

One added information SAS Help Center: BULKLOAD= LIBNAME Statement Option Notes: Support for Microsoft SQL Server, Spark in HDFS, and Yellowbrick was added in SAS 9.4M7. Support for Informix and for Spark in Databricks was added in SAS 9.4M9. ... View more

Thank you for your interest in my post! In my data, "Leave" comes first (records are EMS transports from scene to a hospital). Choosing which variable to change is based on the DAY, MONTH, and TIME parts (the "Leave" field should be earlier then 'Arrived"). The logic if deciding a different year is a problem is that a transport from scene to hospital should not logically be that long. I have identified a specific number of records where the YEAR difference is a problem, none of those records had dates between Dec. and Jan. or contained Feb. 29. I wanted to use the code for changing the specific records identified and not the entire data set. Thanks again 🙂 ... View more

For further clarification: RESOLVE can only resolve macro elements, but it can not execute the resulting code. If macro elements within a macro are meant to be created/changed by non-macro code contained in the macro, this will not work as intended, as the non-macro code will never run. CALL EXECUTE, OTOH, feeds code into the execution queue. But you need to take care when a macro is involved, as macro triggers will be resolved immediately while all other code can only run after the current step finishes. Therefore, %NRSTR is needed to mask macro code during the insertion into the execution queue; it will then be resolved when code is fetched from the queue.  ... View more