Top down: data _null_; j=24; do i=0 to 24 by 2; b=repeat('*',i); c=floor((j-i)/2); put +c b; end; run; * *** ***** ******* ********* *********** ************* *************** ***************** ******************* ********************* *********************** ************************* Bottom up: data _null_; j=24; do i=24 to 0 by -2; b=repeat('*',i); c=floor((j-i)/2); put +c b; end; run; ************************* *********************** ********************* ******************* ***************** *************** ************* *********** ********* ******* ***** *** * ... View more

Enterprise Guide (EG) 8.3 is available and here is a list of my favorite news. This list is not a complete, but just my favorites; have a look at the additional news in the on-line documentation.   You can now work in EG without using a project As a programmer it has been quite annoying to work with projects, but now you can just open a program or a table without adding these to a project. You can still use projects if that’s what you prefer, but can in the same project open data or programs without “destroying” the process flow Options No more restrictions for windows layout Each window can now be layout exactly as you wish, floating or docked. Just drag an open window around to undock, or dock a window using this “helper”: Data unlocked automatic. This error: ERROR: You cannot open WORK.FEMALE.DATA for output access with member-level control because WORK.FEMALE.DATA is in use by you can now be avoided Options : Or you can easy close all open items, either from the window Open Items : or right-click on the tab you want to keep and close all other: Tree remains opened after refresh. If you have opened a tree in e.g. server window, you might have noticed that a refresh previous collapsed the tree. In EG8 the tree remains opened: Drag/Drop enabled You can now use drag and drop from e.g. server list,  either table or column names can be draged: You can even use drag and drop from output and log window ... View more

The REPEAT function can be really helpful, especially when you want to add 0’s before a character variable. The REPEAT function, REPEAT (argument, n), returns a character value consisting of the first argument REPEATed n times.   So first, a simple example on how to use the REPEAT function: data temp; x=REPEAT('Merry Christmas ', 1); put x=; run; The REPEAT function returns a character value that consist of the first argument repeated n+1 time, hence this output: The second demonstration show an example on how to add 0’s before a numeric variable ‘y’: data numeric; input x y; cards; 7896 44 5431 243 8934 1 6541 675 ; run; data temp; set numeric; yy = put(y, z5.); run; The z5. is adding ‘k’ zeros before the variable y, so the total length of the new variable ‘yy’ is 5. The ‘k’ is calculated as 5 - number of values in the variable ‘y’, and gives us the following results: The Zw.d format doesn’t work on a character variable, so the last example show how to add 0’s before a character variable ‘y’: data character; input x y$; cards; 7896 A44 5431 A243 8934 A1 6541 A675 ; run; data temp; set character; yy = cats(REPEAT('0',5-length(y)-1), y); run; The CATS function concatenates 0’s with the variable 'y', while the REPEAT function calculates the repeat of 0’s. The repeat of 0’s is calculated as: 5 - number of letters and values in the variable y - 1. This SAS code gives the following output:     ... View more

Everybody knows how big mistake it would be if you open a present which is aimed to somebody else. Even quite young children learn to read their own names and they can find the right presents in a glimpse of an eye. Likewise trying to include all autoexec_usermod.sas files may cause errors, if you have many properly secured application folders in one SAS Application Server instance. But how to effectively find out if any given user have access to a given application folder?   Think about a scenario where you will have hundreds of independent SAS applications used side by side. You really don't want to have hundreds of SAS Application Servers in your metadata. Therefore you define many applications into one SAS Application Server instance. You also want to give the possibility to make some application level usermods. You include application level usermods programs into SAS Application Server usermods program by using include command with a reference to a file in application folder.   Then you find out that some applications needs to be secured tightly because of GDPR or any other requirements. Securing is done at operational system level by restricting the read and write rights only to the given user groups. Then you think that all is done? Nope. When application level usermods programs are in application folders, those files can not be seen by anybody else but the members of the given user groups. All other users will get error message from the autoexec every time they start SAS session. It's just like they have tried to open a wrong present, oops!   But how to check the file availability as fast as possible? And without any errors? If you build any logic to check the user rights from metadata, it will take too long time. Think that you may have tens or hundreds of applications and these checks should be done for every application at every start of SAS session of any SAS user ...   Please find below a sample macro, which is one light and fast solution for this case. I have used it when including SAS autoexec_usermods.sas files from secured application folders, but the same idea can be used also in any other similar use cases.   %macro include_file(filename);     /*     Include command may give an error if access right is not given.     To avoid the error from include command, we first try to open the file for read     and check if it is possible.     If it is ok, then also include will succeed and it can be done.     If file open does not give valid file id (fid), then also include command would not work     and therefore it should not be done.     This functionality is normal also in that case when we do not get valid fid.     Therefore this macro does not print any system error message to the log.     */     %local fileref rc fid;     %let fileref=usermods;     %let rc=%sysfunc(filename(fileref,&filename));     %let fid=%sysfunc(fopen(&fileref));     %if &fid gt 0 %then %do;         %let rc =%sysfunc(fclose(&fid));         %include "&filename";     %end; %mend include_file;   Sample macro call: %include_file(/sasfolders/lev1/my_application/my_application_autoexec_usermods.sas)     ... View more

Photo by  David Beale  on  Unsplash     With the new Nordic sentiment models and the holiday season, we decided to run some text and sentiment analysis on  one of the oldest Christmas songs, Silent Night,  in Danish, Norwegian, and Swedis h .   The  song is originally written in German  and the three translated songs do therefore have the same  starting point . Differ e nces between the languages should occur as  the translators had to get creative to get the rhymes and rhythm of the songs to  work.   Let us investigate who will be the happiest when singing the song on Christmas.   In order to extract sentiment from the text and understand which language  had the most positive version of the classic Christmas song “Silent Night”, we started with a text analytics pipeline in SAS  Viya  Visual Text Analytics .        Since we were interested in making use of our new Nordic sentiment models in SAS  V i ya  3.5, w e selected the new appropriate language specific model.  To get  access   to the  models for  the Scandinavian languages, contact  Ina or Rune as described in the end of this post.    In  Viya  4   these models will be readily available without having to specify them in the node themselves .        Below are the results of our sentiment analysis of “Silent Night”.      Verse   Danish   Norwegian   Swedish   1   0.60   0.60   0.77   2   0.60   0.77   0.69   3   0.88   0.92   0.60   Average   0.69   0.76   0.69   The sentiment is  here reported as the likelihood that the verse is positive , defined as a number between 0 and 1  with  0,5 being neutral. The closer to  1  the  more likely  it is  that  the   document  has a  positive   sentiment .     All three verses of Silent Night were positive for Danish, Norwegian, and Swedish, however, there were differences in the degree of positive sentiment between the languages.  So ,  the creative process of the translators  does change the sentiment of  the verses .  And  the Norwegian  translator was in gen eral able to make the most positively phrased song , peaking in the  third verse.     We are excited  for the Nordic sentiment models and for the functionality there!  If you want to learn more about text analytics or the  Nordic sentiment models, feel free to reach out to Rune Hjorth Nielsen ( RuneHjorth.Nielsen@sas.com ) or Ina Conrado ( Ina.Conrado@sas.com ). We are happy to chat about it some more  😊   Happy holidays!    ... View more