Today is a really giving day, as you today get 2 tips: EG tasks in SAS Studio Knowledge sharing links & LLM’s   EG tasks in SAS Studio Now the 30 most popular EG tasks are available in SAS Studio – release 2025.11   Knowledge sharing links & LLM’s I often Google to find SAS information, code examples and inspiration. But it is not always I find the right place. So here are my most used sites for SAS knowledge: FANS Previous Events SAS Communities Past Conference Proceedings: 1976-onward Support/Documentation by Product Webinars SAS Video Portal Youtube @SASUsers Github – SAS Software Lex Jansen   And LLM’s are of course also getting more and more useful. So here is a list of possible LLM’s: Gemini Gemini 3 Pro, should be good at proc or even light macro generation ChatGPT GPT 5.1, is close in performance to Gemini 3 Pro Qwen 2.5 coder Microsoft Copilot Claude Llama Cohere Read more about LLM’s here: 27 of the best large language models in 2025   And last but not least: Our own SAS Viya Copilot will be available with the SAS Viya 2025.12 release 😊   Merry Christmas Frans Holm Knowledge sharing with FANS ... View more

Hi and happy Juletip!   Introduction In Viya4, SAS Studio can interact with GitHub using authentication over SSH or HTTPS. From cadence 2025.11, SAS Studio require a file with “known hosts” in the users HOME folder, in addition to the private and public key. This is a bit cumbersome to create when users cannot SSH/login to the compute host in Viya4. An alternative is to use HTTPS and GitHub PAT (Personal Authentication Token) as logon method for integrating GitHub and SAS Studio, see links:   GitHub setup of PAT etc.:   https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/managing-your-personal-access-tokens   https://stackoverflow.com/questions/18935539/authenticate-with-github-using-a-token   SAS Git: https://go.documentation.sas.com/doc/en/sasstudiocdc/v_068/webeditorcdc/webeditorug/p0urbfmbb9lkpdn15yzavxdk1lgk.htm   Steps to setup at GitHub for PAT: Go to github.com and log in using your GitHub ID. Then: Settings -> Developer Settings  (in top right corner and at the bottom of the new menu list) Select "Person Access Token" -> Fine-grained tokens and Generate Key Give the Token a name, description and an expiration date. Set Repository Access to all of your owned repositories (or select one or more specific repos). Permissions. This is the uncertain area. At first, with default settings, it seems on READ for Metadata. And that may not be enough. When it worked, I used:                - READ to everything                - READ and WRITE to: Access to code, codespaces secrets, merge queues, pull requests, and workflows Unclear what the minimum permissions need to be.   Steps to setup in SAS Studio: Create a folder in the "SAS Server" for holding the files that will be under Git control. You can choose any folder you like.      Here the example path is: /srv/nortok/Git_Repos Right mouse-button on the folder and "initialize" it for Git usage. Then the icon for the folder changes to the Git icon. Go to Options -> Manage Git Connections and make a profile. a) Create a HTTPS profile with values adapted to your name:                               Profile name: TorOve_gitHub_HTTPS                               Commit author Name: <Your Name>                               Comit author email: <Your Email>                               Domain ID: GitAuth   The domain ID can be anything, but it is good to have a domain name representing the content in the domain. This domain will hold you GitHub logon ID and the PAT token.                 b) Create a repository with the following settings:                             Remote Repository: https://github.com/<your-selected-repository>.git                            (You can find this link in GitHub: go to the repository → Code → Clone → HTTPS)                            SAS Server Location: /srv/nortok/Git_Repos                            (Note: This path cannot be changed later and must remain as specified above)                            Profile: TorOve_gitHub_HTTPS                           (This is the HTTPS access profile created in the previous step)         4. c) In Manage Environment -> My Credentials set up:                             Domain: GitAuth                             User ID: <Your GitHub user ID>   (The userid used to login to GitHub)                             Password: <PAT>  The generated GitHub Personal Access Token.                                 Testing: Test this using the Git GUI in SAS Viya from the left handside menu (Git Repositories). Try to Pull/Push and add a file (new file stored in the Git folder) and the stage and comit it. The files should then be visible in SAS Viya and in the GitHub repository.   NOTES: SAS Git seems to only accept the branch names "main". So, a branch named "master" may create problems. Set the "main" branch as Default branch in Settings for the actual Repository. Files can be downloaded/Uploaded via GitHub upload/download to move/copy files between branches, like Master and Main. ... View more

REGION isn’t actually a statement but a comment. When you start a comment with region, SAS Studio recognizes it and treats the following lines – up to a comment marked /* endregion */ - as a collapsible block of code. This feature works in VS Code as well. It’s a practical way to group together data steps and procedures, allowing you to submit all the code in one go (Run Region) without having to manually select large sections. Regions can also be nested, as shown in the demo and example code. In SAS Studio, these regions are collapsible also in the SAS Log, making code management even easier. For a demonstration, see the video. You can find the code further down.      /* region main */ /* region connect to data */ libname xl XLSX "%sysget(HOME)/data/CARS.xlsx"; /* endregion */ /* region Fetch data */ proc sql noprint; create table WORK.CARS_Europe_raw as select * from XL.cars where Origin='Europe'; quit; proc sql noprint; create table WORK.CARS_Asia_raw as select * from XL.cars where Origin='Asia'; quit; /* endregion */ /* region Cleanse and enrich */ data WORK.CARS_Europe; set WORK.CARS_Europe_raw; L10kmCity = 23.521 / 'MPG (City)'n; L10kmHighway = 23.521 / 'MPG (Highway)'n; run; data WORK.CARS_Asia; set WORK.CARS_Asia_raw; L10kmCity = 23.521 / 'MPG (City)'n; L10kmHighway = 23.521 / 'MPG (Highway)'n; run; /* endregion */ *Clean up...; proc datasets lib=WORK nodetails nolist ; delete CARS_Asia_raw CARS_Europe_raw; run; /* endregion all steps */ ... View more

When working with SAS Viya you will eventually run into situations where the dataset you've worked with has dropped from CAS memory to a file in the CASLIB- library. In order to load the file or files back to memory, you typically need to list the names of the files with CASUTIL- procedure to find out the actual names of these files. And then you need ro re-run the same CASUTIL- procedure in order to load the files to memory using their full names including the correct suffixes. If you are like me, you end up copy-pasting all the names needed from Log- window to program editor one at the time. And, if you are like me, you never remember to promote the tables and you do this yet one more time. One feature I've been missing in CASUTIL- procedure is that it would be really nice if I could load all files with some specific text in the file name with one go. For example, I would like to load all my ABT- tables from files to memory.  I've now done a SAS Macro called LoadToCAS which loads all files contaiting a specific text from files to CAS memory.  The default CASlib is PUBLIC, but it can be over-ridden. Examples: %LoadToCAS(TableNamesWith="HMEQ_"); - macro call would load all files in PUBLIC -caslib containing text "HMEQ_" to memory. %LoadToCAS(CASLIB=CASUSER, TableNamesWith="ABT_"); - macro call would load all files in CASUSER -caslib containing text "ABT_" to memory.   %macro LoadToCAS(CASLIB=PUBLIC, TableNamesWith=""); /* This SAS Macro loads the wanted files from files to CAS. The default CASLIB- is PUBLIC, but it can be overridde by the macro's CASLIB- parameter. This Macro loads all files containing the example text in TableNamesWith- parameter. Note! TableNamesWith- parameter should be provided within the douple quotes. Ex1. %LoadToCAS(TableNamesWith="HMEQ_"); - macro call would load all files in PUBLIC -caslib containing text "HMEQ_" to memory. Ex2. %LoadToCAS(CASLIB=CASUSER, TableNamesWith="ABT_"); - macro call would load all files in CASUSER -caslib containing text "ABT_" to memory. */ /* Let's capture the file listing of CASUTIL- procedure. */ ods output FileInfo=FileInfo; proc casutil incaslib= &CASLIB.; list files; run; /* Create a dataset containing all rows with the table names we want. */ /* Also, count the number of rows. */ data WORK._FILES; set WORK.FILEINFO; where Name contains &TableNamesWith.; call symput("_CASDATA_" || left(put(_N_, 4.)), compress(Name)); call symput("_CASOUT_" || left(put(_N_, 4.)), compress(scan(Name, 1, "."))); call symput("_N_", _N_); run; /* Loop the rows and load the wanted files to CAS. */ proc casutil incaslib= &CASLIB. outcaslib= &CASLIB.; %do i = 1 %to &_N_.; load casdata="&&_CASDATA_&i.." CASOUT="&&_CASOUT_&i.." promote; %end; run; %mend;     ... View more

It’s that magical time of the year again! Our beloved tradition is back, and we’re excited to launch this year’s #SASJuletip Advent Calendar on Monday, December 1st.   This year, we’re sharing 14 unique and inspiring Juletips - one every workday - each packed with ideas to spark curiosity and help you discover fresh ways to explore SAS products and features. ... View more