SAS Studio currently (Viya 2024.08) uses the same icon for SAS Program nodes in flows regardless whether it is embedded code or referenced external file. Looking at a flow, there is no way to distinguish between those two types. If you want to know, you need to click on every single node and look to Code or Node tabs. It would be very beneficial SAS Studio can use different icons for SAS Program nodes based on whether it is embedded code or referenced external file. The same holds for Python Program node. I believe it takes just a little effort with the great impact on usability.    Here are just two examples: 1. Embedded code (as there is no path shown at the top of Code tab) Embedded code 2. Code referenced externally (the path to the external file is shown at the top of Code tab) Code referenced externally   ... View more

According to this usage note, there's an old defect for CONNECT USING together with OLEDB access: 48877 - A pass-through connection invokes the Data Link Properties dialog box even though an assigned OLEDB libref is also present (sas.com) CONNECT USING was imoplemnted because it's conventient to avoid keeping connection information in many places. Also, this especially nice when you want to generete the connections dynamically in PROC SQL. ... View more

SAS Mirror Manager can upload/push SAS images to some container registries like AWS ECR. I suggest to add an option to allow to delete/remove SAS images in order to be able to cleanup the container registry. For example, if I use SAS Mirror Manager to upload three SAS releases to AWS ECR: Stable 2024.04, 2024.05 and 2024.06. Now I want to delete all SAS images that belong to release Stable 2024.04 to cleanup my AWS ECR but there is no option in SAS Mirror Manager to do it. ... View more

Prometheus currently only tracks active jobs, with no visibility into completed ones. Once jobs finish, their status information from the Workload Orchestrator is not available to prometheus, limiting post-completion monitoring and analysis. It would be beneficial to externalize this data to Prometheus for better job status tracking.  ... View more

I discovered in my latest tech support track: CS0136992 that client IP Address is only passed to the metadata logs for calls related to: New Client Connections  Redirects for the client in the cluster  AND - it cannot be configured to be added for other logging events! SAS Help Center: X Conversion Character I would like to request the ability to add Client IP Address as a descriptive characteristic to various Metadata Audit logging events such as:  Group/Role Changes Logon/Logoff Role Permissions Changes  Client IT teams often work with Splunk & other tools that query logs using a NoSQL query language. As duplicative as it may be - this allows IT teams to easily comply with security and tracking requirements to be able to easily track and identify cyber security events and get all information in a single query.    Without this - what I'm being told, is security teams would first need to identify the event in question, then using the request ID - track that all throughout the logs until they found the initial new client connection in order to find the source IP. This kind of sub-query is incredibly tedious behavior and creates a significant disconnect that must be manually traversed in order to create the wholistic picture cyber security teams are looking for.    Please include this as a feature in the next release!  ... View more

Expand the DNUM and DREAD functions with an additional "modifier" parameter, so hidden files (where the name starts with a dot) are included in the count and can be found.   Such files can be listed by using the -a parameter of the ls command, so their "hidden" status is relative, and the additional parameter of the functions would avoid the need to use external commands (which, in turn, forces SAS admins to enable XCMD). ... View more

PRX-functions (prxmatch(), prxparse(), etc) that deals with regular expressions still cannot process multi-byte character sets like UTF-8 in the SAS 9.4 releases. It is an essential feature that would facilitate a lot a migration to a MBCS context. ... View more

Starting with SAS Viya version Stable 2024.05, PROC PRINT will now display an empty table in the Results window if the source data set has 0 rows. This is a change from previous versions going back to SAS 9. Prior to this change, running PROC PRINT from a data set with 0 rows sent no output to the Results window.   Some users generate "check" data sets in their code that create rows of data and display results only if something unexpected has happened. This recent change in SAS Viya 2024.05 now causes "false alarms" to go off in previously stable code, as results are now generated from empty data sets.   Although this new behavior can be avoided by adding extra code to check the number of rows in the data set prior to running PROC PRINT, end-users' code would be less impacted if an option was added to allow the user to decide if results should or should not be generated when the source data set has 0 rows. Please consider adding such an option. Thanks! ... View more

In large enterprise organizations, it would be beneficial to be able to hide, or make visible, folders from the SAS server files that are mounted from the Studio Explorer pane based on the groups the users are in.   See Case: CS0115567 for the original support question and example. ... View more

There have been quite a few spam questions over the past few months where it took sometimes days until they got removed.   Proposal Implement an automated process that moves questions from the visible list to some hidden queue for investigation if marked by multiple users an inappropriate content.  To avoid that this process can get misused: Only include users that marked the question if they are not New Users but already on a certain level (to exclude fake users).   And just thinking: Eventually request from New Users for their first few post some additional identification step when posting a question to make it a bit harder to automate posting such questions. ... View more

To be consistent with those longtime users coming from the SAS 9.4 world, it would be very useful to have an option to append to the results window rather than replace its contents when resubmitting a program or a portion of a program.  I see that SAS Studio 3.8 added an option to append to the log window rather than replace and it would be great to have a similar option for the results window.  We will soon be moving to SAS Viya so I'm of course most interested in this being implemented in the latest release of SAS Studio running under Viya. ... View more

  The new (and former) site SAS for Developers does not contain proc http examples when using the rest api's I think this is a great omission sincs many customers uses this api's from whitin.     ... View more

Sometimes when a SAS program fails you only want to run from the failing point, it would be nice if there was the functionality in SAS EG to run from where the cursor is to the end of the program. At the moment you have to select the bottom half of the code which is a bit of a faff in a big program. ... View more

Hi Team We are big SAS customer running our SAS on Solaris OS. Currently GIT functionality with in SAS Studio is not supported for Solaris OS. We need this feature.     Thank you ... View more

Currently in the SAS-Viya/SAS Studio (Version Stable 2024.02) QUERY interface, it is difficult to control where new columns are inserted in the Query column listing.   The interface is set up for a user to drag a new column from the tables/Columns window on the left into the QUERY>SELECT>Columns Tab on the right for insertion.   It highlights the row the user hovers over, but upon insertion, the new column is not inserted there.  Instead it shows up at the bottom of the Query column listing.  SAS Technical Support informs me this is the "expected" behavior in the current version.   Request:  Please improve this interface to allow users to insert new columns where they are desired in the overall Query column listing.   i.e. provide ability to: - Select where in the column listing to insert new column(s) and put them there when executed. - Allow selection of multiple columns by holding SHIFT or CNTRL keys and selecting them.  Then allow user to drag them to the desired position in the listing.    - When insertion/move occurs, leave the display showing the same section of the column listing so the user can confirm where they went.  Currently, when a move is done, the entire listing changes to a different view so you have to go looking for what was done.  ... View more

Hello.  When working with PROC OPTMODEL and PROC IML, we're actually working with a solver.  Between versions of SAS, this solver is constantly changed for various reasons.  When the solver is updated, a problem that works fine with the current version of SAS then no longer works with the new version.  Since this is for multiple clients, it is not possible to optimize the solver and please everyone.  Would it be possible to add a new option to the PROCS so that when we invoke them, we pick which solver we want to be using?  This way, if someone wants a solver optimized, then a new solver would be shipped and they could use it while it would leave all other users stable as they could keep using the solver they were using before.  This of course means that SAS needs to ship several solvers but I'm sure it can be done. ... View more

If an external object (f.e. Custom Step) is added to the Flow then it is referenced by the full path. We would like to reference it by a variable or relative path like "../../Prompts.step" instead. This functionality constitutes the Git integrability and allows for a simultaneous work of more than 1 person while keeping the project in the Git.   We would like to see a following set up – each developer works on own local repository. Inside a repository we have number of Flows. Some Flows in a folder contain Custom Steps.        One developer changes something in such Custom Step and commits the changes. Other developer clones the repository and tries to change the same Custom Step. Dev2 is now unable to save it, because the path of the Custom Step points to dev1’s local repository. It means that even though the Flow is in repo2, the object inside (Custom Step) is from repo1. The path to DEV1's repo is visible if we hover over opened inside object from DEV2's Flow.   It is like this:   We want it to be like this:   Now we have to remove the Custom Step from each Flow and attach the new version, then select every custom option inside. It takes a lot of time, since one Custom Step can be in more than one flow. To solve this we could include objects with relative path (f.e. ../../Prompts/1_Prompts.step) or variable (f.e. ${path}/Prompts/1_Prompts.step) so it works across repositories. We don’t want to keep those Custom Steps outside repository as there is no DI’s check-out option and anyone could change the code of the Custom Step anytime. recording - ref: CS0060147 ... View more

We're using SAS 3.5 VIYA; user's account account are from our company LDAP. When a user connect to our SAS, due to LDAP specificity, the account domain may be lower case our upper case (depending from connection method) This is currently not handled by SAS: some functions are case sensitive, some other case insensitive. We suggest than adding a configuration to manage this.   More details : case CS0064251 ... View more

SAS is responsible for a huge number of users in our organization, and there is not an easy way for administrators to post messages to users upon connection to our back end services across all the various UI's.  It would also be nice to have a "stub" service that could be left running for maintenance notification.   It would be good upon connection to a Meta server to have a way to post a message before, and after authentication in a standard way. It would also be good to provide real-time announcements.   Message pre-authentication: "There will be maintenance on Saturday from 4:00pm to 7:00am."  Message post-authentication: "Due to issues with the NAS interface, there may be slow connections when browsing the file system." Real time/Flash messages: "The system will go down in 45 minutes for patching."   (in this scenario it would be nice to stop allowing new connections, and then gracefully quit running sessions to acquiesce)   It would also be nice to provide a system is down for maintenance message.   Perhaps the client could default to checking a simple web service on the meta server for maintenance notification.   HTTP/HTTPS providing a response of: Maintenance: "The SAS environment is currently down for maintenance; SAS will be available by 7am ET."   Client could show "The SAS environment is currently down for maintenance; SAS will be available by 7am ET."   If you use a simple HTTP/HTTPS interface, then you can use a DNS failover or a simple web service on the host to provide information when the service is down.  (Even a simple Python http service 'cd /sas/maintenance; python -m SimpleHTTPServer 80') ... View more