BrunoMueller Tracker

Get connection information for a SAS9 server in Enterprise Guide and use it in VS Code

BrunoMueller — Wed, 13 May 2026 05:40:36 GMT

The other day someone asked me how can I connect from the SAS extension in VS Code to the same workspace servers I can see in Enterprise Guide. Using the information from the profile used in Enterprise Guide did not work. So it seems those two applications work slightly different.

This post will provide a step by step guide how to lookup the information in Enterprise Guide and use it when defining a new connection in the SAS extension for VS Code.

Re: Programmatically finding a user's own repository in the SAS metadata repository

BrunoMueller — Tue, 05 May 2026 17:36:38 GMT

Find below some general code to list all project repositories and their owner.

data md_project_repo_owner;
  length
    nRep 8
    type etype $ 32
    id eId $ 32
    repSearch $ 128
    repURI $ 128
    repName $ 128
    repPath $ 1024
    extSearch $ 1024
    ownerURI $ 128
    ownerId $ 64
  ;
  call missing(type, etype, id, eid, repSearch, repURI, repName, repPath, extSearch, ownerURI);


  * 
  * search for Project repositories
  *;
  repSearch = "OMSOBJ:Repository?@RepositoryType = 'Project'";
  nRep = metadata_resolve(repSearch, type, id);
  putlog "NOTE: Number of Objects found for" +1 repSearch= nRep=;

  *
  * get the individual repURI for each entry
  *;
  do objN = 1 to nRep;
    rc = metadata_getnobj(repSearch, objN, repURI);
    *
    * get Object type and id from objURI
    *;
    eObjType = scan(repURI, 2, ":\");
    eObjId = scan(repURI, 2, "\");

    *
    * get repository details
    * more attributes are available
    *;
    rc = metadata_getattr(repURI, "Name", repName);
    rc = metadata_getattr(repURI, "Path", repPath);

    *
    * search Extension objects
    *;
    extSearch = cats("OMSOBJ:Extension?@Name='ProjectRepository' and @Value=", quote(strip(eObjId), "'"));
    nExt = metadata_resolve(extSearch, etype, eid);

    *
    * get the owner object
    *;
    rc = metadata_getnasn(eId, "OwningObject", 1, ownerURI);

    *
    * get owner id
    *;
    rc = metadata_getattr(ownerURI, "Name", ownerId); 
    output;
  end;
  keep repURI repName repPath ownerURI ownerId;
run;

Who’s been using my CAS tables? Now with activity records

BrunoMueller — Mon, 04 May 2026 09:19:07 GMT

You might remember my post on Who's been using my CAS tables? where I wrote about the audit entries being created for access to CAS tables and how to read them. Recently I read in the documentation for Upcoming Critical Changes for SAS® Viya® Platform the section on Recording Audit Entries Disabled at Stable 2026.05. Does this mean, this information is no longer available?

Do not worry, we still have the same CAS access information available. We just need to read them using a different command. This article will provide the details needed to continue using these audit information.

Get the Code From a SAS Studio Flow using pure SAS program code

BrunoMueller — Mon, 04 May 2026 09:06:59 GMT

Sometimes we want to get the code that is behind a SAS Studio flow. See this post How To Get the Code From a SAS Studio Flow and Why It Matters for more details. The post provides a SAS program to read the generated code. Now often not all the code will fit into the max length of character variable in a SAS data set.

Help me type, the sas-viya CLI supports auto-complete

BrunoMueller — Tue, 21 Apr 2026 11:23:32 GMT

In a Linux terminal we are used to auto-complete for commands, filenames and so on. However when using the sas-viya command-line interface (CLI), pressing the tab key would do nothing. Of course you do have the built-in help, but it is not the same experience.

Beginning with Stable 2026.02 the sas-viya CLI now supports auto-completion for plugins and commands. This article will introduce you how to enable this feature and how to use it.

Re: How to change the format in VA in a 100% stacked bar chart

BrunoMueller — Tue, 21 Apr 2026 06:52:07 GMT

You are welcome and thanks for the feedback.

Re: Multiples box plots

BrunoMueller — Mon, 02 Mar 2026 07:42:01 GMT

About title and footnote. See sample program below. Depending on where the title and footnote statements are used, the apply to the whole page, or just within the region.

ods _all_ close;

ods graphics / reset=all height=110mm width=160mm outputfmt=png;

ods pdf file="%sysfunc(pathname(work))/layout.pdf" startpage=never;

title "Title outside of layout";
footnote "Footnote outside of layout";

ods layout gridded columns=1;
ods region;
title "Title inside region 1";
footnote "Footnote inside region 1";
proc sgplot data=sashelp.cars;
vbar type / group=origin response=invoice stat=mean;
run;
ods region;
title;
footnote;
proc sgplot data=sashelp.cars;
vbar type / group=origin response=invoice stat=mean;
run;
ods layout end;

ods pdf startpage=now;

title "Title outside of layout";
footnote "Footnote outside of layout";

ods layout gridded columns=1;
ods region;
title "Title for table";
footnote "Footnote for table";
proc print data=sashelp.class(obs=10);
run;

ods layout end;
ods pdf close;

Re: Multiples box plots

BrunoMueller — Thu, 26 Feb 2026 15:02:11 GMT

As long as two rows of images will fit on the page you just continue with

ods region;
proc sgplot ...;

for the next 4 plots.

Re: Multiples box plots

BrunoMueller — Thu, 26 Feb 2026 13:26:59 GMT

Since you output to a PDF, you can set some options like papersize and margins. You could use a A3 papersize, so there is more space for the graph, but still print it as A4. I would set the height and width using mm, as you can measure on the paper with a ruler. ODS LAYOUT has support for COLUMN_GUTTER=dimension and ROW_GUTTER=dimension

Here is a sample to test things out. I used macros for some of the options to make it easier to change

ods _all_ close;
options
  papersize="A3"
  topmargin=1cm
  bottommargin=1cm
  leftmargin=1cm
  rightmargin=1cm
  orientation=landscape
  nodate nonumber;
;
ods graphics / reset=all width=90mm height=135mm outputfmt=png;
ods pdf file="%sysfunc(pathname(work))/layout_gridded.pdf";
ods layout gridded columns=4 column_gutter=0;

%macro dataLabel;
  datalabel=model 
    labelfar
%mend;

%macro displayStats;
  displaystats=(n mean median max datamax min datamin  iqr range std q1 q3)
%mend;

%macro datalabelAttrs;
  datalabelattrs=(size=6pt color=blue Style=Normal Weight=Normal) boxwidth=1
%mend;

ods region;

  proc sgplot data=sashelp.cars;
    vbox invoice /
      %dataLabel
      %displayStats
      %dataLabelAttrs
    ;
  run;

ods region;

  proc sgplot data=sashelp.cars;
    vbox horsepower/
      %dataLabel
      %displayStats
      %dataLabelAttrs
    ;
  run;

ods region;

  proc sgplot data=sashelp.cars;
    vbox mpg_city/
      %dataLabel
      %displayStats
      %dataLabelAttrs
    ;
  run;

ods region;

  proc sgplot data=sashelp.cars;
    vbox mpg_highway/
      %dataLabel
      %displayStats
      %dataLabelAttrs
    ;
  run;

ods layout end;
ods pdf close;

The PDF produced is attached.

Re: Multiples box plots

BrunoMueller — Thu, 26 Feb 2026 08:15:29 GMT

You are not using ODS LAYOUT and you do not specify the size of the image to create, so I guess that is why you do not have them on the same page.

Re: Multiples box plots

BrunoMueller — Wed, 25 Feb 2026 12:36:27 GMT

How about using ODS LAYOUT, see sample below.

ods _all_ close;

ods graphics / width=80mm height=80mm outputfmt=static;
ods pdf file="c:\temp\layout_gridded.pdf";

ods layout gridded columns=2;

ods region;
proc sgplot data=sashelp.cars;
vbox invoice;
run;

ods region;
proc sgplot data=sashelp.cars;
vbox horsepower;
run;

ods region;
proc sgplot data=sashelp.cars;
vbox mpg_city;
run;

ods region;
proc sgplot data=sashelp.cars;
vbox mpg_highway;
run;

ods layout end;
ods pdf close;

Re: How to use the Sample: Load cas-shared-default Public data job to load SAS7BDAT files?

BrunoMueller — Mon, 09 Feb 2026 09:50:43 GMT

Please note the filter is case sensitive.

I am not aware of a way to see the actual code being used.

it will help if we see the filter definition.

A look into the future, Upcoming Critical Changes

BrunoMueller — Thu, 11 Dec 2025 12:44:47 GMT

The SAS documentation provides a "What's New" section. This will tell you about new functionality and features within the SAS software. You will see this in the doc for a Stable or Long-Term Support release when it comes out.

Being able to know what the future holds can be an advantage. For SAS Viya there is now a way to see upcoming critical changes. Having this information will allow for better planning when you think about your next SAS Viya release upgrade.

Re: PROC SGPLOT axis break with different distribution/scale before and after the break

BrunoMueller — Wed, 26 Nov 2025 13:49:03 GMT

As a start have a look here https://blogs.sas.com/content/graphicallyspeaking/2012/05/31/broken-y-axis/ it provides a code sample that does the break of an axis "manually". You will have to use the GTL (Graph Template Language). I have taken the above example and adapted it for a broken X axis. Since you did not provide any data and what kind of plot you want I assumed something.

See this sample:

data plotdata;
  do month = 1 to 14;
    value = rand("integer", 10, 20);
    output;
  end;
run;

/*-- manual break in axis --*/
proc template;
  define statgraph series_break;
    begingraph;
      /* the columweights determine the space available */
      layout lattice / columns=2 rowdatarange=union columnweights=(0.9 0.1) ;
        layout overlay / 
          xaxisopts=(
            griddisplay=on display=(ticks tickvalues label) 
            tickvalueattrs=(size=7)
             linearopts=(viewmin=1 viewmax=11 TICKVALUELIST=(1 2 3 4 5 6 7 8 9 10))
             offsetmax=0
          );
          seriesplot x=month y=value;
        endlayout;
        layout overlay / 
          xaxisopts=(
            griddisplay=on display=(ticks tickvalues) 
            tickvalueattrs=(size=7)
            linearopts=(viewmin=11 TICKVALUELIST=(11 12 13 14))
            offsetmin=0
          )
          yaxisopts=( display=none)
        ;
          seriesplot x=month y=value;
        endlayout;
      endlayout;
    endgraph;
  end;
run;

ods graphics / reset=all;

proc sgrender data=plotdata template=series_break;
run;

Re: Guidance on Visualizing Postal Code Areas in SAS Visual Analytics

BrunoMueller — Wed, 12 Nov 2025 13:46:41 GMT

It is best to start with the documentation: https://go.documentation.sas.com/doc/en/vacdc/v_035/vareportdata/p031vp9uc5y5iun0zipy3c1trkqn.htm please choose the proper version that you are using.

Basic Steps:

First you will need the corresponding polygon information for your PLZ8 regions. The easiest would be to have the data as SHP file.
The SHP file needs to be imported as a CAS table, see Loading Geographic Polygon Data as a CAS Table
Define your own Geographic Data Provider, you need to be an administrator (by default only Administrators can do this). See Create a Geography Data Item By Using Custom Polygonal Shapes, Lines, or Coordinates from a Geographic Data Provider

CAS Table State Management for everybody

BrunoMueller — Mon, 18 Aug 2025 08:10:27 GMT

The CAS Table State Management, is a functionality available to the SAS Administrator. The CAS table state management enables you to manage the import and load of source files as well as the unload of tables from memory in caslibs. It is made available to the SAS administrator using predefined sample jobs.

This article will show how these jobs can be made available to a normal user, so that he can take advantage of the functionality provided by these jobs.

Re: Modify column length of table in a caslib

BrunoMueller — Thu, 31 Jul 2025 10:04:11 GMT

Yes, I forgot about that when input and outout are both from CAS it runs in CAS, thanks for the correction. I will make a change to my post.

Re: Modify column length of table in a caslib

BrunoMueller — Thu, 31 Jul 2025 10:05:57 GMT

I was wrong about this "The previous example with the DATA step runs on the compute server, all data will be read from CAS and written back to CAS. Depending on the size this may take some time." thanks for the correction @yabwon .

The CAS server supports executing FedSQL, not all statements are supported see https://go.documentation.sas.com/doc/en/pgmsascdc/v_065/casfedsql/p0ipyarz6fe3mvn1hc8ega07irv6.htm for more details.

We can make use of the casting functionality within FedSQL to change the length of char or varchar column. Below is an example program that does this. It is executing on the CAS server so no data transfer between CAS and compute.

/*
 * change length of a char/varchar column
 * one needs to make a copy of an existing table
 *
 * the create table does support the {options replace=true} table option
 * however FedSQL will always create a session scope table
 */
cas sugus sessopts=(caslib="casuser");

/*
 * load some sample data, has session scope
 */
proc casutil;
  load data=sashelp.cars casout="mycars" replace;
run;
quit;

title "columns before";
proc cas;
  action table.columninfo / table="mycars";
  run;
quit;
title;

proc cas;
  /*
    * build column expressions
    * we use the CAST()/:: function to change the length
    */
  action table.columninfo result=cols / table={caslib="casuser" name="mycars"};
  sqlCols=" ";
  do colInfo over cols.columnInfo;
    colName=colInfo.column;

    colExpression=colName;

    /* change as needed */
    if colName="Model" then do;
      colExpression=catx(" ", quote(colName), "::varchar(64) as", quote(colName));
    end;
    /* change as needed */
    if colName="DriveTrain" then do;
      colExpression=catx(" ", quote(colName), ":: varchar(32) as", quote(colName));
    end;

    sqlCols=catx(", ", sqlCols, colExpression);
  end;
  print(note) sqlCols;
  run;
  action table.droptable / caslib="casuser" name="mycars2" quiet=true;
  run;
  /*
    * create complete SQL statement
    * be aware that fedsql will always create a session scope table
    */
  sqlStmt=catx(" " , "create table casuser.mycars2 as select" , sqlCols ,
      "from casuser.mycars;" );
  print(note) sqlStmt;
  action fedsql.execdirect / query=sqlStmt;
  run;
quit;


title "columns after";
proc cas;
    action table.columninfo / table="mycars2";
    run;
quit;
title;

Re: Alternative to Auditing for Report Usage in SAS Viya Visual Analytics

BrunoMueller — Mon, 21 Jul 2025 08:22:23 GMT

In SAS Viya 3.5 the activity entries are not available.

See this article https://communities.sas.com/t5/SAS-Communities-Library/Auditing-Report-Actions-in-SAS-Viya-3-x/ta-p/959712 by for more details on auditing report operations.

Or go the way as outline by @gwootton .

Re: Alternative to Auditing for Report Usage in SAS Viya Visual Analytics

BrunoMueller — Wed, 16 Jul 2025 14:16:49 GMT

Please provide the version information on SAS Viya you are interessted in.

As of SAS Viya 20nn.nn there are no longer traditional log files.

The information on which reports have been accessed, can be returned using the sas-viya audit list-activities command (tested with Stable 2025.06).

Here is an example:

sas-viya --output json audit list-activities --application-contains Visual --sort-by timeStamp --limit 100 --action read --user-interaction-contains Open

The command will return JSON output. A activity entry will look like below.

    {
      "action": "read",
      "administrativeAction": false,
      "application": "SAS® Visual Analytics",
      "correlator": "dfd7180f-c19f-4dd8-847a-2b35f1dbebc3",
      "httpContext": {
        "method": "GET",
        "statusCode": 200
      },
      "id": "7fca62e6-ccc1-4aba-8897-2d10efdffb1a",
      "links": [
        {
          "href": "/audit/activities/7fca62e6-ccc1-4aba-8897-2d10efdffb1a",
          "method": "GET",
          "rel": "self",
          "type": "application/vnd.sas.audit.activity",
          "uri": "/audit/activities/7fca62e6-ccc1-4aba-8897-2d10efdffb1a"
        }
      ],
      "objectName": "cars-simple",
      "objectType": "report",
      "properties": {
        "referringApplication": "SASVisualAnalytics"
      },
      "references": [],
      "secure": true,
      "serviceName": "reports",
      "sessionId": "a8bbbb4b-d8f0-4e38-ab06-df83840d430b",
      "state": "success",
      "timeStamp": "2025-07-16T13:43:50.037Z",
      "type": "resource",
      "uri": "/reports/reports/130a9b58-d93b-4f24-8478-ff02d416ff3b/content",
      "user": "sean",
      "userInteraction": "Open report cars-simple",
      "version": 8
    }

A non admin user will only see his entries.

A SAS admin will see all entries

There are many more options to the mentioned command. For instance to select a specifc time window using --after and|or --before. Note the --limit option, by default only 50 entries are returned.