About BrunoMueller

BrunoMueller

@touwen_k glad this helps you. I am not aware of a way to display a different message. As you can see in the example, there is a field called "reason" but it is not surfaced to the user.

BrunoMueller

I see. These options are not currently available for custom graphs, but can be set while building the custom graph. Here are the basic steps: create a new custom graph using the "New Custom Graph" button From the elements on the left add bar chart Go to the options of the bar chart on the right, set grouping scale and data label as shown Add the group role to the bar chart as shown before Add a line graph on top of the bar chart Set options of the line chart accordingly (choose right Y axis for the line chart) Maybe rename the "shared" role to category Check what you would like to have in the legend Proceed as mentioned before (save custom graph, import it, use it). It will look like this: If you need to make changes to the custom graph. Delete it from the object window. Make the changes. Import it again. Build new report object.

BrunoMueller

This is possible using the SAS Graph Builder. In the SAS Graph Builder you will find a "template" called "Dual Axis Bar-Line Chart" Use it as starter, add the group role to the the bar chart The default for group display is cluster, as you need it. Save the graph template, you will be asked to remove the data driven lattice roles. In a SAS Visual Analytics report import the graph template You are now ready to use the graph template like any other object

BrunoMueller

During a recent class on SAS Viya Platform Administration, I was asked, if there was a way to put SAS Viya into "admin mode". Meaning only the SAS Administrators should have access, all other users would not be able to use the SAS Viya environment. I guess the question originated from the fact, that this is possible in SAS9.4 with pausing a metadata server for Administration. So lets look at a way of doing this in SAS Viya. How to approach this The general authorization system controls all access to applications, services and SAS Content (folders, reports etc.). It uses a set of rules that determine the individual access for a user. So we start here. The general authorization system is based on these principles: implicitly disallow any access that is not granted a Prohibit always wins A rule has these basic elements: Element Description Target Can be a service, application, individual object etc. The target is represented as an URI (Uniform Resource Identifier). Principal The user, group or custom group to which the rule is assigned. Permission Specifies the type of access, such as read, create, update, delete, add, remove, and secure. Setting Determines whether access is provided (grant) or not allowed (prohibit). The setting can be conditional, using a constraint expression. So we need to find a rule that affects everything, all applications, services etc. Looking at the documentation we will find this: Rule's Target URI Rule's Relevance /** Relevant to all requests. The rule to control all access You might try to create a new rule with the following elements: Element Content Target /** Principal Authenticated Users Permission Read Setting Prohibit But wait, SAS Administrators are Authenticated Users too, so this rule would also block access for SAS Administrators. So nobody would be able to use the system. Never do this. Please read on to find a better way to control access, with conditional prohibit. Using a conditional prohibit, we can set the setting based on a condition. We only want to apply the Prohibit if the requesting user is not a member of the of the SAS Administrators group. So our new rule would have the following elements: Element Content Target /** Principal Authenticated Users Permission Read Setting Conditional Prohibit Condition !(groupsForCurrentUser().contains('SASAdministrators') || groupsForCurrentUser().contains('sasapp')) Conditions are written using the Spring Expression Language (SpEL). Special functions are available like the groupsForCurrentUser(). You will find more functions in the documentation. Let's look at the condition in detail: Element Description ! Stands for NOT ( Start of a group groupsForCurrentUser().contains('SASAdministrators') check if current user is member of the SASAdministrators group || Stands for OR groupsForCurrentUser().contains('sasapp') check if current user is member of the sasapp group. This is needed for internal users. ) End of a group When using conditions in a rule this will happen: A rule that has a condition that evaluates to true for a particular access request is applied in the authorization decision process for that access request. A rule that has a condition that evaluates to false for a particular access request is ignored in the authorization decision process for that access request. So any user that is not a member of the SAS Administrators or the sasapp group, the rule will apply. Create the actual rule The rule can be created using the Rules page in the SAS Environment Manager or the sas-viya command line interface. We are going to use the sas-viya authorization create-rules command together with a file. This has the following advantages: we can specify a rule-id for easier handling afterwards we can specify the state of the rule to be disabled, so it does not have an immediate effect The file content looks like this, note the id and enabled keys and values: [ { "op": "add", "value": { "objectUri": "/**", "principalType": "authenticatedUsers", "type": "prohibit", "condition": "!(groupsForCurrentUser().contains('SASAdministrators') || groupsForCurrentUser().contains('sasapp'))", "permissions": [ "read" ], "description": "disallow access except for SASAdministrators, sasapp groups", "id": "offline-mode-sasadministrators-only", "reason": "SAS Viya only available to SAS Administrators", "enabled": false } } ] This JSON format is documented under Patch authorization rules. Please note the id uses 3 hyphens in the name, this is important as otherwise the rule will not be found. To create this rule we use this command: sas-viya authorization create-rules --file offline-mode.json The result of the command will tell us, that 1 rule has been created. To check the rule just created, we use this command: sas-viya --output fulljson authorization show-rule --id offline-mode-sasadministrators-only. Note the --output fulljson to get back the complete JSON structure. Enable the rule Since we created the rule as disabled, it will not have any immediate effect. To enable it we use the following command: sas-viya authorization enable-rule --id offline-mode-sasadministrators-only The response looks like this: Select any image to see a larger version. Mobile users: To view the images, select the "Full" version at the bottom of the page. Now every user that is not part of the groups SAS Administrators or sasapp will no longer have any access to any of the applications, services etc. If a user wants to access a SAS Application the following message is displayed in the browser: Likewise, if you want to access any of the SAS Viya API's you will receive this response: { "version": 2, "httpStatusCode": 403, "message": "Forbidden", "details": [ "Unauthorized", "path: /folders/folders/@myFolder", "correlator: adcd63b3-55b4-4caa-b65d-c2e57cec9b9f" ] } The HTTP status code 403 stands for Forbidden. A user can still authenticate to SAS Viya but then no further access is possible Disable the rule To disable the rule use the following command: sas-viya --yes-to-all authorization disable-rule --id offline-mode-sasadministrators-only All users can now work as before. Summary We have seen that by using a specific rule with a condition in the general authorization system we can block non administrator users from working with SAS Viya. This rule can easily be enabled or disabled as needed. Always be very careful when using Prohibit together with the Authenticated User principal as this will include any SAS administrator as well. Find more articles from SAS Global Enablement and Learning here.

BrunoMueller

I would start here Why you need a log and metric monitoring solution for the SAS Viya platform for an introduction. SAS provides the SAS Viya Monitoring for Kubernetes to get started.

BrunoMueller · ‎06-24-2024

Just run proc javainfo all; run; And you will see more details.

BrunoMueller · ‎06-12-2024

Have a look at the sample code below. The original target variable contents is split into two variables so that we can use the nesting of the across usage. /* * create dummy data */ data have; infile cards dlm=","; input week : date9. target : $32. value : 8. ; /* * split target into two variables */ target2 = scan(target, 1, " "); strain = scan(target, 2, " "); format week date9.; cards; 27May2024,Pox S1,19 27May2024,Pox S2,25 27May2024,Ebola A,4 27May2024,Ebola B,. 27May2024,Prion B,. 27May2024,Prion H,. 28May2024,Pox S1,10 28May2024,Pox S2,30 28May2024,Ebola A,4 28May2024,Ebola B,1 28May2024,Prion B,2 28May2024,Prion H,. ; /* * create a format to print missing values * as requested */ proc format; value mynumber . = "not detected" other = [8.] ; run; /* * proc report code * nocompletecols only print the actual values * group usage for week so that multiple rows are reduced to one * use of " " as label to suppress printing * target2 and strain used as across, the comma in the column * statement indicates the nesting */ proc report data=have nocompletecols; column ("week" week) target2, strain, value; define week / " " group; define target2 / " " across ; define strain / " " across; define value / " " analysis format=mynumber. center; run; /* * alternate way using proc tabulate */ proc tabulate data=have format=mynumber.; class week target2 strain; var value; table week = " ", target2 = " " * strain = " " * value = " " * sum = " "; run;

BrunoMueller · ‎06-11-2024

There is a rule that looks like this, see below. By default Authenticated Users is the principal type. If like in the example below you assign a certain group, only members of this group can run programs. Other users will receive an error when they try too. There is a similar rule for background processing using the ObjectURI /studio/sessions/*/background/submissions. By default Authenticated Users are allowed to do this. This might help, however you still have the complete SAS Studio GUI available which might be a bit to much for the casual data viewer. Also what was mentioned before about ACCESS=READONLY is important. As an alternative have a look at https://datacontroller.io It might just do what you need (I do not have personal experience with it).

BrunoMueller · ‎06-10-2024

You might have noticed, that the id value is only in the first row (when a new id value starts), after that it is not visiable and therefor the id variable is empty. You need to put the value of the first row for each id value aside and use this in the compute block. To put the value aside, use a COMPUTE BEFORE columnName compute block. See the example below. I use the _dummy variable to display the contents of the idkeep variable. proc report data=sample nowd; column id value1 value2 _dummy; define id / 'ID' display order order=data; define value1 / sum 'Value 1'; define value2 / sum 'Value 2'; define _dummy / computed; compute before id; idkeep = id; endcomp; compute _dummy; _dummy = idkeep; if idkeep = 1 and value1.sum > value2.sum then do; call define("value1.sum", "style", "style={background=Red}"); end; endcomp; run;

BrunoMueller · ‎06-10-2024

Made some changes to the Proc REPORT code, see below. Here are the details for the changes: Added a _dummy column as the last column. In a compute block for a given column you can only access the values of the columns to the left of the current column. So by adding the _dummy column, I can access all the column values to the left in the _dummy compute block. The value of the _dummy column is not printed (NOPRINT) added ORDER to the ID variable to display each value Since value1 and value2 are analysis variables you need to access the value using columnName.statistic in our case value1.sum or value2.sum This name can also be used in the call define to reference the proper column change foreground to background in the call define proc report data=sample nowd; column id value1 value2 _dummy; define id / 'ID' order order=data; define value1 / sum 'Value 1' ; define value2 / sum 'Value 2' ; define _dummy / computed noprint; compute _dummy; if value1.sum > value2.sum then do; call define("value1.sum", "style", "style={background=Red}"); end; endcomp; run;

BrunoMueller · ‎06-06-2024

When you change your code to this, adapted to your names used proc cas; action table.tableInfo result=r2 /caslib="CASUSER", name="myClass"; do row over r2.tableinfo; print(note) row; end; run; quit; You will find that you get the following dictionary back. Note you have formatted values that take the timezone into consideration and unformatted values that are the UTC values. You are using the UTC values. { Name=MYCLASS, Rows=19, Columns=5, IndexedColumns=0, Encoding=utf-8, CreateTimeFormatted=2024-06-06T17:45:28+02:00, ModTimeFormatted=2024-06-06T17:45:28+02:00, AccessTimeFormatted=2024-06-06T17:45:28+02:00, JavaCharSet=UTF8, CreateTime=2033307928.2, ModTime=2033307928.2, AccessTime=2033307928.2, Global=0, Repeated=0, View=0, MultiPart=0, SourceName=, SourceCaslib=, Compressed=0, Creator=someuser, Modifier=, SourceModTimeFormatted=, SourceModTime=., TableRedistUpPolicy=Not Specified }

BrunoMueller · ‎06-05-2024

You can adjust the timezone information when creating a CAS session, see here for more information https://go.documentation.sas.com/doc/en/pgmsascdc/v_049/casref/n02215l0eb3lxtn13piutithjc32.htm Have look at the code below, there are two CAS statements, use only one of them. One sets the timezone to the same value as in your SAS session, the other goes with the default. Have a look at the results of the table.tableinfo action for the difference. %let timeZone = %sysfunc(getoption(timezone)); %put NOTE: &=timezone; cas sugus sessopts=(caslib="casuser" timezone=&timeZone); * cas sugus sessopts=(caslib="casuser"); data myclass; set sashelp.class; run; proc casutil; load data=myclass casout="myclass" replace; run; quit; proc cas; action table.tableinfo / caslib="casuser"; run; quit; cas sugus terminate;

BrunoMueller · ‎05-27-2024

I do not understand the question. The bar chart to the left uses the "normalize groups to 100%", it filters the list table below using an action. The bar chart to the right uses the technique with the calculated data items as described above, it filters the list table below. Are you doing something different?

BrunoMueller · ‎05-15-2024

Sure just add the variables needed specify NOPRINT for the once you use in a COMPUTE BEFORE variable block. proc report data=sashelp.orsales; column year quarter product_line quantity; define year / group noprint; define quarter / group noprint; define product_line / group; define quantity / analysis; compute before year; line year 4.; endcomp; compute before quarter; line quarter $6.; endcomp; run;

BrunoMueller · ‎05-15-2024

You are welcome. Maybe it makes sense to buy this book https://support.sas.com/content/dam/SAS/support/en/books/the-sas-programmers-proc-report-handbook/69155_excerpt.pdf

Online Status	Offline
Date Last Visited	yesterday