About BrunoMueller

BrunoMueller · ‎05-29-2020

Give the code below a go. We do need to create some output to get the first bookmark ods pdf file="c:\temp\want.pdf" startpage=no; ODS PROCLABEL 'ods proc label'; data _null_; declare odsout ODS(); ODS.note(data: " ", style_attr: 'font_size=1pt'); run; ods pdf nobookmarkgen; proc sgplot data=sashelp.class; scatter x=age y=height; run; ods pdf startpage=yes; proc sgplot data=sashelp.class; scatter x=age y=weight; run; ods pdf close;

BrunoMueller · ‎05-27-2020

You could build your own tool for this, using the REST API provided. The basic structure would be: 1. Get the report content from the source system: GET https://yourserver/reports/reports/14990fe5-6a32-409c-bdd2-3da6c89479f2/content Doc: https://developer.sas.com/apis/rest/Visualization/#get-report-content 2. Next create a new Report in a new folder on the target system: POST https://yourserver/reports/reports?parentFolderUri=/folders/folders/75d5f145-80ae-4271-9628-ac2f620f92dd Doc: https://developer.sas.com/apis/rest/Visualization/#create-report 3. Update the newly created report: PUT https://yourserver/reports/reports/3a200423-7a9d-4785-8a02-02b20b513373/content Doc: https://developer.sas.com/apis/rest/Visualization/#update-a-report The calls above contain URI's like "/reports/reports/3a200423-7a9d-4785-8a02-02b20b513373" My colleague @XavierBizoux created some python scripts to work as outlined above. The scripts are currently missing: create a folder if it does not exist script does not cater to put report into a new folder, this would need to be added But it is a starting point. You will find the scripts here: https://github.com/xavierBizoux/SAS/tree/master/CICD/Reports Also have a look at https://github.com/sassoftware/pyviyatools for additional examples on how to do REST calls from Python Find attached a SAS program I used for verification of the three steps. You will need to adapt the URI's for you environment.

BrunoMueller · ‎05-18-2020

You could go with the approach outlined here https://blogs.sas.com/content/graphicallyspeaking/2012/05/31/broken-y-axis/ this technique was used when the RANGES= option was not available. See a sample using SERIESPLOT below: data broken_axis; infile cards; x = _n_; input y y2; cards; 100 8 120 15 655 70 688 69 180 08 120 15 655 70 688 73 180 12 ; /* * just to get basic graph and GTL */ proc sgplot data=broken_axis tmplout="c:\temp\sgplot.sas"; series x=x y=y; series x=x y=y2 / y2axis; run; /*--Full Panel Break--*/ proc template; define statgraph Panel_Break; begingraph; entrytitle 'Example broken axis'; layout lattice / rows=2 columndatarange=union rowweights=(0.35 0.65) rowgutter=2; columnaxes; columnaxis / display=(ticks tickvalues) tickvalueattrs=(size=7); endcolumnaxes; /* upper part */ layout overlay / cycleattrs=true yaxisopts=(griddisplay=on display=(ticks tickvalues) tickvalueattrs=(size=7) offsetmin=0 linearopts=(viewmin=650 viewmax=750 tickvaluelist=(650 700 750))) y2axisopts=(griddisplay=on display=(ticks tickvalues) tickvalueattrs=(size=7) offsetmin=0 linearopts=(viewmin=65 viewmax=75 tickvaluelist=(65.0 70.0 75.0))); seriesplot x=x y=y / datalabel=y; seriesplot x=x y=y2 / yaxis=y2 datalabel=y2; endlayout; /* lower part */ layout overlay / cycleattrs=true yaxisopts=(griddisplay=on display=(ticks tickvalues) tickvalueattrs=(size=7) offsetmin=0 linearopts=(viewmin=0 viewmax=210 tickvaluelist=(0 100 200))) y2axisopts=(griddisplay=on display=(ticks tickvalues) tickvalueattrs=(size=7) offsetmin=0 linearopts=(viewmin=0 viewmax=21 tickvaluelist=(0 10 20))); seriesplot x=x y=y / name="y" datalabel=y; seriesplot x=x y=y2 / yaxis=y2 name="y2" datalabel=y2; DiscreteLegend "y" "y2"; endlayout; endlayout; endgraph; end; run; proc sgrender data=broken_axis template=Panel_Break; run;

BrunoMueller · ‎05-13-2020

Have a look here https://developer.sas.com/apis/rest/CoreServices/#get-a-resource-39-s-ancestors to get the folder information of a report SAS sample code looks like: proc http url="&BASE_URI/folders/ancestors?childUri=/reports/reports/72c7e418-cd68-49c7-8f2f-6e9548639821" method='GET' oauth_bearer=sas_services out=resp ; run; %put NOTE: resp json pretty print; %put %sysfunc( jsonpp(resp, log)); libname resp json fileref=resp; proc copy in=resp out=work; run;

BrunoMueller · ‎05-11-2020

It is a while since the question came up, there is a way to do this. The beginning will be the same: create a dropdown to select the one to highlight and put the value into a parameter then create a new calculated item as follows: IF ( 'Continent Name'n = 'selectedContinent'p ) RETURN '_#_' ELSE Concatenate('Continent Name'n, '_#_') Use the new item as group element in your line chart, do not display any legend Create a new display rule using the new calculated item: use your favorite color for the "_#_" value and some other for all the others

BrunoMueller · ‎04-27-2020

@Ivan555 the big table being in the DBMS, explains the long runtime, as all the rows have to be passed from the DBMS to SAS to then do the join. Do avoid the transfer of all the rows from the DBMS to SAS, you have several possibilities: copy the small table to the DBMS, so the join can be made within the DBMS, might no be an option depending on your rights create the list of IDs using the Proc SQL select id into :idList separated by "," and use the macro var in the IN operator, similar to how you typed it in, but this time all automatic If you have control over the LIBNAME statement pointing to the DBMS, have a look at the MULTI_DATASRC_OPT, it will do something similar, just all automatic

BrunoMueller · ‎04-27-2020

Question: are both tables stored as SAS Data Set, or is the big table coming from a DBMS?

BrunoMueller · ‎03-31-2020

To change the appearance of text on axis and legends check for the appearance options in the doc. To insert text you can use the INSET statement. See also example below proc sgplot data=Microbiolog; vbox AWCD / Group=Species category=Age_class; inset ( "Confidence level (*ESC*){unicode alpha} =" = "0.05" "some label" = "some value" ) ; xaxis discreteorder=data labelattrs=(color=red size=20pt) valueattrs=(color=green size=15pt) ; keylegend / valueattrs=(color=blue size=10pt) titleattrs=(color=black size=20pt) ; run; Also check out this blog entry for more details: https://blogs.sas.com/content/sgf/2017/09/15/proc-sgplot-theres-an-attrs-for-that/

BrunoMueller · ‎03-19-2020

I suggest you contact tech support on this.

BrunoMueller · ‎03-01-2020

The example first creates a SAS data set with a primary key, and then shows how the output of Proc CONTENTS for this SAS data set, along with several queries against different dictionary tables and how this primary key shows up in this dictionary tables. So you can use the queries in the example to get all SAS data sets with primary key definitions, just change the where clause as necessary

BrunoMueller · ‎02-26-2020

You can use Proc CONTENTS or use some of the dictionary tables avaialble with Proc SQL to display information on contraints defined for a SAS data set. Have a look at the code example below. See here for a listing of available dictionary tables https://go.documentation.sas.com/?docsetId=sqlproc&docsetTarget=n02s19q65mw08gn140bwfdh7spx7.htm&docsetVersion=9.4&locale=en data myClass; my_pk = _n_; set sashelp.class; run; proc sql; alter table myclass add primary key (my_pk) ; quit; ods trace on; ods output IntegrityConstraints=my_ic; proc contents data=work.myclass; run; ods trace off; proc sql; create table TABLE_CONSTRAINTS as select * from dictionary.TABLE_CONSTRAINTS where table_catalog = "WORK" and table_name = "MYCLASS" ; create table CONSTRAINT_COLUMN_USAGE as select * from dictionary.CONSTRAINT_COLUMN_USAGE where table_catalog = "WORK" and table_name = "MYCLASS" ; create table CONSTRAINT_TABLE_USAGE as select * from dictionary.CONSTRAINT_TABLE_USAGE where table_catalog = "WORK" and table_name = "MYCLASS" ; create table columns as select * from dictionary.columns where libname = "WORK" and memname = "MYCLASS" ; quit;

BrunoMueller · ‎02-17-2020

You show log with a table named class_format_test, but the message above is for a different table as well as a different format. Have you checked in VA with the table class_format_test in cas library public?

BrunoMueller · ‎02-17-2020

Please run the below code, it will show information about the cas table (similar to what you get with Proc contents). It will also run something similar to "proc freq" using the column with the user defined format. You might need to adjust names for caslib= and name= etc. depending on how you are testing. proc cas; action table.columninfo / table={caslib="public" name="class_format_test"}; action freqTab.freqTab / table={caslib="public" name="class_format_test"} tabulate={ {vars={'ageGroup'}} } display={ names={"OneWayFreqs"} } ; run; quit;

BrunoMueller · ‎02-13-2020

You are missing two critical elements I mentioned before: a global scoped table (promoted) needs to be deleted first so that it can be written again proc delete data=public.class_format_test; run; you have to promote the table so it has a global scope and can be seen in Visual Analytics data public.class_format_test(promote=yes); The NOTE message about the format "NOTE 484-185: Format MPGRATING was not found or could not be loaded." just says it can not be found in the SAS environment where the program is running, it will be found on the CAS server.

BrunoMueller · ‎02-12-2020

Your program will work, just a few things to note: numeric variables with a user defined format will be treated as category you have to promote the table so it has a global scope and can be seen in Visual Analytics a global scoped table (promoted) needs to be deleted first so that it can be written again when running the DATA Step the format has to be available to SAS, or use the NOFMTERR system option This code sample works for me: cas sugus sessopts=(caslib="casuser"); cas sugus listformats scope=global members; libname public cas caslib="public"; options nofmterr; proc delete data=public.class_format_test; run; data public.class_format_test(promote=yes); set sashelp.class; ageGroup = age; format ageGroup class_age_group.; run;

Online Status	Offline
Date Last Visited	‎12-17-2025 09:19 AM