About BrunoMueller

BrunoMueller · ‎08-12-2015

Hi Ram Find below as sample logconfig that I have used sucessfully. Please make sure, that all users starting a Workspace server do have write access to the location where the log file are written. If something is wrong with the XML then one get an entry in the system log giving the name of a file that indicates the type of error. The sample blow was made for Windows, so you have to change the location of the log file, also make sure, that you get unique filename for each process. <?xml version="1.0" encoding="UTF-8"?> <logging:configuration xmlns:logging="http://www.sas.com/xml/logging/1.0/" debug="true">    <appender name="AuditLibraryFile" class="FileAppender"> <param name="Append" value="true"/> <param name="ImmediateFlush" value="true"/> <param name="fileNamePattern" value="c:\temp\audit_dataset\audit.dataset_%d_%S{hostname}_%S{pid}.log"/> <layout> <param name="ConversionPattern" value="%d!%c!%u!jobid=%S{jobid} Action=%E{Audit.Dataset.Action} Status=%E{Audit.Dataset.Status} Libref=%E{Audit.Dataset.Libref} Engine=%E{Audit.Dataset.Engine} Member=%E{Audit.Dataset.Member} NewMember=%E{Audit.Dataset.NewMember} MemberType=%E{Audit.Dataset.Memtype} Openmode=%E{Audit.Dataset.Openmode} Path=%E{Audit.Dataset.Path} Sysrc=%E{Audit.Dataset.Sysrc} Sysmsg=%E{Audit.Dataset.Sysmsg} username=%S{user_name} startup_cmd=%S{startup_cmd} " /> </layout> <filter class="StringMatchFilter"> <param name="StringToMatch" value="Libref=SASHELP"/> <param name="AcceptOnMatch" value="FALSE"/> </filter> </appender>  <logger name="Audit.Data.Dataset.Open" additivity="false"> <appender-ref ref="AuditLibraryFile"/> <level value="Trace"/> </logger>  <logger name="Audit.Data.Dataset.Delete" additivity="false"> <appender-ref ref="AuditLibraryFile"/> <level value="Trace"/> </logger>  <logger name="Audit.Data.Dataset.Rename" additivity="false"> <appender-ref ref="AuditLibraryFile"/> <level value="Trace"/> </logger>  <root>  </root> </logging:configuration>

BrunoMueller · ‎08-11-2015

BrunoMueller · ‎08-07-2015

Hi Find below a code example where the FDELETE function is used in a macro %macro fdel(file); %local fref; %let fref = __fdel; %if %sysfunc( fileexist(&file) ) = 0 %then %do; %put WARNING: &sysmacroname &file does not exist; %return; %end; %let rc= %sysfunc( filename(fref,&file) ); %let rc= %sysfunc( fdelete(&fref) ); %let rc= %sysfunc( filename(fref, ) ); %mend; %let delfile = /tmp/deleteTest.txt; data _null_; file "&delfile"; put "test"; run; %fdel(&delfile) Bruno

BrunoMueller · ‎07-30-2015

Ody The VARTYPE function can not be used on its own, one has to know for which data set you want to check. See this example code for how to do it: SAS-macros/attrv.sas at master · friendly/SAS-macros · GitHub There is an autocall macro that checks the type of the contents of a macro variable, have a look here: SAS(R) 9.4 Macro Language: Reference, Fourth Edition Bruno

BrunoMueller · ‎07-15-2015

Hi You can use Proc HTTP to get the SAS Data Set and write it directly to the WORK Library. See sample code below. filename sample "%sysfunc(pathname(work))/sample_data.sas7bdat"; proc http method="get" url= 'http://www.unc.edu/~mfunk/dr/v1dot1/sample_data.sas7bdat' out=sample ; run; data new; set sample_data; run;

BrunoMueller · ‎07-09-2015

Carmen Is your time_L_ value a decimal number, meaning 0.01 is minute 1 and 0.59 is minute 59? Bruno

BrunoMueller · ‎07-08-2015

Hi Have a look at this blog post http://blogs.sas.com/content/sasdummy/2012/03/20/sas-program-by-processing/ by Chris Hemedinger It shows very nicely on how to approach this. It will show how to find unique values for a given variable and then have a loop for each value and run some code. Since you already have char variables that are the value for spliting up the data sets I would use either Month or YEARMONTH_KEY Bruno

BrunoMueller · ‎07-08-2015

Hi Carmen I tried to simulate the data based on your description. Find below sample code to creates some test data and creates a time variable. Please have look and tell us whether the sample data looks like yours. The checkHour is only for cross checking the result, and to generate the data. data have; do myDate = today() -1 to today(); do checkHour = 0 to 23; do min = 0 to 59; output; end; end; end; format myDate date9.; run; data want; set have; by myDate; if first.myDate = 1 then do; newMin = -1; end; newMin + 1; time = hms(0, newMin, 0); format time time8.; run; Bruno

BrunoMueller · ‎06-17-2015

Hallo basefan Ein Stored Process kann als ein RESTful WebService aufgerufen werden. Im Folgenden werden die einzelnen Schritte erläutert. Dabei sollte man auch noch die Docu ( SAS(R) 9.3 BI Web Services: Developer's Guide) beiziehen. Der STP akkzeptiert einen SAS Data Set Namen als Input und kann eines der geforderten XML Files zurückgeben. Code für den STP: libname outxml xml; libname mvarinfo xml ; data outxml.outputTable; set &tableName; run; proc sql; create table mvarinfo.mvarinfo as select scope , name , resolve(value) as value from dictionary.macros order by name ; quit; Der STP hat folgende Parameter Definition: Folgende Data Targets werden definiert: Dieser STP kann mit folgendem SAS Program (siehe unten) als RESTful WebService aufgerufen werden. Der Name des STP (voller Pfadname) lautet: /stp_samples/myWS/ws_returnTable Der Name des STP muss im XML welches die Input Parameter enthält mitgegeben werden. Mit hilfe des Zusatzes streams/... kann gewählt werden welches XML file zurückgegeben werden soll. Im Beispiel unten entweder .../streams/outxml oder .../streams/mvarinfo verwenden filename restin temp; data _null_; infile cards; input; file restin; put _infile_; cards4; <ws_returnTable> <parameters> <tableName>sashelp.class(obs=1)</tableName> </parameters> </ws_returnTable> ;;;; filename restout temp; filename resthead temp; proc http ct="application/xml" method="post" url="http://chcent94:7980/SASBIWS/rest/storedProcesses/stp_samples/myWS/ws_returnTable/streams/outxml" in=restin headerout=resthead out=restout verbose ; run; data _null_; infile resthead ; input; putlog _infile_; run; data _null_; infile restout ; input; putlog _infile_; run; filename restin clear; filename restout clear; filename resthead clear; Viel Spass mit SAS Bruno

BrunoMueller · ‎06-17-2015

FriedEgg, Many thanks for the great example. Bruno

BrunoMueller · ‎06-16-2015

Currently I can not think of a simple solution to get around the lost card message. As someone already mentioned there are pure Lua based JSON parsers, maybe this will work together with Proc LUA. Bruno

BrunoMueller · ‎06-14-2015

You get the lost card, because not all the "records" have all the fields. The sample code provided runs fine, as long as all the "records" have the same fields. Maybe there is a way to add additional logic to find out whether a field exists or not. This is a sample record { "connectedTo": "08003164071", "serviceName": "05 Direct", "callGuid": "01456958-bf2e-48ea-b22c-d3f6f5a9a1c2", "origin": "02082481633", "stateChangedAt": "2014-04-16T07:20:22Z", "sequence": 6, "appletName": "TM Out Of Hours", "event": "End", "state": "Caller", "duration": "00:00:04" }, Record with missing fields { "connectedTo": "01179817946", "serviceName": "Rhiannon Green", "callGuid": "01456973-0ddd-4438-9f56-eb41490f2e2d", "origin": "01173003658", "stateChangedAt": "2014-04-16T07:46:20Z", "sequence": 1, "appletName": "AN Calls Recorded", "event": "NewApplet", "state": "", "additionalParameters": { "applet Type": "Announcement" } },

BrunoMueller · ‎06-12-2015

Hi Have a look at the example below, it can read your json string quite well. I guess the key is to have modified list input (the colon before the informat name) together with the DLM= option to say data values are delimited either by a , or a } filename sugus dataurl 'data: { "count":8141, "calls":[ { "connectedTo":"01179817787", "serviceName":"05 Direct", "callGuid":"014cc38e-1ac5-44ee-8fdc-1176b9d83632", "origin":"", "stateChangedAt":"2015-04-17T00:19:25Z", "sequence":5, "appletName":"TM Out Of Hours", "event":"End", "state":"Caller", "duration":"00:01:13" }, { "connectedTo":"123456789", "serviceName":"05 Direct2", "callGuid":"014cc38e-1ac5-44ee-8fdc-1176b9d83632", "origin":"", "stateChangedAt":"2015-04-17T00:18:12Z", "sequence":1, "appletName":"AN Welcome Message", "event":"NewApplet", "state":"", "ringDuration":"00:00:00", "duration":"00:00:00", "additionalParameters":{ "applet Type":"Announcement" } } ] }' ; data TEST; INFILE sugus LRECL = 3456677 TRUNCOVER SCANOVER dsd dlm=",}" ; INPUT @'"connectedTo":' connectedTo : $50. @'"serviceName":' serviceName : $50. @'"callGuid":' callGuid : $50. @'"origin":' origin : $50. @'"stateChangedAt":' stateChangedAt : $50. @'"sequence":' sequence : $50. @'"appletName":' appletName : $50. @'"event":' event : $50. @'"state":' state : $50. @'"duration":' duration : $50. @@; run; Bruno

BrunoMueller · ‎06-11-2015

If you post the SAS log it is easier to see what is going on

BrunoMueller · ‎06-11-2015

The double quotes should work just fine in Proc SQL, since SAS accepts either single or double quote. You are missing a semicolon at the end of the SELECT statement. If you want to go with the single quotes, you can use the %TSLIT() autocall macro like below: %macro temp(val=A); proc sql; select * from sashelp.class where name like %tslit(&val%); quit; %mend; %temp(val=A) Bruno

Online Status	Offline
Date Last Visited	Thursday