About SASJedi

SASJedi · ‎02-08-2021

I no longer have access to a mainframe to test my code, but something like this should work: data _null_; /* Assign a fileref to the PDS */ _rc=filename("myPDS","mydata.my.pdsfile"); _did=dopen("myPDS"); if _did ne 0 then do; putlog "ERROR: Unable to open PDS to look at members"; _rc=filename("myPDS"); stop; end; do _i=1 to dnum(did); /* Get info for each PDS member */ MemberName=dread(did,i); _rc=filename("thisMem",cats("myPDS(",MemberName,")","S"); _fid=fopen("thisMem"); do _j=1 to foptnum(_fid); infoname=foptname(_fid,_j); infovalue=finfo(_fid,infoname); output; end; _rc=fclose(_fid); _rc=filename("thisMem"); end; _rc=dclose(_did); _rc=filename("myPDS"); run; Hope this at least gets you started 🙂

SASJedi · ‎02-08-2021

You say silly, I say prudent. Potato, poTAHto.... In my 25 years of macro programming, I've expended many manhours troubleshooting my programs (an other people's programs) only to find that the root problem was a failure to quote text resolved from a macro process. There is no reason that a macro process should produce unquoted text unless the text being produced is executable macro code. So over the years, I've developed the habit of quoting all text produced by my macro processes unless I to execute it. And it's proved to be an excellent habit. It has significanlty reduced programming bugs, with a very satisfactory correlated reduction in debugging time. I dont intend to be precriptive here - feel free to use SYSFUNC if you prefer. But since I didn't intend to execut the result produced by the SAS function, I chose QSYSFUNC, and I stand by my choice...

SASJedi · ‎02-04-2021

Moved to Stat Procs for better visibility.

SASJedi · ‎02-02-2021

If you always analyze a range of consecutively named variables, try something like this: /* Set the macro variable value to the correct variable range */ %let range=analysis_202001 - analysis_202004; data test; /* Wherever you need the variable list, reference the macro variable. */ set temp (keep = product_number &range); run;

SASJedi · ‎02-02-2021

The simplest method is to use the SAS LIBNAME XMLV2 engine. Something like this would work: filename classXML 'C:\temp\class_data.xml'; libname classXML xmlv2; proc copy in=sashelp out=classXML; select class; run; libname classXML clear; filename classXML clear; This produces the following XML file: <?xml version="1.0" encoding="windows-1252" ?> <TABLE> <class> <Name>Alfred</Name> <Sex>M</Sex> <Age>14</Age> <Height>69</Height> <Weight>112.5</Weight> </class> <class> <Name>Alice</Name> <Sex>F</Sex> <Age>13</Age> <Height>56.5</Height> <Weight>84</Weight> </class> </TABLE>

SASJedi · ‎02-02-2021

Moved to Analytics -> Data Mining forum for better visibility.

SASJedi · ‎02-02-2021

So true, @ChrisNZ - versions of SAS before SAS 9.4M5 did not allow conditional macro statements in open code.

SASJedi · ‎02-01-2021

Tables of the same name must be written to a different SAS library or the new table will overwrite the existing table. The tables written by SAS code generated from a macro are in no way associated with the macro itself.

SASJedi · ‎02-01-2021

You need to use %QSYSFUNC to execute the EXIST function for this. For example: data a; input id x; cards; 1 10 2 20 ; run; data b; input id w; cards; 1 12 2 23 ; run; data d; input id y; cards; 1 30 2 50 ; run; data wanted; merge %if %qsysfunc(exist(a)) %then %do; a %end; %if %qsysfunc(exist(b)) %then %do; b %end; %if %qsysfunc(exist(c)) %then %do; c %end; %if %qsysfunc(exist(d)) %then %do; d %end; ; by id; run;

SASJedi · ‎02-01-2021

I think you would be better off using a DATA step for this. Check out the examples on the FILENAME EMAIL docs page.

SASJedi · ‎02-01-2021

I can't see what you are hoping to accomplish with your macro code. Remember, MACRO programs are used to write DATA steps. It's the DATA step program that processes the data. Before the DATA step can run, the MACRO must have already finished writing it. So I'm not seeing what it is your macro code is trying to accomplish. Can you describe the result you want?

SASJedi · ‎02-01-2021

The LAG function is you friend here, but make sure you always execute the LAG - never put a call to the LAG function in a conditional code block. Try this: data have; infile datalines truncover; input Group:$1. Year:4. Rate:10.; datalines; A 2019 9.090909 A 2020 27.27273 A 2021 31.81818 B 2018 22.72727 B 2019 72.72727 B 2020 95.45455 B 2021 100 C 2018 31.81818 C 2019 54.54545 C 2020 68.18182 C 2021 68.18182 ; data want; set have; by group; /* ALWAYS execute the LAG function - never make it conditional */ rate_2=sum(Rate,-lag1(rate)); if first.group then rate_2=rate; run; proc print; run;

SASJedi · ‎01-04-2021

Moved this post to Statistical Procedures, where it should get better exposure.

SASJedi · ‎12-14-2020

Use a macro variable to designate the attachment file when you create it. For example: /* Create the attachment file in the WORK library so it will be automatically deleted when the SAS session ends */ %let myFile=%sysfunc(pathname(work))\Report.pdf; ods pdf body="&myFile" style=Journal; PROC SQL; title "Details from sashelp.class where age > 10"; SELECT * from sashelp.class where age > 10; quit; ods pdf close; /* Set up this email using the PDF created above as an attachment*/ filename mymail email "my.recipient@example.com" subject="Report for sashelp.class" lrecl=256 type="TEXT/HTML" /* Attachment information */ attach="&myFile"; ods html5 body=mymail style=sasweb; options nocenter; title; /* Write the body of the email */ data _null_; file print; put "Here is the information you asked for."; put "May the SAS be with you!"; put "Mark"; run; ods html5 close;

SASJedi · ‎12-03-2020

If you could explain the desired logic for the calculated fields, I might be able to make some suggestions. .

Online Status	Offline
Date Last Visited	Saturday