About Tom

Tom · ‎10-05-2025

Before trying to use macro code and macro variables to generate code make sure you know what code you want to generate. Can you provide a simple example with 2 or three reference lines without any macro variables? Then explain what parts of the code needs to vary. Are you sure you need to use code generation to produce the graph you want? If you could explain the graph, perhaps include a picture of what you want it look like, then perhaps there is a way to make it directly from a dataset (or multiple datasets) instead. Also can you explain what the graph represents? Perhaps SAS already has a procedure to produce the graph you want.

Tom · ‎10-05-2025

Since the question says which of them COULD you import with PROC IMPORT than D is the right answer. For all of them you can use the DELIMITER= statement to set the appropriate delimiter. And for B you can use the GETNAMES= statement to tell there is no header line to use for making guesses about what names to use for the variables. But as to SHOULD you use PROC IMPORT to read them? I would say NO since none of them have so many variables or observations that would require you to use a guessing procedure like PROC IMPORT. Instead you could just write the data step to read them yourself. And for B the guesses that PROC IMPORT makes proc import file=b dbms=csv out=b replace; getnames=no; run; for how to write the data step to read it data WORK.B ; infile B delimiter = ',' MISSOVER DSD lrecl=32767 ; informat VAR1 $16. ; informat VAR2 $6. ; informat VAR3 $13. ; informat VAR4 $4. ; informat VAR5 $9. ; informat VAR6 $10. ; informat VAR7 $8. ; format VAR1 $16. ; format VAR2 $6. ; format VAR3 $13. ; format VAR4 $4. ; format VAR5 $9. ; format VAR6 $10. ; format VAR7 $8. ; input VAR1 $ VAR2 $ VAR3 $ VAR4 $ VAR5 $ VAR6 $ VAR7 $ ; run; will be much worse than what you would do. data b; infile b dsd truncover ; input Region :$20. Product :$20. Country :$20. Count (Price1-Price3) (:comma.); format price1-price3 dollar10.; run; Not only did PROC IMPORT not make any attempt to make meaningful names for the variables it mistakenly made every variable character even though the last four clearly look like nunbers. And the code it writes does silly things like: using the side effect of having the names first appear in INFORMAT statements to force the data step compiler to guess that the variable's type and length should match the informat type and informat width. adding the $ modifier in the INPUT statement for variables that have already been defined as character. attaching informats and formats to variables that do not need them. Also note that the code you need to write to read the file is almost smaller than the code you need to write to have PROC IMPORT guess how to write code to read the file. data a; infile cards dsd dlm='&' firstobs=2 truncover ; input Region :$8. State :$2. Month :anydtdte. Expenses Revenue; format Month monyy7. ; cards; Region&State&Month&Expenses&Revenue Southern&GA&JAN2001&2000&8000 Southern&GA&FEB2001&1200&6000 Southern&FL&FEB2001&8500&11000 Northern&NY&FEB2001&3000&4000 Northern&NY&MAR2001&6000&5000 Southern&FL&MAR2001&9800&13500 Northern&MA&MAR2001&1500&1000 ; data b; infile cards dsd truncover; input Region :$20. Product :$20. Country :$20. Count (Price1-Price3) (:comma.); format price1-price3 dollar10.; cards; "Africa","Boot","Addis Ababa","12","$29,761","$191,821","$769" "Asia","Boot","Bangkok","1","$1,996","$9,576","$80" "Canada","Boot","Calgary","8","$17,720","$63,280","$472" "Eastern Europe","Boot","Budapest","22","$74,102","$317,515","$3,341" "Middle East","Boot","Al-Khobar","10","$15,062","$44,658","$765" "Pacific","Boot","Auckland","12","$20,141","$97,919","$962" "South America","Boot","Bogota","19","$15,312","$35,805","$1,229" "United States","Boot","Chicago","16","$82,483","$305,061","$3,735" "Western Europe","Boot","Copenhagen","2","$1,663","$4,657","$129" ; data c; infile cards dsd dlm=' ' firstobs=2 truncover ; input Region :$7. State :$20. Capital :$30. Bird :$30.; cards; Region State Capital Bird South Georgia Atlanta 'Brown Thrasher' South 'North Carolina' Raleigh Cardinal North Connecticut Hartford Robin West Washington Olympia 'American Goldfinch' Midwest Illinois Springfield Cardinal ;

Tom · ‎10-05-2025

Are you asking us to do your homework? Please show what you have tried and if you still have questions explain them more completely.

Tom · ‎10-04-2025

I take it as a side effect of SAS's desired to maintain backwards compatibility. Quotes weren't required before, so they still aren't. The syntax for the LABEL statement is LABEL followed by zero or more NAME = LABEL pairs. Note that you do need quotes when not having them would change how the statement is interpreted (parsed?). For example if I had wanted a label with an equal sign in it. For example: label sex= 'M=Male F=Female' ; If I remove the quotes label sex= M=Male F=Female; and try it the SAS log will show: NOTE: Variable M is uninitialized. NOTE: Variable F is uninitialized. Because now I have three name=label clauses instead of one. But if the thing before the = is not valid as name label age=11=low 14=high; then it works. Note that where I see this syntax quirk used more often is in PROC FORMAT steps. proc format fmtlib ; value $sex M=Male F=Female; run;

Tom · ‎10-04-2025

Note that the ARRAY is not needed for this solution since you only need the list of variables once you can just place the list into the function call directly. select (max(of result1-result5));

Tom · ‎10-04-2025

@Ksharp wrote: John.King, But concatenation operator || would appear in graph. Wouldn't it get you a surprise ? Because at that point || is not an operator, it just another unquoted set of characters to add to the label. Just like if I used other "operators" as part of the label. data have; set sashelp.class (obs=1); label age= 'Word one' two 'three four' five sex= + | & and or || ; run; proc print label; run;

Tom · ‎10-03-2025

Remove the nesting. Just use three independent %IF blocks. %let current_month_num = %sysfunc(month(%sysfunc(today()))); %if (&current_month_num=1 or &current_month_num=2 ) %then %do; %put no report will be generated; %end; %if (&current_month_num=9 or &current_month_num=8 or &current_month_num=7) %then %do; %put quarter to date comparison report; %end; %if (&current_month_num=3 or &current_month_num=4 or &current_month_num=5 or &current_month_num=6 or &current_month_num=10 or &current_month_num=11 or &current_month_num=12 ) %do; %put not in Q3 and not in Jan or Feb; %end;

Tom · ‎10-03-2025

So it sounds like you have these two datasets (aka "tables"). MAILS mailno,model,date CLIENTS clientno,mailno How long does it take to run this query? create table sent_503106 as select distinct clientno from clients where mailno in (select mailno from mails where model='503106' and date between '01SEP2025'd and '30SEP2025'd ) ; How long does it take to run this inner join? create table client_model as select distinct A.clientno,B.model from clients A inner join mails B on a.mailno = b.mailno and B.model in ('503106') and B.date between '01SEP2025'd and '30SEP2025'd ;

Tom · ‎10-02-2025

How is your code structured? Are you using the autocall facility to define your macros? Or do you have the macro definitions and plain SAS code mixed together like in your example? It might make it work better if you define all of the macros first. Then put the plain SAS code that calls them last. Then there would be much less chance that cancelling the submission would interrupt the compilation of the macros.

Tom · ‎10-01-2025

I know this thread only got "reanimated" because of a SPAM message, but this post is a perfect example of why you should (almost) never be using the ancient MISSOVER infile option. @zekeT_sasaholic wrote: First. This process looks strange to me. The reading of input criteria as card statements. I wont judge as maybe you are using that as a mock up. but try this... data layout; infile cards missover; input a $1-500 ; cards; /** your card stuff **/ ; run; hth zeke torres www.wcsug.com Using the MISSOVER option will prevent the input statement from moving to the next line when the current line does not have 500 remaining characters. But it will also discard any line that has less than 500 characters (actually less than 481 since in-line data lines are always padded with spaces to the next multiple of 80 bytes). Instead use the newer TRUNCOVER option (I think it is only 30 years old).

Tom · ‎10-01-2025

Note that the %INCLUDE command also supports the IgnoreDosEof option. IGNOREDOSEOF is used in the context of I/O operations on variable record format files. When this option is specified, any occurrence of ^Z is interpreted as character data and not as an end-of-file marker.

Tom · ‎10-01-2025

Did you review the full logs of the entire SAS session to find the FIRST error? Did you try setting the SOURCE2 option (or add the /SOURCE2 option to the %INCLUDE statement) so that you can see the statements that are being executed? Another thing to double check is the lengths of the lines in those %INCLUDE files. I have seen situations where long lines are truncated. Another thing to double check is if the files have physical tab characters in them. When you submit SAS code from the Program Editor window in SAS Display Manager the tabs are automatically converted into spaces for you. But when you use %INCLUDE to read the files that does not happen. So the program might execute differently.

Tom · ‎10-01-2025

Note that you can do the same thing without the data step. %let mth_s = %sysfunc(intnx(month,%sysfunc(today()),-1,B),date9.); %let mth_e = %sysfunc(intnx(month,%sysfunc(today()),-1,E),date9.);

Tom · ‎10-01-2025

@sc5 wrote: Thanks for replying! Every script has that piece of code in it. So the main script is like %include "mydir/report_1.sas"; %include "mydir/report_2.sas"; %include "mydir/report_3.sas"; and each of those scripts has the code snippet in them. Is "mydir" a fully qualified path? For example if SAS is running on Unix does it start with / ? Does the path mentioned exist on the machine where SAS is running? Does the user that is being used to run the job have permission to read the SAS programs?

Tom · ‎10-01-2025

You have a typo (or perhaps I did?). Notice that SAS tells you about it: NOTE: Variable nobs is uninitialized. Fix this line: if 0 then set fixed4interleave nobs=nob;

Online Status	Offline
Date Last Visited	Friday