About ArtC

ArtC · ‎10-07-2012

As Art suggested the first step is to convert all dates to SAS dates. You do not need to do so here, but it is almost always a best practice. Once dates are converted they can be used to merge or join the tables. The formats currently assigned have nothing to do with the problem it is the stored values that you need to deal with. Conside the following simple example of conversion of both numeric and character dates. data char; input cdate $6.; date = mdy(input(substr(cdate,5,2),2.),01,input(substr(cdate,1,4),4.)); datalines; 199101 199202 run; data num; input ndate 6.; date = mdy(mod(ndate,100),01,int(ndate/100)); datalines; 199202 199303 run; data want; merge char num; by date; run; proc print data=want; run; If you want the SAS date to appear in the YYYYMM format use the YYMMN6. format.

ArtC · ‎10-07-2012

If you need macro variables, I would suggest not using the DATA step. %* Macro window; %WINDOW INPUTS #3 @35 "Inputs" COLOR = BLUE #6 @10 "Percents" COLOR = RED // @15 "Red" COLOR = BLACK +2 RED_1 4 COLOR = GREEN REQUIRED = NO ATTR = UNDERLINE / @15 "Black" COLOR = BLACK +2 BLACK_1 4 COLOR = GREEN REQUIRED = NO ATTR = UNDERLINE / @15 "Blue" COLOR = BLACK +2 BLUE_1 4 COLOR = GREEN REQUIRED = NO ATTR = UNDERLINE; %DISPLAY INPUTS; %let dec_red = %sysevalf(&red_1/100); %let dec_black = %sysevalf(&black_1/100); %let dec_blue = %sysevalf(&blue_1/100); %put &red_1 &dec_red; %put &black_1 &dec_black; %put &blue_1 &dec_blue; OTOH if you are going to use these values in a DATA step then use the WINDOW statement and avoid the use of the macro language altogether. *************************** * DATA step window; data colors; WINDOW INPUTS #3 @35 "Inputs" COLOR = BLUE #6 @10 "Percents" COLOR = RED // @15 "Red" COLOR = BLACK +2 RED_1 4. COLOR = GREEN REQUIRED = NO ATTR = UNDERLINE / @15 "Black" COLOR = BLACK +2 BLACK_1 4. COLOR = GREEN REQUIRED = NO ATTR = UNDERLINE / @15 "Blue" COLOR = BLACK +2 BLUE_1 4. COLOR = GREEN REQUIRED = NO ATTR = UNDERLINE; DISPLAY INPUTS; dec_red = red_1/100; dec_black = black_1/100; dec_blue = blue_1/100; put red_1 dec_red; put black_1 dec_black; put blue_1 dec_blue; stop; run;

ArtC · ‎10-04-2012

Am I missing something or is it as simple as adding a wildcard into the FILENAME statement: filename intxt "c:\temp\xx*.txt"; data want; infile intxt; input var $; run;

ArtC · ‎08-24-2012

It sounds like you want to embed the formula into the worksheet so that it will be active. This can be accomplished using the Output Delivery System, ODS. There have been a number of papers written on the topic. Try google: sas ods excel formula

ArtC · ‎08-21-2012

Without the outer TRIM function, the second set of characters are appended to the end of the 10,000 characters of the first (most of which are blank). The TRIM removes the trailing blanks and allows you to append the values of the next record to immediately follow the '/n' of the preceding record.

ArtC · ‎08-20-2012

This error indicates to me that the file list.txt is not in the current directory (root of H:). From the earlier posts it is still unclear to me that the OP knows the actual location of the file.

ArtC · ‎08-20-2012

Your e-mail settings may prevent you from receiving HTML. As an alternative you could have the file as an attachment. Here it is PDF - just to be different. Also I have left the HELLO as part of the email. Leaving it as ODS TEXT as FriedEgg has it will include it in the attached file. A final DATA _NULL_ step is then used to generate the e-mail. filename eml email to=sam@caloxy.com subject="From SAS" content_type='text/html' attach="c:\temp\results.pdf"; ods pdf file="c:\temp\results.pdf" style=minimal; data info; length infoname infoval $60; drop rc fid infonum i close; rc=filename("abc","c:\temp\testinc.sas"); fid=fopen("abc"); infonum=foptnum(fid); do i=1 to infonum; infoname=foptname(fid,i); infoval=finfo(fid,infoname); output; end; close=fclose(fid); run; ods text=' '; option nocenter nodate; proc print data=info noobs; title1 "Attributes obtained for a file"; run; ods pdf close; data _null_; file eml; put 'Hello,'; put 'The weekly code has completed.'; run;

ArtC · ‎08-16-2012

The following macro, which I found in Appendix 1 a macro book**, does the conversion that you are looking for. It has the advantage of maintaining variable names and labels and uses any numeric formats (such as dates) in the conversion. **********************; ***A1.6.4 ***%ALLCHAR **********************; %macro allchar(dsn); * Determine the numeric vars in &dsn; proc contents data=&dsn out=cont noprint; run; * Create the macro variables for each numeric var; data _null_;; set cont(keep=name type format formatl formatd label); length fmt $15; where type=1; * Count the numeric vars and save the total number; i+1; ii=left(put(i,3.)); call symput('n',ii); * create a format string; fmt = 'best.'; if format ne ' ' then fmt = trim(format) ||trim(left(put(formatl,3.))) ||'.'||left(put(formatd,3.)); fmt = compress(fmt); call symput('fmt'||ii,fmt); * Save the variable name; call symput('name'||ii,name); * Save the label for this variable; if label = ' ' then label = name; call symput('label'||ii,label); run; * Establish a data set with only character variables; * &n number of numeric variables in &dsn; * __aa&i temporary var to hold numeric values; * &&name&i name of the variable to covert from numeric; * * The numeric value of &name1 is stored in __aa1 * by renaming the variable in the SET statement. __aa1 * is then converted to character and stored in the * 'new' variable &name1 in the data set CHARONLY. * ; data charonly (drop= %* Drop the temp. vars used to hold numeric values; %do i=1 %to &n; __aa&i %end; ); length %* Establish the vars as character; %do i=1 %to &n; &&name&i %end; $8; set &dsn (rename=( %* Rename the incoming numeric var to a temp name; %* This allows the reuse of the variables name; %do i=1 %to &n; &&name&i=__aa&i %end; )); * Convert the numeric values to character; %do i=1 %to &n; &&name&i = left(put(__aa&i,&&fmt&i)); label &&name&i = "&&label&i"; %end; run; proc contents data=charonly; proc print data=charonly; run; %mend allchar; %allchar(sasclass.biomass) ** Appendix 1 of Carpenter's Complete Guide to the SAS Macro Language, 2nd Edition. Used with the permission of the author.

ArtC · ‎08-16-2012

In this solution the TRANSLATE function is changing the special characters to blanks, and then the COMPBL function converts multiple blanks to a single blank. There are other functions that could have also been used COMPRESS and DEQUOTE among them. Also this soution uses single quotes to delineate the text strings in the TRANSLATE function. When using macro language elements this technique can have unintended consequences. Macro language quoting functions such as %STR and %BQUOTE, shown below, do not have these issues. %let xx2 =%sysfunc(compbl(%sysfunc(translate(&x,%str( ),%bquote(" ; ,))))); I once had got a number of consulting hours to debug a program, ultimately the problem was found in the following code fragment %scan(&name, &i, ' ') which works fine unless the &name contains an apostrophe such as O'Conner

ArtC · ‎08-14-2012

For more information on the UPDATE approach (with some alternate examples) shown by @Astounding take a look at: https://communities.sas.com/thread/32502 and http://www.sascommunity.org/wiki/Tips:Using_UPDATE_to_Collapse_a_Data_Set

ArtC · ‎08-14-2012

If it is assumed that the problem is as specified then the following will collapse the data. data have; infile datalines missover; input id x y z; datalines; 123 8 123 . 7 456 . 9 456 . . 2 run; * for this specific example; proc summary data=have nway; class id; var x y z; output out=want(drop=_type_ _freq_) sum=; run; proc print data=want; run; This solution will not work if there are other conditions such as more than one row with a non-missing value.

ArtC · ‎08-14-2012

We should also ask whether the subset is selected with or without replacement. For small subsets of a large data set it probably does not matter a lot, but....

ArtC · ‎08-14-2012

The success or failure (of most PROC or DATA steps) is stored in the automatic macro variable &SYSERR. This macro variable can be interrogated to determine further processing. Assuming that you are running the GENMOD within a macro loop, checking &SYSERR should be straightforward.

ArtC · ‎08-14-2012

If these are just key values can't they be treated as character strings? If you truly need them to be numeric, you will need to tell us what it is that you are trying to accomplish with them.

ArtC · ‎08-14-2012

Mike, I would like to say that I did not know of your article on sasCommunity.org, but I know I did because I signed off on its associated sasCommunity.org tip http://www.sascommunity.org/wiki/Tips:Create_an_Informat_from_a_User-Defined_Function back in January. Hey that was 8 months ago...sigh The solution in your sasCommunity article addresses a broader problem and is a bit more sophisticated than mine. I suggest readers take a look at Mikes article for a better FCMP example. Again the link to his article is http://www.sascommunity.org/wiki/Create_an_Informat_from_a_User-Defined_Function. Folks, if you are new to PROC FCMP take a look at Mike's article and some of the others on sasCommunity.org. The idea of using a function within a format [starting in SAS 9.3] is extremely powerful as it makes most DATA step functions (and user defined functions) available in procedure steps and wherever formats are used.

Online Status	Offline
Date Last Visited	‎06-20-2021 07:11 PM