As stated before there is the datetime value in SAS. See Log from SAS program 39 %put NOTE: &=sysvlong &=sysscpl %sysfunc(datetime(), 20.6); NOTE: SYSVLONG=9.04.01M2P072314 SYSSCPL=X64_7PRO 1727371518.875000 As you see running on windows you get down to milliseconds but not more, might be different on Unix/Linux, can not test now Bruno Message was edited by: Bruno Müller result from running in the SAS University Edition 41 %put NOTE: &=sysvlong &=sysscpl %sysfunc(datetime(), 20.8); NOTE: SYSVLONG=9.04.01M1P120413 SYSSCPL=Linux 1727351691.72906000
... View more
First Question The outline is controlled by the "contrast" color, see code example below that shows two techniques using a style and using the new STYLEATTRS statement. The gdata1 and gcdata1 define the color and contrast color This works with SAS9.4M2 There might be better ways of doing this. ods path (prepend) work.template (update) ; proc template; define style styles.mystyle; parent=styles.listing; style GraphColors from GraphColors / 'gdata1'=yellow 'gcdata1'=black 'gdata2'=green 'gcdata2'=green 'gdata3'=blue 'gcdata3'=blue ; end; run; ods graphics / reset=index; ods html path="c:\temp" file="sample.html" style=styles.mystyle; ods rtf path="c:\temp" file="sample.rtf" style=styles.mystyle; proc sgplot data=sashelp.cars ; hbar type / group=type datalabel; run; proc sgplot data=sashelp.cars ; styleattrs datacolors=(yellow red green) datacontrastcolors=(black red green) ; hbar type / group=type datalabel; run; ods html close; ods rtf close;
... View more
See also this example proc groovy; submit "p1" "p2"; public class ArgsTest { public static void main(String[] args) { int index; for (index = 0; index < args.length; ++index) { System.out.println("args[" + index + "]: " + args[index]); } } } endsubmit; quit;
... View more
Well not exactly what you are looking for, but it might help you towards what you want to achieve. The below code passes two arguments to a groovy program that uses Java classes to read out width and height of an image. Find more documentation on Proc Groovy in Base SAS(R) 9.4 Procedures Guide, Third Edition %let imgName = c:\temp\f.JPG; proc groovy; submit "&imgName" "name=value"; import java.awt.image.BufferedImage import java.io.File import javax.imageio.ImageIO args.each { println "NOTE: ----> arg ${it}" } def img = ImageIO.read(new File(args[0])) println("NOTE: Width:" + img.getWidth() + " Height:" + img.getHeight()) exports.put("xWidth", img.getWidth()) exports.put("xHeight", img.getHeight()) endsubmit; quit; %put _user_;
... View more
You can take advantage that with SAS9.4 we can create formats, that use functions to return the value. here is an example: * * create a function to round *; proc fcmp outlib=work.func.sample; function xround100(value); return( round(value, 100) ); endsub; run; * * make function available *; options cmplib=work.func; * * create a format that uses the function *; proc format; value xround100x low - high = [xround100()] ; run; * * apply the format *; proc tabulate data=sashelp.cars; class origin type; var invoice; table type, origin*invoice*f=xround100x.; table type, origin*invoice; run;
... View more
During compilation of the DATA step the number of observations is read from the input data set and stored in the variable _nObs. During execution for each iteration of the DATA step loop _n_ is increased by one. The IF without a THEN is called a Subsetting IF.
... View more
Find below a DATA Step example. The first steps creates data for validation, the second steps has the actual row selection using a subsetting IF data have; rowId + 1; set sashelp.Class; run; data want; set have nobs=_nObs; if _n_ = 1 or _n_ >= (_nObs - 4); run;
... View more
I would go for a regular expression, see below: The \d idenifies digits, the + matches one or more times, the $ matches the position at the end of the input string It is important to trim the string first data have; infile cards truncover; input error $200.; cards; 1 LLLL MMMMM NNNN Missing0048165a 2 LLLL MMMMM NNNN Missing 3 AAAAAA BBBBB CCCCCCCC segment Required0048158 4 AAAAAA BBBBB CCCCCCCC segment Required 5 XXXX YYYYYY ZZZZZZ (Required) Missing0000047 6 XXXX YYYYYY ZZZZZZ (Required) Missing ; data want; set have; digitStart = prxmatch("/\d+$/", trim(error)); part1 = substr(error, 1, digitStart-1); part2 = substr(error, digitStart); run;
... View more
You can take advantage of the NLEVELS option in the Proc FREQ procedure. See sample code below: * create some test data ; data newClass; set sashelp.class; if ranuni(0) < 0.2 then do; call missing(name); end; if ranuni(0) < 0.5 then do; call missing(age); end; height = .; empty = .; run; * count values for each variable ; ods output NLevels=nlevels; proc freq data=newClass nlevels; tables _all_ / noprint; run; * check for empty variables ; data colInfo; set nlevels; vComment = (nLevels = 1 and nMissLevels = 1 and nNonMissLevels = 0); run; * format to print ; proc format; value nullComment 1 = "null" 0 = "not null" ; run; proc print data=colInfo; format vComment nullComment.; run; Now you need to do this for every table in your library, a good starting point is the blog entry Implement BY processing for your entire SAS program - The SAS Dummy by To get a list of tables in a library you can use the dictionary tables like this: proc sql; create table myTables as select catx(".", libname, memname) as tableName length=41 from dictionary.tables where libname = 'WORK' ; quit;
... View more
