The ITERATOR object allows you to control the ordered retrieval of information from the hash table. Loading the table is much more straightforward. From your question it sounds like you will want to load the table one record at a time. The beauty of the hash object is that hash table is independent of the incoming order. There are some references to information at: http://www.sascommunity.org/wiki/Hash_object_resources ... View more

You have placed a semicolon after the macro call. This semicolon is eventually added to the second END, which is then END;; This null statement interrupts the IF-THEN/ELSE sequence thus isolating the ELSE. The solution is to call the macro without the semicolon. ... View more

You are very close. Just a couple of things: 1) Double quotes must be used to surround macro variables that appear inside of quoted strings. 2) The &PARM1 is part of the path and as such must be quoted. The ODS statement becomes: [pre] ods rtf file= "&parm1\Boxplot.rtf"; [/pre] Art ... View more

Since you cannot use DATALINES, you could still process against _INFILE_ after it has been brought into the PDV. The nested functions become a bit awkward but should not be too bad. [pre] * TRYCAP.TXT contains the data line from above.; filename txtline "c:\temp\trycap.txt"; data a; infile txtline ; input ; aftera = scan(substr(_infile_,index(upcase(_infile_),'A')+1),1); put _infile_; put aftera =; run; proc print data=a; run; [/pre] This has the advantage of a case insensitive search, but the original case is preserved in the final data set. ... View more

I hope this is not a step backwards, but you mention the CAPS option. Does this not do what you want? [pre] options caps; data a; input @'A' aftera $; datalines; d s e r a qwert qawetrqq A222qwerrt run; proc print data=a; run; [/pre] ... View more

For data sets of your size (assuming a reasonable number of columns) the SQL step ought to do the trick. If the SQL look-up is too slow, there are other alternatives that generally require more coding, but can usually process quicker. The DATA Step Components (HASH) Objects are about the fastest. An overview of look-up techniques can be found at: http://caloxy.com/papers/43-i_how_table_lookups_from_ift.pdf ... View more

This is my test of being able to write to the EB forum. ... View more

In SAS9 the %ABORT statement has similar capabilities to the DATA step's ABORT statement. It is executable so it can be used with an %IF. Different combinations of %ABORT and its options perform differently depending on the OS and the operating environment. Experiment. ... View more

When you specify the width on the format, be sure to specify it when the variable is created. In your case the FORMAT statement would need to come before the SET statement. The LENGTH statement would also do the trick. [pre]data Fruit_frmt; format Fruit $f_name50.; *length fruit $50; set sorted_Fruit; run; [/pre] ... View more

In your first post make sure that the WHERE construction gives you the result that you are expecting. It could be perfectly fine (just what you want), but many users get in trouble with this non-standard syntax. [pre]where y_&i=nvalue=1;[/pre] the WHERE is true when: y_&i=1 and nvalue=1 y_&i=0 and nvalue^=1 ... View more

Double check to make sure that the length of the datetime variable has not been shortened. ... View more

I really like Data_NULL_'s SUMMARY solution. It has the added advantage of allowing the top n values through the IDGROUP. Since the CLASS statement is being used the result is sorted on CLAIMID, but the incoming data need not be sorted. ... View more

Try something like: [pre] proc sql; select claimid, date, rownumber from claims group by claimid having date=max(date) ; quit; [/pre] ... View more

Assuming that your table is not too very large, one way: [pre] data claims; input claimid date mmddyy10. rownumber; format date date9.; datalines; 111 01/31/2010 1 111 02/28/2010 2 111 03/31/2010 3 112 1/31/2010 2 112 3/31/2010 1 ; run; proc sort data=claims; by claimid descending date; run; data latestclaims; set claims; by claimid; if first.claimid; run; proc print data=latestclaims; run; [/pre] ... View more

Since I have never actually needed these values it never bothered me sufficiently to ask. Certainly the information is available, we see it when we use the CNTLOUT. The MIN= and MAX= options (on the VALUE and INVALUE statements) have some effect on the stored min/max values (as shown by the CNTLOUT), but not on those seen in the VIEWs. ... View more