Rolland,

A still sleazier trick would be to write

by &by ;

SAS will accept empty BY statements. Some older verisons did not and perhaps some procs still do not. So testing may be advised. Questions: Should SAS be able to change this sort of behavior at whim from version to version? Should sleazy programmers take advantage of this behavior? When do we get a stable documented language? Perhaps someone could do a short paper on the advantages of having a stable language.

717 data w ; do x = 1 to 5 ; output ; end ; 718 run ;

NOTE: The data set WORK.W has 5 observations and 1 variables. NOTE: DATA statement used: real time 0.04 seconds

725 data w ; do x = 1 to 5 ; output ; end ; 726 run ;

NOTE: The data set WORK.W has 5 observations and 1 variables. NOTE: DATA statement used: real time 0.05 seconds

727 728 data _null_ ; 729 set w ; 730 by ; 731 put _all_ ; 732 run ;

x=1 _ERROR_=0 _N_=1 x=2 _ERROR_=0 _N_=2 x=3 _ERROR_=0 _N_=3 x=4 _ERROR_=0 _N_=4 x=5 _ERROR_=0 _N_=5 NOTE: There were 5 observations read from the data set WORK.W. NOTE: DATA statement used: real time 0.00 seconds

IanWhitlock@westat.com

-----Original Message----- From: Roland [mailto:roland@RASHLEIGH-BERRY.FSNET.CO.UK] Sent: Wednesday, December 11, 2002 7:39 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Re: recruiting cheesy, sleasy SAS tricks

Here is my "byline trick". Suppose you have a by= parameter in your macro so that the user can specify the "by" variables. If this is set then you will probably need to add the line "by &by ;" underneath some of the procedure calls. This is best handled by setting up a macro variable like in the following:

%local byline; %if %length(&by) %then %let byline=by &by %str(;);

and then you can add this line under your procedure calls as follows:

proc summary nway data=ds; &byline

In this way you either get nothing if the by= parameter was not set or you get the correct "by" line.

"Roland" <roland@rashleigh-berry.fsnet.co.uk> wrote in message news:at7bfc$o6q$1@news7.svr.pol.co.uk... > Another one I call the "global declaration trick". If you want to work out > if a global macro variable has been set or not then declaring it as global, > if it already exists, will not affect its contents. So you can declare it > global and then test its length to see if it has been set. > > And another one, a dummy macro declaration. This is a very cheesy and sleazy > trick. It is for when you want to run a macro once but never again during a > session. Sure, you are going to call it maybe dozens of times, but you do > not want to run it if it has already been run. You could set up a global > macro variable and test its setting and branch out of the macro, but that > wouldn't be either cheesy or sleazy. So what you do instead (and I think > this only works for members of an autocall library) is to put all the macro > code before the macro declaration. and just end the macro with: > > %macro mymacro; > %mend mymacro; > > So all the macro code gets run and the stored macro contains nothing. That > way you call it as many times as you like afterwards and it will do nothing. > > > "Roland" <roland@rashleigh-berry.fsnet.co.uk> wrote in message > news:at7b0g$nsj$1@news7.svr.pol.co.uk... > > Another one I have seen a lot and used to use a lot but not recently is a > > way of disabling a chunk of macro code within a larger macro. That is to > use > > something like: > > > > %macro dontdo; > > .... > > ... > > .. > > %mend dontdo; > > > > This is better that using comment lines since these may already exists > > within the code you are trying to disable and so might terminate your > > commentary section earlier than you thought. > > > > > > > > "Roland" <roland@rashleigh-berry.fsnet.co.uk> wrote in message > > news:at7aqm$i0s$1@newsg3.svr.pol.co.uk... > > > Another sleazy trick I use a lot is when I am assigning a value to a > macro > > > variable in proc sql. I use "separated by ' '" after it, not because I > > have > > > more than one item, but just to cut off the trailing blanks. > > > > > > "Roland" <roland@rashleigh-berry.fsnet.co.uk> wrote in message > > > news:at719h$do0$1@newsg3.svr.pol.co.uk... > > > > Another trick I use for when I want the results from some system > > commands > > > is > > > > write the results out to a file and then read in that file and assign > it > > > to > > > > a macro variables using my %readfile macro. If changes the line throws > > to > > > > singla spaces and it all comes out perfect and ready for further > > > processing. > > > > > > > > http://mysite.freeserve.com/rolandrb/sasmacros/readfile.sas > > > > > > > > > > > > Another sleazy trick is to ensure something is in quotes. To do this I > > > > compress for both single and double quaotes and enclose the whole lot > in > > > > double quotes. > > > > > > > > %let file="%sysfunc(compress(&file,%str(%'%")))"; > > > > > > > > > > > > Another one which is actually quite sensible is for defaulting an > output > > > > dataset name. I scan for the first string up to a left parenthesis > > whether > > > > the parenthesis exists or not. That way I don't get the modifiers that > > > might > > > > have been defined with the input dataset. > > > > > > > > %let dsout=%scan(&dsin,1,%str(%()); > > > > > > > > > > > > "Roland" <roland@rashleigh-berry.fsnet.co.uk> wrote in message > > > > news:at70qi$lvh$1@news6.svr.pol.co.uk... > > > > > Another one I use is "sum()=" in proc summary literally as written. > > This > > > > > calculates the sum of all the numeric variables and puts the result > in > > > > > variables of the same name. I'm not sure if this is common practise. > > > > > > > > > > Another is to use the 0 or 1 for a boolean expression and part of a > > > > > calculation as in the following where I have subtracted 1 from a > year > > > > count > > > > > but will add it on again if a condition is true.: > > > > > > > > > > %macro age(dob,date); > > > > > year(&date)-year(&dob)-1+((month(&date)>month(&dob)) > > > > > or ((month(&date)=month(&dob)) and (day(&date)>=day(&dob)))) > > > > > %mend; > > > > > > > > > > > > > > > > > > > > "Roland" <roland@rashleigh-berry.fsnet.co.uk> wrote in message > > > > > news:at6vs8$hss$1@newsg4.svr.pol.co.uk... > > > > > > put (_all_) (=); > > > > > > > > > > > > ..is a good one to get sas to drop the _N_ and _ERROR_ automatic > > > > variables > > > > > > from the put _all_ list. > > > > > > > > > > > > Another good one is to trick sas into "flattening" a "proc report" > > > which > > > > > is > > > > > > to use a dummy "noprint" variable after an across variable. > > > > > > > > > > > > > > > > > > > > > > > > "Michael L. Davis" <michael@BASSETTCONSULTING.COM> wrote in > message > > > > > > > > > > > > > > > > > > > > > news:20021210185607.GDAN17506.imta01a2.registeredsite.com@smtp.registeredsit > > > > > > e.com... > > > > > > > Hello Friends, > > > > > > > > > > > > > > I'd like to do a Coder's Corner presentation at NESUG next year > on > > > the > > > > > > subject of "Cheesy, Sleasy SAS Tricks". As an example, when > coding > > > SCL > > > > > > programs, I'm in the habit of including the statement: > > > > > > > > > > > > > > rc=rc ; > > > > > > > > > > > > > > to suppress the compiler warning that I'm creating the variable > rc > > > but > > > > > not > > > > > > using it. > > > > > > > > > > > > > > I'd welcome additional candidates for the role of "cheesy, > sleasy > > > > > tricks". > > > > > > To qualify, they need to be coding declarations or other > > constructions > > > > > that > > > > > > suppress the literal mindedness of SAS or its sometimes > > > over-solicitous > > > > > > manner of trying to protect us from ourselves. > > > > > > > > > > > > > > I'll be sure to acknowledge useful contributions, selected or > > > > otherwise, > > > > > > in the acknowledgement section of the presentation. > > > > > > > > > > > > > > BTW, thanks to all for the help last week with regards to > > reordering > > > > the > > > > > > PDV. Unfortunately, the friend for whom I was trolling wanted as > > easy > > > a > > > > > > method as the RETAIN statement. The ATTRIB statement was to > change > > > the > > > > > > length of a variable brought in by a SET statement. So combining > > the > > > > > RETAIN > > > > > > and ATTRIB statements did not work (the length was not changed by > > the > > > > > ATTRIB > > > > > > statement). > > > > > > > > > > > > > > - Michael "Mad Doggy" Davis > > > > > > > > > > > > > > Michael L. Davis > > > > > > > Bassett Consulting Services, Inc. > > > > > > > North Haven CT 06473-3712 > > > > > > > web: http://www.bassettconsulting.com > > > > > > > email: michael@bassettconsulting.com > > > > > > > tel: 203-562-0640 > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >