%let searchStr=BorrowerName; %let buffLen=%sysfunc(length(&searchStr.)); data _null_;   length buff $ &buffLen.;   retain buff '';   retain inBuff 0;   infile "c:\temp\file.txt" RECFM=N eof=eof;   file "c:\temp\outfile.txt" RECFM=N;   input char $char1. @@;   buff=substr(buff,2,&buffLen.-1)||char;   inBuff=min(inBuff+1,&buffLen.);   if buff="&searchStr." then do;   /*do not put*/   buff='';   inBuff=0;   end;   if inBuff=>&buffLen. then do;   outChar=char(buff ,1);   put outChar $1. @@;   end;   return; eof:   if inBuff>0 then do;   buff=substr(buff,&buffLen.-inBuff+1,inBuff );   put buff $varying&buffLen..inBuff @@;   end;   stop; run; This code is inefficient, but if your xml files are small, it works. Reading bigger blocks and doing search on those (but solving the issue on the block boundaries) would be better. Some comments: 1. end= option is not working in this setting on the infile statement, so I had to use the eof= option. 2. I had to use the $char1. informat, because $1. informat can skip blanks. 3. I needed a variable: inBuff, which stores, how many characters are actually in buff. I couldn't solve this without that variable, again because of the blanks. 4. $varying. format is needed at the end, if you don't want additional blans in the output file (compared to the input file). Thank you sharing this problem, I wonder if anyone could suggest impovements (make it more effective or simplify it) to the above code. ... View more

I would use regexp, if I have to remove more then one thing data have;   val="papaya,mango,mango,orange,papaya,fruit,veg,apple,kiwi,orange,orange,papaya,papaya";   output; run; data want (keep=processed);   array removes{4} $20. ("mango","fruit","veg","orange");   length processed $2000;   length tmp       $200 ; /*constructing the regexp*/   if _n_=1 then do;   do i=1 to dim(removes);   tmp=catx('|',tmp,'(,?'||strip(removes)||')');   end; putlog tmp=;   regexp=cats('s/',tmp,'//'); putlog regexp=;   regexpid=prxparse(regexp);   end;   set have;   processed=prxchange(regexpid,-1,val);   processed=prxchange('s/(^,)//',-1,processed);/*removing ',' from beginning*/ putlog processed=; run; ... View more

Just add a comma (,) to the end of mango. data test;   have="mango,orange,papaya";   want = tranwrd(have,"mango,","");   putlog want=; run; But if you want to handle situations, where mango can be at the beginning, end, and in the middle of the string... Maybe you need to use tranwrd 2 times. Once with "mango," then with ",mango". ... View more

Small improovement by using array, view and by group processing: options fullstimer msglevel=i; %let datetime_start = %sysfunc(TIME()); %put START TIME: %sysfunc(datetime(),datetime14.); data original (drop = i);   do i = 1 to 1000000;     lon = -78 + ranuni(i);     lat = 37 + ranuni(i+1);     output;   end; run; %let numrep=10; %let sampsize=5500; proc surveyselect data = original method = SRS rep = &numrep.   sampsize = &sampsize. out = sampledata noprint;   id _all_; run; options vbufsize=500M; data geodist (keep=Replicate geodist) / view=geodist;   array alat[&sampsize.] _temporary_;   array alon[&sampsize.] _temporary_;   do i=1 to &sampsize.;   set sampledata;   alat=lat;alon=lon;   end;       do i=1 to &sampsize.;     lat=alat;lon=alon;     do j=i+1 to &sampsize.;          geodist=geodist(lat, lon, alat , alon , "m");         output;   end;       end;     run; proc kde data=geodist(spill=no);       univar geodist/method=srot out=kout(keep=density) plots=NONE NOPRINT;    by Replicate; run;   %put END TIME: %sysfunc(datetime(),datetime14.); %put PROCESSING TIME:  %sysfunc(putn(%sysevalf(%sysfunc(TIME())-&datetime_start.),mmss.)) (mm:ss); ... View more