Have a look at this code COUNTW will count the "words" delimited by a blank, SCAN will get the proper string data postalcodes; postal_codes = "T1S0C3 T1S1A1 T1S1A2 T1S1A4 T1S1A5 T1S1A6 T1S1A7 T1S1A8 T1S1A9 T1S1B1 T1S1B2"; length postalCode $ 16; n_postalCodes = countw(postal_codes, " "); do i = 1 to n_postalCodes; postalCode = scan(postal_codes, i, " "); output; end; drop postal_codes /* n_postalCodes i */; run;
... View more
Fleroo I might have to do with the length, I am not sure, I generally put quotes around userid and password. Please mark the discussion as answered, so others can see immediately that this has been answered Enjoy using SAS
... View more
pcfriendly use the defaults for the data and assign the roles as stated earlier the Ranking method should be "smallest to largest" check the box "Reverse Ranking from smallest to largest" leave the default for "if values tie"
... View more
The ANYDTDTM informat was introduced with SAS9.2 please use the NLDATM informat in SAS9.1 from my code sample change the line Time_date : anydtdtm. to Time_date : nldatm. this should then give you the correct date and time. Please note that the behavior of the NLDATM informat depends on the LOCALE= SAS System option
... View more
SAS Enterprise Guide also provides a task to do ranking under Tasks -> Data -> Rank You will use start_date as "column to rank" and job_role as "rank by" Unter options you check "Reverse Ranking from largest to smallest" so the youngest date gets the ranking 1. You can then use a query to select only those with ranking value 1
... View more
The following code will read your data into a SAS Data Set. I have added code to extract the date and time from time_date variable, just in case you need them to be separate filename mycsv "c:\temp\test.csv"; data test; infile mycsv dlm="," firstobs=2; input ID : 8. Time : $1. Rate : $12. Time_date : anydtdtm. Bid : 8. Ask : 8. ; justDate = datepart(time_date); justTime = timepart(time_date); format time_date datetime19. justDate date9. justTime time8. bid ask 10.5 ; run; filename mycsv clear;
... View more
The ANYDTDTE. informat almost does what you need but you need to tweak the data a bit (add a day), see example below: data have; infile cards truncover; input @1 dateString $32. @; input @1 mydate anydtdte32. @; newDate = tranwrd( strip(dateString), " ", " 1, "); newDate2 = input(newDate, anydtdte32.); format mydate newDate2 date9.; cards; January 2011 February 2011 March 2011 April 2011 May 2011 June 2011 July 2011 August 2011 September 2011 October 2011 November 2011 December 2011 ;
... View more
SAS Data Integration Studio can be extended with your own user written transformations, so you would need to code it once and then just use it again and again. The code for such a transformation could look like this: %macro myCheckData; %global trans_rc; data _A_%sysfunc( compress(&transformID, .)); set &_input(obs=1); run; %put NOTE: &=sysnobs &=syserr; %if &sysnobs = 1 %then %do; %let trans_rc = 0; %put NOTE: OKOKOKOKOK; %end; %else %do; %let trans_rc = 8; %put NOTE: NOTOK NOTOK NOTOK NOTOK; %end; %put NOTE: &sysmacroname &=trans_rc; %mend; %myCheckData You can use then the normal Return Code Check transformation.
... View more
You can add a Job either after the Loop End transformation or within the loop, see also picture below. If the job does not have any parameter it will do its work for each iteration of the loop
... View more
SAS Data Integration Studio provides a loop transformation that can help with this. Basic Steps: create a job for the "detail" work, add parameters to the job (using job properties) the parameter values will be available as Macro variables so they can be used anywhere create the control job with the loop transformation that will provide the parameter values to the "detail" job
... View more
Please note, that the DELETE function as outline above, can only be used within a SCL (SAS Component Language) program. To delete files or directories from within the DATA Step or a Macro program you need to use the FDELETE function. Bruno
... View more
Hi Wendy You are right you have to name all entries you want to delete. But you do not have to do it manually, with the help of some code this is done automatically, see below for an example ods output Properties=_properties; proc document; doc name=mydoc; list / levels=all details follow order=insert ; run; quit; proc sql number; select path into :pathList separated by ", " from _properties where type = "Dir" and countw(path, "\") = 1 ; quit; %put NOTE: &=pathList; proc document; doc name=mydoc; title "before delete"; list / levels=all details follow; run; delete &pathList; run; list / levels=all details follow; run; quit;
... View more
