Removing spaces occured due to group in proc report

[JasonNC]

Hi,

In proc report i want to display as shown before

Name       Address            plans

john         6325 holly dr      XXX

               raliegh,NC         YYY

               27540               ZZZ

In my data set i have the address field as three variables and the plans in one variable. 

Name      address 1           address 2         address 3      plans  

john         6325 holly dr      raliegh,nc          27540           xxx

john         6325 holly dr      raliegh,nc          27540           YYY

john         6325 holly dr      raliegh,nc          27540           ZZZ

i am doing it as

PROC REPORT DATA=test NOWD MISSING  SPLIT='*' 

columns Name address 1  address 2   address 3  plans

define name / group

define address / computed

define plans / display

compute address / char length =105

address=address 1||"~{newline}"||address 2 ||"~{newline}"||address 3;

endcomp;

run;         

The problem i am having is it is displaying as shown below

Name         address               plans

john           6325 holly dr         XXX

                 raliegh,NC           

                 27540                  

                                            YYY

                                             ZZZ

I want to remove the extra space between Plans.Can you please help me regarding that            

[Cynthia_sas]

Hi:

  I have a couple of observations and some questions.

  Are the variables actually named "address 1", "address 2" and "address 3" with spaces in the variable name, or are your showing the variable label?? Given something similar to your COLUMN statement, I would expect your LOG to show an ERROR message or two:

279 column name address 1 address 2 address 3 address PLANS;

  I don't see your DEFINE statements for  "ADDRESS 1", "ADDRESS 2" and "ADDRESS 3" -- I do see your DEFINE statement for the computed item ADDRESS. But I don't see how your code is working. I would expect to see ADDRESS1, ADDRESS2, ADDRESS3 as DISPLAY items with NOPRINT specified in the DEFINE statement. Can you show ALL your code, including the ODS statements? And/or, can you post your complete SAS Log?

  And, how is this data file being created so that PLANS is a different value on every row, but address values 1, 2, and 3 (however they are named) are all the same on every row? It might actually be easier to pre-process the file with a DATA step program rather than expecting PROC REPORT to deal with the input dataset that you show.

  I don't understand how you are getting the results you posted, especially since you didn't post your ODS statements -- what is your destination of interest? {NEWLINE} won't work in the LISTING destination -- so your output must be to some other destination. Your code should not have produced any output with spaces in the variable names and you didn't show the use of "Address 1"n construct.

  Also,  I don't understand how you are getting output the way you describe with NAME as a GROUP item, but another item (such as PLANS or ADDRESS1, ADDRESS2, ADDRESS3) as DISPLAY item(s). I would expect to see something like this in the LOG:

NOTE: Groups are not created because the usage of plans is DISPLAY. To avoid this note, change

all GROUP variables to ORDER variables.

  So...given my incomplete understanding of your data and your actual code, my idea would be to pre-process your DATA with a DATA step program using BY group processing and FIRST.byvar/LAST.byvar logic, so you can easily concatenate the various values for PLANS and then create ADDRESS when you are on the last observation for NAME. Then, after the data is "cleaned up" you can pass the data to PROC REPORT.

  You can also open a track with Tech Support, they could look at your actual data and your actual code and determine your destination of interest and help you come to the best solution.

cynthia

                        -

                        22

                        76

ERROR 22-322: Syntax error, expecting one of the following: a name, ;, (, ',', -, :, =, _ALL_,

_CHARACTER_, _CHAR_, _NUMERIC_.

ERROR 76-322: Syntax error, statement will be ignored.

[JasonNC]

Thanks Cynthia for your reply.It's my bad not to post the whole code.i appreciate your patience and support.

[Ksharp]

You need to pre- process the dataset firstly.

data temp;
infile datalines dlm=' ';
input (Name      address1           address2         address3      plans) ( & $20.);
datalines; 
john         6325 holly dr      raliegh,nc          27540           xxx
john         6325 holly dr      raliegh,nc          27540           YYY
john         6325 holly dr      raliegh,nc          27540           ZZZ
Patrick        6325 holly dr      raliegh,nc          27540           xxx
Patrick         6325 holly dr      raliegh,nc          27540           YYY
Patrick         6325 holly dr      raliegh,nc          27540           ZZZ
;
run;
data want(drop=address: plans);
 set temp;
 by name notsorted;
 length add pla $ 200;
 retain pla ;
 add=catx('~n',of address:);
 pla=catx('~n',pla,plans);
 if last.name then do; output; call missing(of _all_);end;
run;
ods listing close;
ods html file='c:\x.html' style=sasweb;
ods escapechar='~';
proc report data=want nowd;
run;
ods html close;
ods listing;

Ksharp

[JasonNC]

Hey,

Thanks ksharp it worked for my present scenario .My output destination is PDF.I appreciate your patience and support.