| View previous topic :: View next topic |
| Author |
Message |
vijay
Beginner
Joined: 09 May 2003
Posts: 131
Topics: 64
|
 Posted: Mon Jul 17, 2006 10:09 am Post subject: File Match/Merge in SAS |
 |
|
Hi,
When I try to run this sample code I get an error.ID1 has been defined as char and numeric and same error with id2.I'm new to SAS.Help me in figuring out this error
Vijay
| Code: |
DATA FILEIN1;
INFILE CARDS;
INPUT @01 ID1 $CHAR2.;
CARDS;
12
11
;
DATA FILEIN2;
INFILE CARDS;
INPUT @01 ID2 $CHAR2.
@04 NAME $CHAR10.;
CARDS;
12 NAME10
11 NAME11
;
PROC SORT DATA=FILEIN1;
BY ID1;
PROC SORT DATA=FILEIN2;
BY ID2;
DATA COMBINE;
LINK GETFILEIN1;
LINK GETFILEIN2;
DO WHILE (1=1);
IF (ID1=ID2) THEN
DO;
OUTPUT;
END;
IF FILE1EOF AND FILE2EOF THEN STOP;
END;
GETFILEIN1:
IF EOF1 THEN
DO;
FILE1EOF=1;
RETURN;
END;
SET FILEIN1 END=EOF1;
RETURN;
GETFILEIN2:
IF EOF2 THEN
DO;
FILE2EOF=1;
RETURN;
END;
SET FILEIN2 END=EOF2;
RETURN;
PROC PRINT DATA=FILEIN1;
TITLE 'PRINT FILEIN1 RECORDS ';
PROC PRINT DATA=FILEIN2;
TITLE 'PRINT FILEIN2 RECORDS ';
PROC PRINT DATA=COMBINE;
TITLE 'PRINT COMBINE RECORDS ';
VAR ID3 NAME3;
/*
|
|
|
| Back to top |
|
 |
jetson
Beginner
Joined: 07 Oct 2005
Posts: 30
Topics: 2
Location: Texas
|
| Posted: Mon Jul 17, 2006 11:35 am Post subject: |
|
|
| It's not clear what you are trying to accomplish. Joining the tables is much simpler than what you are doing...perhaps you want more than to just join the two tables by ID? Pleas explain. |
|
| Back to top |
|
|
vijay
Beginner
Joined: 09 May 2003
Posts: 131
Topics: 64
|
| Posted: Mon Jul 17, 2006 11:50 am Post subject: |
|
|
I'm just trying to compare files without merge jetson
Vijay |
|
| Back to top |
|
|
jetson
Beginner
Joined: 07 Oct 2005
Posts: 30
Topics: 2
Location: Texas
|
| Posted: Mon Jul 17, 2006 1:03 pm Post subject: |
|
|
PROC SQL should be much easier than data step (unless data step is required.) It looks like you're trying to process each record from the first table against each record from the second table...definitely a painful process if you have a lot of data.
PROC SQL can do the cartesian join and filtering... |
|
| Back to top |
|
|
advoss
Beginner
Joined: 23 Aug 2005
Posts: 26
Topics: 0
|
| Posted: Wed Jul 19, 2006 11:27 am Post subject: |
|
|
Why not use a merge? What are you looking for that it will not provide?
| Code: |
DATA COMBINE;
merge filein1(rename=(ID1=ID) in=in1)
filein2(rename=(ID2=ID) in=in2);
by ID;
if in1 then ID1 = ID;
if in2 then ID2 = ID;
drop ID;
run;
|
_________________
Alan Voss |
|
| Back to top |
|
|
|
|
|
FAQ
Search
Quick Manuals
Register
Profile
Log in to check your private messages
Log in
