Date:         Tue, 25 May 2004 13:59:24 -0400
Reply-To:     Peter Crawford2 <peter.crawford@BLUEYONDER.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Peter Crawford2 <peter.crawford@BLUEYONDER.CO.UK>
Subject:      Re: Input comma delimited file

As this is a csv file, I would like to recommend a very useful
feature of sas infile statements   DSD
It makes this step a lot simpler, as long as you declare
your variables before using them... See this proposal

DATA TEST;
  INFILE FILEIN1 dsd /* implies comma delimiter */;
  length EMAIL $46 CUST_NUM 8;
  INPUT  EMAIL     CUST_NUM ;
RUN;
PROC PRINT DATA = TEST (OBS=250);
RUN;

Not only does the DSD option imply that comma delimiter,
it also adds support for
    consecutive delimiters - implying a missing column,
    quotes that protect delimiters embedded in data
    embedded blanks in character data ( delimiters take precedence )

There may be more reasons to applaud DSD, but I would have to start
thinking.....

Peter Crawford
Crawford Software Consultancy Limited
UK


On Tue, 25 May 2004 11:55:19 -0400, Gerhard Hellriegel <ghellrieg@T-
ONLINE.DE> wrote:

>On Tue, 25 May 2004 11:10:14 -0400, Lu Liu <Lu.Liu@TALBOTS.COM> wrote:
>
>>Hi,
>>
>>I have a beginner's question.  I used the following codes to input a
comma
>>delimited file with 2 variables, email (min 19, max 46) and customer
number
>>(min 7, max 8).  But I am only getting the first 8 columns in the email
>>field.  How do I correct it?
>>
>>DATA TEST;
>>  INFILE FILEIN1 DLM=',';
>>  INPUT EMAIL $ CUST_NUM;
>>PROC PRINT DATA = TEST (OBS=250);
>> RUN;
>>
>>
>>
>>Obs     EMAIL      CUST_NUM
>>
>>  1    RJBEAN10     1005661
>>  2    CAROLGEL     1050773
>>  3    JLCRIBBS     1063455
>>  4    UTA@SONG     1073812
>>  5    MYRNA@EP     1114518
>>
>>Thank you very much for your help!
>>
>>Lu
>
>you should use:
>
>DATA TEST;
>  INFILE FILEIN1 DLM=',' missover;
>  INPUT EMAIL : $46.  CUST_NUM : 10.;
>PROC PRINT DATA = TEST (OBS=250);
> RUN;