LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (April 2003, week 5)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Tue, 29 Apr 2003 16:30:35 +0000
Reply-To:     Peter Crawford <peter.crawford@DB.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Peter Crawford <peter.crawford@DB.COM>
Subject:      Re: A strange problem.
Comments: To: Gerhard Hellriegel <ghellrieg@T-ONLINE.DE>
Content-Type: text/plain; charset="us-ascii"

I don't think Ron's "pad" option will help, because the CR of CRLF is being treated as data on unix. Pad will add trailing blanks to that line, which won't make that CR problem go away. Adding CR to the defimiter set should solve the issue If no delimiter is already defined, adding CR requires only in infile ........ dlm='200D'x .........; If it is a comma delimiter, then dlm='2C0D'x If the data is being read by proc import, then catching the generated code provides a "code base" in which to modify the infile statement (as well as other things like var lengths, types and informats)

By the way, does anyone else think SI should at least fix their generated code so that it doesn't give a bad example ! Copied from the generated code if _ERROR_ then call symput('_EFIERR_',1); /* set ERROR detection macro variable */ generates the too familiar.. bad example "converted" messages in the log NOTE: Numeric values have been converted to character values at the places given by: (Line):(Column). 17:44 So easily fixed by generating, instead if _ERROR_ then call symput('_EFIERR_','1');

Regards Peter Crawford

Gerhard Hellriegel <ghellrieg@T-ONLINE.DE> Sent by: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> 29/04/03 15:50 Please respond to Gerhard Hellriegel

To: SAS-L@LISTSERV.UGA.EDU cc: Subject: Re: A strange problem.

On Tue, 29 Apr 2003 08:28:27 -0400, Fehd, Ronald J. (PHPPO) <rjf2@CDC.GOV> wrote:

>> From: tina [mailto:tyang@ARCHES.UGA.EDU] >> I am not very good with SAS and have no idea how to solve >> this problem. I sincerely hope that someone can help me with this. >> >> I have a dataset that I created from a text file, > >classic gotcha for WinDoze: >variable length records, which result in >"SAS went to new line" >warnings >read your log and see what the widest/longest record length is >and put that as the value of lrecl: > >infile TEXTFILE > lrecl = 80 pad <---<<< > ; > >Ron Fehd the macro maven CDC Atlanta GA USA RJF2@cdc.gov > >My computer must be broken: >whenever I ask a wrong question, > I get a wrong answer. >-- Pot-Shots by Ashleigh Brilliant > >My computer must be broken: >whenever I read a file with variable length records > I get a wrong answer. >-- Tina

... you should also try out the options for infile like MISSOVER or TRUNCOVER, which normally gets you out of "SAS went to a new line..." The default max-lrecl is 255, maybe too small. Set it to a higher value in that case!


Back to: Top of message | Previous page | Main SAS-L page