Date:         Mon, 13 Sep 2004 08:42:29 -0700
Reply-To:     kurts@field.com
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Kurt Schutzmann <kurts@FIELD.COM>
Subject:      Re: About Import Procedure, ASK FOR HELP!!!
Comments: To: "Rex.Lv" <rex.lv@126.com>
In-Reply-To:  <000801c49888$a3d5f9a0$0a00020a@rex>
Content-Type: text/plain; charset="iso-8859-1"

filename infile "c:\myfile" recfm=v lrecl=10000; %* assuming file is on a
PC;

proc import out=sasfile data=infile dbms=csv;
guessingrows=100; %* this option may not be available in v8. Run it to find
out;
run;

-----Original Message-----
From: Rex.Lv [mailto:rex.lv@126.com]
Sent: Saturday, September 11, 2004 10:23 PM
To: kurts@field.com
Subject: Re: About Import Procedure, ASK FOR HELP!!!


It doesn't work. Could you give me an example SAS code? I use SAS 8.2. Thx.!


Best Regards,
Rex
----- Original Message -----
From: "Kurt Schutzmann" <kurts@field.com>
To: "Myra Oltsik" <my_sas_spot@YAHOO.COM>; <SAS-L@LISTSERV.UGA.EDU>;
<KUMARA@ESSILOR.COM.SG>; <rex.lv@126.COM>
Sent: Saturday, September 11, 2004 3:05 AM
Subject: RE: About Import Procedure, ASK FOR HELP!!!


> If you have a CSV or DLM file, then GUESSINGROWS= can be helpful.
>
> For excel files, try specifying the exact range (RANGE=) or the entire
> spreadsheet.
>
> There is even one for access to specify the size of memo fields
(MEMOSIZE=).
>
> Regards,
> Kurt
>
> -----Original Message-----
> From: Kurt Schutzmann [mailto:kurts@field.com]
> Sent: Friday, September 10, 2004 11:39 AM
> To: Myra Oltsik; SAS-L@LISTSERV.UGA.EDU; KUMAR Arun
> [KUMARA@ESSILOR.COM.SG]; Rex.Lv [rex.lv@126.COM]
> Subject: RE: About Import Procedure, ASK FOR HELP!!!
>
>
> Better option: use the GUESSINGROWS option when reading in the file.
>
> Kurt
>
> -----Original Message-----
> From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU]On Behalf Of Myra
> Oltsik
> Sent: Friday, September 10, 2004 8:04 AM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Re: About Import Procedure, ASK FOR HELP!!!
>
>
> Rex,
>
> I've had the same experience. When SAS reads in an Excel file it gets
> the length and format from the first "few" records it reads. I don't
> know how many records it reads before it decides on the attributes,
> however. If you put at least one of your longest records near the top
> of your data you should get an appropriate length. If you don't care
> what order the data are read in -- if you have an "ID" variable on
> which you can sort -- then rearranging the records won't matter. If
> the order you read in the records is important, then create a new
> column with the row numbers on which you can sort the data later, and
> then manually rearrange your records so SAS gets the right attributes.
>
> I hope this helps.
>
> Myra
>
> KUMARA@ESSILOR.COM.SG (KUMAR Arun) wrote in message
> news:<07123CDF07D82249A6A840A789BC2EEEDDA5CB@sges0004.essilor.com.sg>...
> > Hi Rex,
> > I remember facing the same type of problems. Perhaps SAS decide the
length
> > of columns based on few observations(I don't remember exactly perhaps
20).
> > So one way out is go to the sas log, copy the relevant proc import codes
> and
> > change the length according to your need for the columns.
> >
> > Regards,
> > Arun Kumar
> >
> > -----Original Message-----
> > From: Rex.Lv [mailto:rex.lv@126.COM]
> > Sent: Friday, September 10, 2004 3:04 PM
> > To: SAS-L@LISTSERV.UGA.EDU
> > Subject: About Import Procedure, ASK FOR HELP!!!
> >
> >
> > Hi,
> >
> >
> > I want to import some data from excel files, but I'm lost. The columns
> that
> > have over 25 characters are automatically truncated to 25 character.
> Anybody
> > could give me a hand? Thx. a lot!
> >
> >
> > The code that I used to import data is as followed:
> >
> > PROC IMPORT OUT= WORK.t
> >             DATAFILE= "C:\datafile.xls"  /*The excel file includes the
> > data*/
> >             DBMS=EXCEL2000 REPLACE;
> >      RANGE="Sheet1$";                    /*The sheet I want to import*/
> >      GETNAMES=YES;
> > RUN;
> >
> >
> > Best Regards,
> > Rex
>
>
>