|
In article <K6XaaMAqq5N2EwaT@crawfordsoftware.demon.co.uk>,
Peter Crawford <Peter@CRAWFORDSOFTWARE.DEMON.CO.UK> wrote:
> In article <909601402.2129690.0@vm121.akh-wien.ac.at>, Quan Li
> <qxl4@PSU.EDU> writes
> >Dear SAS-L,
> >
> >I have multiple external files (about 100 ) that are in the same format
> >(tab-delimited) and with the same variables that I would like to merge into
> >one sas-dataset. Instead of using multiple data steps and infile
> >statements, I wonder if there is an easier solution, such as reading them
> >in one datastep and then merge into one.
> >
> >Thanks for any help.
> >
> >Best,
> >Quan
> >Quan Li
> >Assistant Professor
> >Department of Political Science
> >The Pennsylvania State University
> >Phone: 814.865-6575
> >Fax: 814.865-8979
> >Email: qxl4@psu.edu, quanli@psu.edu
>
> lookup what FILEVAR option of infile satatement does
>
> basically, it names the variable containing the external file name from
> which you want to read
>
> filename fileref '.'; /* just make any dummy (but existing) fileref */
> data allofit(label='all 100 inputs together'
> keep=var1 -- varz) ;
>
> infile cards; /* or somewhere else with your 100 files named */
> length from $66; /* allowing (slightly) long file names */
> input from &; /* make it more complex to support multi-blanks */
>
> length var1 - var25 8 varA $16 /* ...... */ varZ $1;
> /* whatever */
>
> infile fileref filevar =from ..etc.. end=eof dsd dlm=',' truncover;
>
> do until( eof);
> input var1 -- varZ ;
> output; ****---------------<<<<<<<<;
> end;
> cards;
> first filename
> second filename
> third filename
> ;
> --
> Peter Crawford
Why make it so complicated ? Surely it's simpler to concatenate the files
into a single fileref? e.g. filename all ('c:\files\file1.dat'
--
Black Cat Solutions Ltd.
SAS Software Specialists
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
|
