|
Nick,
You have to consider that, while in Unix piping and running SAS may be two
different simultaneous asynchrone processes interchanging data, in DOS and
Windows a pipe always is sequentially (synchroneously) processed before the
calling program (SAS) proceeds further. This means that, though you do not
see the intermediate scratch file from the pipe, it actually is there (in a
temp directory) and you need sufficient scratch space to store the whole
copy of the original file in. I have stressed this once or twice more on the
list already.
Regards - Jim.
--
Y. (Jim) Groeneveld, MSc IMRO TRAMARKO tel. +31 412 407 070
senior statistician, P.O. Box 1 fax. +31 412 407 080
head IT department 5350 AA BERGHEM IMRO TRAMARKO: a CRO
J.Groeneveld@ITGroups.com the Netherlands in clinical research
My computer does not need me at all, but I can't do without it anymore.
> -----Original Message-----
> From: Nick Paszty [SMTP:npaszty@ORGANIC.COM]
> Sent: Tuesday, June 27, 2000 8:28 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Reading from a pipe UNIX and NT
>
> Hello.
>
> I would like to read data from a file without completely opening it. I
> have some very large gz files that are zipped simple text files. I would
> like to unzip them into a pipe and read the data from the pipe into
> SAS. This is being done using a Perl process as the analysis tool but I
> would like to replace Perl with SAS. The other problem is that I'm on PC
> SAS. I would rather not open the file completely which could turn into an
> 800 MB file and then read that file in. Can SAS create a program data
> vector without knowing what all records look like? Is this at all
> possible? Sure would be a nice space saver!
>
> Thanks,
>
> Nick
|