SAS-L archives -- February 2003, week 1 (#274)

Date: Wed, 5 Feb 2003 16:33:51 -0600 Reply-To: Kevin Myers <KevinMyers@AUSTIN.RR.COM> Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> From: Kevin Myers <KevinMyers@AUSTIN.RR.COM> Subject: Re: Antwort: Re: comma delimited field parsing Content-Type: text/plain; charset="iso-8859-1"

Thanks for the great tip Peter! Even after 20+ years of writing SAS code, it seems that I still learn new tricks very day.

s/KAM

----- Original Message ----- From: "Peter Crawford" <peter.crawford@db.com> To: "Kevin Myers" <KevinMyers@AUSTIN.RR.COM> Sent: Wednesday, February 05, 2003 4:20 PM Subject: Antwort: Re: comma delimited field parsing

in v8 you can write into _infile_ That is as simple as it gets...

Of course, you only get "one of those" in a datastep

To impose the dsd rules, you need an infile statement, but that _almost_ can point at dummy

Here is a trial ...... 245 data test1; 246 input var1 : $15. ; 247 put _all_; 248 cards;

var1=a,b,c,d,e,f _ERROR_=0 _N_=1 var1=,,,,, _ERROR_=0 _N_=2 var1=,,c,d,e,f _ERROR_=0 _N_=3 var1=a,b,,,e,f _ERROR_=0 _N_=4 NOTE: The data set WORK.TEST1 has 4 observations and 1 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.01 seconds

253 ; 254 data _null_; * ensure a dummy demo file is available; 255 file '!temp/demo'; put 'demo'; stop; 256 run;

NOTE: The file '!temp/demo' is: File Name=C:\TEMP\demo, RECFM=V,LRECL=256

NOTE: 1 record was written to the file '!temp/demo'. The minimum record length was 4. The maximum record length was 4. NOTE: DATA statement used: real time 0.01 seconds cpu time 0.00 seconds

257 data test2; 258 set test1; 259 * ensure an infile buffer is available; 260 infile '!temp/demo' dsd dlm=',' truncover lrecl=32000; 261 input @1 @@; 262 _infile_=var1; * load var1 ready for "input" parsing; 263 *define the vars to parse out of var1 ; 264 length a b c d e f $20; 265 input a b c d e f @@; **use @@ to keep that buffer; 266 put ( a--f )(=) @30 var1= ; 267 run;

NOTE: The infile '!temp/demo' is: File Name=C:\TEMP\demo, RECFM=V,LRECL=32000

a=a b=b c=c d=d e=e f=f var1=a,b,c,d,e,f a= b= c= d= e= f= var1=,,,,, a= b= c=c d=d e=e f=f var1=,,c,d,e,f a=a b=b c= d= e=e f=f var1=a,b,,,e,f NOTE: 1 record was read from the infile '!temp/demo'.

So it works !

Sorry you're not making it to sugi !!

Regards Peter Crawford

Datum: 05/02/2003 23:13 An: SAS-L@LISTSERV.UGA.EDU

Antwort an: Kevin Myers <KevinMyers@AUSTIN.RR.COM>

Betreff: Re: comma delimited field parsing Nachrichtentext:

Hi Peter -

Using dsd works great as long as you are reading values directly from a file, but not if you already have them in a variable. But maybe I am missing something...? You say "Just load _infile_ with var1, ...". I don't know how to accomplish that other than reading var1 directly from a dsd file. Do you have some other trick up your sleeve to get the value of var1 into _infile_? I'm guessing that you must, but I can't imagine what it might be offhand...

Sure do wish that I could make SUGI, but unfortunately I can't afford to make the trip. I've always wanted to go, but was never able to make the trip, even while I was with SAS. :-(

s/KAM

----- Original Message ----- From: "Peter Crawford" <peter.crawford@db.com> To: "Kevin Myers" <KevinMyers@AUSTIN.RR.COM> Cc: <SAS-L@LISTSERV.UGA.EDU> Sent: Wednesday, February 05, 2003 3:47 PM Subject: Re: comma delimited field parsing

Hi Kevin

I think the easy way to do this kind of thing is using the strength of dsd (theme of my sugi coders corner paper "More _Infile_ Magic" )

Just load _infile_ with var1, and then easily "input" those columns It only needs preceeding with a dsd based infile. Any quoted values are fully respected in dsd rules.

Hope you'll be able to come along to SUGI

Regards Peter Carwford

Datum: 05/02/2003 22:41 An: SAS-L@LISTSERV.UGA.EDU

Antwort an: Kevin Myers <KevinMyers@AUSTIN.RR.COM>

Betreff: Re: comma delimited field parsing Nachrichtentext:

If you can insure that your original strings won't include quoted values with imbedded commas, then you can go with Michael's approach of using tranwrd to pre-process the string before passing it into scan(). However, in the more general case that won't work because it will alter any values within quotes that include embedded commas.

s/KAM

----- Original Message ----- From: "Charles Patridge" <Charles_S_Patridge@PRODIGY.NET> Newsgroups: bit.listserv.sas-l To: <SAS-L@LISTSERV.UGA.EDU> Sent: Wednesday, February 05, 2003 2:47 PM Subject: Re: comma delimited field parsing

> Michelle, > > In the documentation of SAS, here is what I found: > > /*** SCAN FUNCTION: > SAS Language - Version 6 - Chapter 11 - Page 595 > Leading Delimiters before the first word have no effect. If there are > two or more contiguous delimiters, SCAN treats them as one. > ***/ > > > I'll see what I can do to get what you would like. > Regards, > Charles Patridge > > If you were reading in VAR1 = "a,b,c,d,e,f" from an external file, then > I would have suggested using the DSD option of the INFILE statement which > should have worked even if you had such strings as ",,c,,e,f". > > but dealing with a SAS variable already in that form requires a different > solution.

Diese E-Mail enthält vertrauliche und/oder rechtlich geschützte Informationen. Wenn Sie nicht der richtige Adressat sind oder diese E-Mail irrtümlich erhalten haben, informieren Sie bitte sofort den Absender und vernichten Sie diese Mail. Das unerlaubte Kopieren sowie die unbefugte Weitergabe dieser Mail ist nicht gestattet.

This e-mail may contain confidential and/or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden.

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

listserv.uga.edu
Enterprise Information Technology Services
The University of Georgia
listhelp@uga.edu