Date: Thu, 21 Apr 2011 14:17:09 -0500
Reply-To: Joe Matise <snoopy369@GMAIL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Joe Matise <snoopy369@GMAIL.COM>
Subject: Re: Time informat don't work for decimal?
In-Reply-To: <201104211914.p3LIrAQI015916@waikiki.cc.uga.edu>
Content-Type: text/plain; charset=ISO-8859-1
No. xx:yy is read in as hh:mm, not mm:ss. I see no reason SAS should be
expected to use a particular format in its log...
-Joe
On Thu, Apr 21, 2011 at 2:14 PM, Ya Huang <ya.huang@amylin.com> wrote:
> Ok,
>
> All these numbers are from SAS log file, real time and cpu time.
> If SAS formatted the time this way, shouldn't it be able to read back in?
>
> Why 12:23.20 is not considered a valid time format? Shouldn't it
> be interpreted as 12 minutes and 23.2 seconds?
>
>
>
> On Thu, 21 Apr 2011 12:00:11 -0700, Schwarz, Barry A
> <barry.a.schwarz@BOEING.COM> wrote:
>
> >The first is not a valid time format and the last does not mean what you
> want it to (it actually means 56:34:00.00 and not the 00:00:56.34 you want)
> so the answer to your question is probably no.
> >
> >-----Original Message-----
> >From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Ya
> Huang
> >Sent: Thursday, April 21, 2011 11:43 AM
> >To: SAS-L@LISTSERV.UGA.EDU
> >Subject: Re: Time informat don't work for decimal?
> >
> >So, if I have a char var, with all the possible time format values:
> >
> > 12:23.20
> >1:23:43.12
> > 56.34
> >
> >Any simple way to read in as SAS time (without having to test
> >the string first)?
> >
> >
> >
> >On Thu, 21 Apr 2011 14:35:44 -0400, Ya Huang <ya.huang@AMYLIN.COM> wrote:
> >
> >>"but not what SAS assumes (as a makes clear)"
> >>
> >>So, SAS assumes that the string has to be in HH:MM:SS.ss
> >>format, or if it is MM:SS.ss format, we have to use time5.?
> >>
> >>This one is even weird:
> >>
> >>6 data _null_;
> >>7 a='50.5';
> >>8 b=input(a,time.);
> >>9 put b=;
> >>10 run;
> >>
> >>b=180300
> >>
> >>I wonder what SAS 'Assumes' this time?
> >>
> >>
> >>On Thu, 21 Apr 2011 13:20:38 -0500, Joe Matise <snoopy369@GMAIL.COM>
> wrote:
> >>
> >>>I think it's an issue of "what is what"... see this:
> >>>data test;
> >>>x='18:25.23';
> >>>y = input(x,TIME.);
> >>>z = input('00:'||x,TIME.);
> >>>a = input(substr(x,1,5),TIME.);
> >>>put _all_;
> >>>run;
> >>>
> >>>y as you note does not work. Z and A both do though, _but are not the
> >>>same_.
> >>>
> >>>So when SAS sees 18:25.25, what is it supposed to think?
> >>>HH:MM.ss or MM:SS.ss ? The former is illegal... the latter is fine, but
> >>not
> >>>what SAS assumes (as a makes clear).
> >>>
> >>>-Joe
> >>>
> >>>On Thu, Apr 21, 2011 at 1:17 PM, Ya Huang <ya.huang@amylin.com> wrote:
> >>>
> >>>> Hi there,
> >>>>
> >>>> I was really really surprised today:
> >>>>
> >>>> 1 data _null_;
> >>>> 2 a='18:25.23';
> >>>> 3 b=input(a,time.);
> >>>> 4 put b=;
> >>>> 5 run;
> >>>>
> >>>> NOTE: Invalid argument to function INPUT at line 3 column 3.
> >>>> b=.
> >>>> a=18:25.23 b=. _ERROR_=1 _N_=1
> >>>> NOTE: Mathematical operations could not be performed at the
> >>>>
> >>>> If I remove the decimal part ".23", it works fine. But clearly the
> >>document
> >>>> says decimal part is OK:
> >>>>
> >>>> The TIMEw. informat reads SAS time values in the following form:
> >>>>
> >>>> hh:mm:ss<.ss> <AM | PM>
> >>>>
> >>>> Something obvious I missed?
> >>>>
> >>>> Thanks
> >>>>
> >>>> Ya
> >>>>
>
| 
