Path: utzoo!attcan!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!zaphod.mps.ohio-state.edu!rpi!leah!wfh58
From: wfh58@leah.Albany.Edu (William F. Hammond)
Newsgroups: comp.sys.amiga
Subject: Re: ARRRRRRRRRRGGGGGGGGGHHHHHHHHHHHH!!!!!!!!!!!
Summary: bra and ket, not T:
Message-ID: <3413@leah.Albany.Edu>
Date: 28 Jul 90 16:21:30 GMT
References: <2649@mindlink.UUCP>
Reply-To: wfh58@leah.albany.edu.UUCP (William F. Hammond)
Organization: Dept of Math & Stat, SUNYA, Albany, NY
Lines: 24

In article <2649@mindlink.UUCP> a217@mindlink.UUCP (Vincent Lim) writes:
> ...
>> >      EXECUTE: No K directive
>> Ah, the dreaded "No K directive".  Try putting the following two lines in
>>     .bra <
>>     .ket >
> ...
>I'm not CBM but I do have an observation which I'd like to share. I've noticed
>that if your script performs an Execute and you have no T: directory assigned
>and no SYS:t directory exists or is not writable then you will get the bogus
   My versions of "execute", both standard and ARP, attempt to create a "T"
directory in the root above the current directory.  (Additionally, ARP's
"execute" will assign "T:" to it.)  I am unable to provoke a "no K directive"
message by eliminating ":T" and "T:".
   My suspicion is that the message is triggered by the appearance in the
script of both the '>' and '<' characters for input/output diversion.  Since
these are the default "bra" and "ket" characters, "execute" is yelling about
the absence of a ".key" at the top of the script.  For this most of the time
redefining the "bra" and "ket" characters will take care of the problem.
----------------------------------------------------------------------
William F. Hammond                   Dept. of Mathematics & Statistics
518-442-4625                         SUNYA, Albany, NY 12222
wfh58@leah.albany.edu                wfh58@albnyvms.bitnet
----------------------------------------------------------------------