Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!iuvax!rutgers!ucla-cs!gast
From: gast@lanai.cs.ucla.edu (David Gast)
Newsgroups: comp.editors
Subject: Re: re: Yet another vi question (summary)
Summary: IT IS POSSIBLE
Message-ID: <16745@shemp.CS.UCLA.EDU>
Date: 12 Oct 88 21:56:33 GMT
References: <270@tijc02.UUCP>
Sender: news@CS.UCLA.EDU
Reply-To: gast@cs.ucla.edu (David Gast)
Organization: UCLA Computer Science Department
Lines: 68

In article <270@tijc02.UUCP> djm408@tijc02.UUCP (David Marks         ) writes:
>I am still getting responses to my question on entering vi with a blank
>template. Most of you are missing the point I was trying to make and why I
>asked the question in the first place.
>
>        around a problem, I usually am curious enough to want an
>        answer to why the original solution would not work.

It seems to be a ``feature.''  But a solution does exist, see below.

>
>     2. I was trying to avoid copying the template file to the report
>        before editing it. Doing this from a program means that I have
>        to invoke a shell which creates another process that really should
>        not be needed.

No need to invoke the shell or to create any new processes if you are
writing a C program.  Just use the function calls described in the
manual. I use BSD so I do not know the precise names you will have to
use for System V.

>
>     3. I feel that vi should be able to this. No manual I have read
>        (no flames on UNIX manuals and books please :-) ) even comes
>        close to explaining why this might not work. All claim that
>        ANY ex command can be executed on entry to vi, including those
>        that take arguments; most show examples of the "set" command.

I just checked one.  It says that "most" or "many" (I forget which one)
commands work.  It does not say that all work.  It is, however, either
a bug or a bad feature that not all do not work.

>
>Those of you that thought that my vi is damaged, should know that we are using
>the same vi that is distributed by AT&T with System V rel. 2.2, and thus AT&T
>must have a damaged vi.

As you probably know, AT&T did not write vi.  There is nothing that says
that Berkeley or AT&T never write code with bugs or features.

>I resent the attitude that some of you took trying to explain to me how to use
>shell scripts to solve my problem.

Please try to excuse us.  How many cpu cycles do you figure you will save
by not using a shell script?  How many CPU cycles have been spent sending
your message and all of the replies across the net?  Is it a positive
number or would have it been better to do it in a shell script to begin
with?  In the Bourne shell or Korn Shell, you could have done
	> file
to create a file without starting any extra processes.  The shell works
great for many programs.

There is a way to do exactly what you want to do and all from within
vi.  If I understand, you want to be able to invoke vi, with something
like ``vi file'', where file does not exist, but when vi is called a template
is copied into file automatically.

Since you resent all suggestions, here is my offer.  Send me a $100
cashiers check and I will send you the solution.  It only uses vi.
I do not claim that it is better than the above suggestions, only
that it works.

David Gast
1446 Federal Avenue #A
Los Angeles, CA 90025
gast@cs.ucla.edu

{uunet,ucbvax,rutgers}!{ucla-cs,cs.ucla.edu}!gast