Re: Problem with (using) autogsdoc

[Richard Frith-Macdonald]

On Wednesday, September 18, 2002, at 10:35 PM, Chris B. Vetter wrote:

> On Wed, 18 Sep 2002 22:27:17 +0100
> Richard Frith-Macdonald <richard@brainstorm.co.uk> wrote:
> > I'm having trouble figuring out what you mean.
> > Are you expecting author information to be inserted into the
> > FunctionsTemplate
> > file?  Nothing apart from function documentation is put there ... it's
> > up to you to create the template you want.  Or use -FunctionsTemplate
> > "" to tell autogsdoc not to merge functions from different files into
> > one place.
>
> I have a standard header for .H and .M files that get parsed by sed(1)
> when I create a new "file set" (that is foobar.m and foobar.h)
>
>   /** FILE: foobar.[hm]
>     *
>     * <title>foobar</title>
>     *
>     * Copyright (c) Foobar, Inc
>     * All rights reserved.
>     *
>     * Written by: author (email)
>     *
>     [...]
>
> I was wondering, why that information IS picked up for files 
> containing a
> class, but NOT for files only containing (C) functions.

I think this confusion is because I left autogsdoc with some loose 
ends ...
I left a hack in place that left the various template flags turned on by
default (I've removed that), and didn't document the use of those flags
(I've added some more documentation now).

If you run autogsdoc normally (ie. not set to combine all functions into
a single file), foo.[hm] should produce a foo.gsdoc and a foo.html with
the function documentation along with author, copyright etc.

If you tell autogsdoc to combine the function documentation into a 
template,
you ideally create that template yourself ... as it contains fragments 
from
possibly very many source files, it does not know what copyright/author 
etc
info would apply - so you put the info you want into the template 
manually.