Date: Mon, 3 Aug 1998 17:20:08 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Michel de Ruiter <mdruiter AT cs DOT vu DOT nl>
cc: dj AT delorie DOT com, djgpp-workers AT delorie DOT com
Subject: Re: Some notes about the 2.02 alpha...
In-Reply-To: <m0z3JtP-000XepC@sloep06.cs.vu.nl>
Message-ID: <Pine.SUN.3.91.980803171320.3493E-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Mon, 3 Aug 1998, Michel de Ruiter wrote:

> Why (special reason?) is libc [efg]cvt\(buf\)?-documentation in
> src/libc/compat/stdlib/itoa.txh, instead of their respective *.txhs?

There's no requirement that each module should have its own .txh file.  
There are quite a few of other modules that share a single .txh file.

> Where is info/libm.inf? Shouldn't it be in djdev202.zip?

Definitely.

> Why is bin/coff2exe.bat a batch-file, and not a symlink
> bin/coff2exe.exe to stubify?

Why is this important?

> (I noticed that the Emacs building procedure uses `coff2exe' in the 
> batch-file, 

This is only true when you build Emacs with DJGPP v1.x.  Building with v2 
doesn't use coff2exe at all, it uses stubify.  See msdos/mainmake.v2.

> which should otherwise be `call coff2exe' to be correct and compatible 
> with v1. 

You don't need (and shouldn't need) to use `call' in a Makefile.  A 
Makefile, conceptually, is just like a command line: you need to type 
commands there exactly as you would from a DOS prompt.  So if a Make 
program requires you to use `call' with a batch file, it has a bug.  
DJGPP's ports of Make don't have this bug, AFAIK.