Message-ID:	<B0000085011@stargate.astr.lu.lv>
From:	"Andris Pavenis" <pavenis AT lanet DOT lv>
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Date:	Tue, 27 Apr 1999 12:50:42 +0300
MIME-Version:	1.0
Subject:	Re: v2.03: wrapping up
CC:	djgpp-workers AT delorie DOT com
References:	<199904261652 DOT MAA06425 AT envy DOT delorie DOT com>
In-reply-to:	<Pine.SUN.3.91.990427101542.18849D-100000@is>
X-mailer:	Pegasus Mail for Win32 (v3.02b14)
Reply-To:	djgpp-workers AT delorie DOT com

On 27 Apr 99, at 10:16, Eli Zaretskii wrote:

> 
> On Mon, 26 Apr 1999, DJ Delorie wrote:
> 
> > Could we have the specs add -i or something, so that gcc always
> > #includes a special version file when compiling?
> 
> That's what I suggested a year ago when this problem was discussed in
> the context of GCC 2.8.x.  I think using "-imacros" to include
> sys/version.h will solve this cleanly and reliably.

I don't think so. Imagine if somebody will want to test something with 
DJGPP-2.00 or 2.01 which didn't have sys/version.h. 

> The problem is that there seems to be no way of having specs expand
> environment variables (last time I looked), so you cannot expand
> $DJDIR or $C_INCLUDE_PATH to know the exact path to sys/version.h, and
> "-imacros" doesn't look along the include parth automatically (again,
> last time I checked).

both egcs-1.1.2 and egcs 2.93.18 19990412 I have tesed does NOT expand 
environment variable in that place 

> Perhaps some of this has changed since then, and the solution is just
> waiting to be discovered?
> 
> Could someone please look at the latest GCC/EGCS sources and see if
> there is some way of including a header whose name is relative to the
> include directory?

I don't think there is any without modifying sources of gcc (perhaps gcc/cccp.c)

As I said even such modification to get sys/version.h to be included 
automatically before other include files will forbid us to even try to use
some earlier DJGPP versions that didn't have sys/version.h (such possibility may 
be possible for testing purposes). Perhaps including stdio.h there is not the thing 
we want.

So I want to ask for some experiments:
	My instalation does not define DJGPP_MINOR in specs (I have removed
 	lib/specs). I haven't met any problems with it. Yesterday I tried to look
	at sources of emacs-19.34 and didn't find any problems when building it.
	Grepping sources also shows that all should be Ok.

	Therefore suggestion: Maybe also somebody else should check is things
	so grave when one should include stdio.h if DJGPP_MINOR is needed.
	I myself haven't seen any problems. Binaries of DJGPP port of egcs-1.1.2
	does not define DJGPP_MINOR in specs. Therefore it would be easy to test
	this with this version.

Andris

- Raw text -