Newsgroups: rec.arts.int-fiction
Path: nntp.gmd.de!news.ruhr-uni-bochum.de!news.uni-stuttgart.de!uni-regensburg.de!lrz-muenchen.de!informatik.tu-muenchen.de!Germany.EU.net!howland.reston.ans.net!torn!watserv3.uwaterloo.ca!undergrad.math.uwaterloo.ca!svanegmo
From: svanegmo@undergrad.math.uwaterloo.ca (Stephen van Egmond)
Subject: Inform 6 porting outcome
Sender: news@undergrad.math.uwaterloo.ca (news spool owner)
Message-ID: <DquCFJ.Jyv@undergrad.math.uwaterloo.ca>
Date: Fri, 3 May 1996 17:54:55 GMT
References: <1996Apr29.085915@oxvaxd> <4m5s3d$2q1@nntp4.u.washington.edu> <4m73fu$djp@peak.org> <4m9d77$1cr@newsstand.cit.cornell.edu>
Nntp-Posting-Host: cantor.math.uwaterloo.ca
Organization: University of Waterloo
Lines: 32

I was unable to do very extensive testing, since I don't have much Inform 
source.  I received a litany of errors when I tried to compile 
magic-toyshop.inf, but I suspect that was because the version of the code 
I was using was using withdrawn directives and statements.

I did manage to compile the portion of "through the looking glass" that 
was posted on Gareth's home page without difficulty or warnings.

How did I do this?

I found that header.h was making a few assumptions that caused trouble:

int32 wasn't defined before the OS-dependent IFDEFs were compiled.  This 
was required since (for me) the Amiga IFDEF had a function that returned 
int32.

More seriously, I think that there is an IFDEF (within the Amiga code, 
not sure about others) that refers to INFORM_FILE that brackets the 
function returning int32 referred to above (specifically, the function 
that returns a magic "temporary" number based on the process ID used for 
temp files or something).  This should be MAIN_INFORM_FILE (or whatever 
is defined at the top of inform.c).

I think that is the extent of significant problems I found.

Oh, yes, I had to include limits.h (ANSI) for the benefit of the int32 
IFDEFs.  Those ingenious comparisons make use of DEFINEs from limits.h.  
The tacit assumption was that the OS-dependent IFDEFs would find those 
constants at some point.

/Steve

