Attempting to compile ADL; a quick question

[Ross Cunniff]

In article <589 at sbsvax.UUCP> greim at sbsvax.UUCP (Michael Greim) writes:
> In article <438 at coplex.UUCP>, dean at coplex.UUCP (Dean Brooks) writes:
> < 
> <    Let me clarify a few of the problems I have had with this.  It seems
> < that ADLRUN does weird things (like core dump or ignore all input) whenever
> < I compile ANY adventure.  This includes EVERY sample game released with
> < ADL (let alone any I would write).
> < 
> <    One interesting note...  Every time I compile a game, it gives two
> < warnings.  It says TELLER and PRINT(?) is undefined.  Other than that
> Really STRING and TELLER are undefined. See the manual for their meaning.

Correct.  STRING is the NOUN used when the player enters a string in
a sentence.  TELLER is the VERB used when the player enters a sentence
like 'Robot, go east then south'.  If these are undefined, nothing
BAD happens, you just don't get access to these features.

> < the COMPILE seems to work fine.  When I RUN the game, it core dumps or
> < ignores all my input; it will say things like "I dont understand you".
> < 
> <    I have set up the adltypes.h correctly (to my knowledge).  We have had
> < no problems porting most other SYS 5.2 programs to our machine, so I
> < would tend to believe other people are having this problem.  Any clues?
> Sounds like you came across some compiler bug. Or in adlrun there might
> be some assumptions like "pointers are the same size as ints" or
> "NULL is all bits zero". ADLRUN works on the computers I have available
> here, which all run some kind of UNIX(TM). I have tried it on VAX 11/780
> (4.3BSD), SUN 3/160 (SUN OS 3.5), SIEMENS PC-MX2 (SINIX 2.1). I have not
> yet tried it on our VAX 86000 (ULTRIX).
> What machine do you have?

It's unlikely that there's any code assuming that pointers are the same
size as ints, since ADL has been ported to IBM-PC's running MS-DOS.
It's quite likely that there's code assuming that "NULL is all bits zero",
although I haven't run across it.  Could you e-mail me a copy of your
modified adltypes.h as well as a brief description of your machine
(i.e. byte order in long ints; what does NULL look like; number of bytes
in an int, a short, a long, a pointer;  what is the operating system
like (BSD, SysV))?  ADL itself has been ported to the following computers
and operating systems (as far as I know):

	Hewlett-Packard 9000, series 300 (68020 based workstation
			running a BSD, SysV hybrid)
	Hewlett-Packard 9000, series 800 (RISC-based computer
			running a BSD, SysV hybrid)
	VAX 11/780 running 4.2 BSD
	Microvax II running VMS
	IBM PC running MS-DOS 2.0
	Amiga personal computer (68000 based PC running a custom O/S)

> My biggest complaint with adlrun is that is does not use the termcap
> capability CS to split the screen in two pieces. The top line is used
> to hold room name and score but it is scrolled off the screen every
> time when the rest of the screen scrolls. I had too much to do
> to fix this. Maybe you could look into it.

Well, that's sort of a feature, since not all terminals have memory lock.
I suppose that I could have some flag for if the terminal have memory
lock and use it if so.  ADL version 4.0, anybody?

> > Likewise.  If you've written a dungeon with ADL, I'd be *GREATLY* interested
> > in seeing a copy of it.  Post it to comp.sources.games, or e-mail me
> > a copy.
> Hm. I have written a small test dungeon with 5 rooms, 2 problems and
> 1 dragon to try some of the features of adl. I am planning with some
> friends to write a big dungeon. Maybe on the AMIGA, maybe on UNIX, well
> we will post if anything comes out of it.

That would be interesting to see!

> Now something different:
> ADL is really quite nice, I enjoyed playing aard (What the heck is
> an "aardvark" anyway ??), and I am really amazed that adlcomp and adlrun
> are so *small*.
> So here's a big hand for the authors.

(*blush*) Thank you.  Oh, an aardvark is a funny animal that eats ants.

> (Even if the programs have bugs. What big program doesn't ??:-)

Bugs?  What bugs?  :-)

>	Michael T. Greim
>	Universitaet des Saarlandes
>	FB 10 - Informatik (Dept. of CS)
>	Bau 36, Im Stadtwald 15
>	D-6600 Saarbruecken 11, West Germany

					Ross Cunniff
					Hewlett-Packard Colorado Language Lab
					...{ucbvax,hplabs}!hpda!cunniff
					cunniff%hpda at hplabs.ARPA