Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!purdue!decwrl!ucbvax!TECHNION.BITNET!PHR00JG
From: PHR00JG@TECHNION.BITNET ("Jacques J. Goldberg")
Newsgroups: comp.os.cpm
Subject: Z80 Pascal Compilers and recursion.
Message-ID: <8807060808.AA24572@ucbvax.Berkeley.EDU>
Date: 6 Jul 88 08:08:25 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 26
X-Unparsable-Date: Wed, 06 Jul 88 08:42:53 IST

Some confusing, misleading, notes have been posted about the simple recursive
example which I quoted only to show that Turbo-Pascal and Polydata Pascal were
behaving in the SAME incorrect way.

1. Pascal IS recursive and the example works perfectly, I repeat, with UNIX
on a VAX, with VAX-VMS, with the IBM and the Waterloo compiler on IBM-VM
systems, with Microsoft Pascal under MS-DOS, and with JRT Pascal on my Z80
system. It does NOT work correctly with Turbo-Pascal and Polydata Pascal under
MS-DOS but this has nothing to do with the language implementation, only with
its I/O interface when the unit is the keyboard.

2. Turbo Pascal and Polydata Pascal DO correctly implement recursion, with the
DEFAULT {A-} compiler option . The alternate option {A+}, under CP/M-80 ONLY,
contrarily to what Erik Seielstad writes, makes the code NOT recursive. Again,
the DEFAULT obeys Pascal RULES, while EXPLICITLY requesting {A+} ( A for
ABSOLUTE code) will assume you don't want recursion and thus save speed and
memory (so says the guide, at least...)

3. Then I will repeat my request: WHO CAN GIVE ME AN ADDRESS WHERE TO BUY
PREFERABLY Z80_Polydata_Pascal ,OR STILL ACCEPTABLE Z80_Turbo_Pascal?

                                            Jacques J. Goldberg
                                            phr00jg@technion.bitnet
                                               ^^
                                               ||
                                               ---zeroes