Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: $Revision: 1.6.2.16 $; site ada-uts.UUCP
Path: utzoo!linus!decvax!bellcore!petrus!scherzo!allegra!mit-eddie!think!ada-uts!richw
From: richw@ada-uts.UUCP
Newsgroups: net.lang.c
Subject: Re: Recoding Lisp programs in C
Message-ID: <10200013@ada-uts.UUCP>
Date: Mon, 7-Oct-85 12:58:00 EDT
Article-I.D.: ada-uts.10200013
Posted: Mon Oct  7 12:58:00 1985
Date-Received: Fri, 11-Oct-85 07:56:21 EDT
References: <324@bcsaic.UUCP>
Lines: 26
Nf-ID: #R:bcsaic:-32400:ada-uts:10200013:000:1108
Nf-From: ada-uts!richw    Oct  7 12:58:00 1985


>> ***** ada-uts:net.lang.c / brl-tgr!gwyn /  9:39 pm  Oct  5, 1985
>> If the code needed to be done in LISP in the first place,
>> then any C translation is going to have to provide garbage
>> collection anyway.

What about having the C version be conscientious about using "malloc"
AND "free"?  Though many people I know feel that dynamic storage
allocation is a "detail" that programmers need never worry much about,
I wonder if making sure to "free" anything you "malloc" is all that
hard.

I wrote my own slightly modified versions of "malloc" and "free" which
remember how many allocated blocks haven't been freed, and then prints
out an error message at the end of the program (I also modified "exit"
to do this) if all allocated blocks haven't been freed.  Unfortunately,
I haven't had time to use these versions in recent work.

Has anyone out there REALLY used "free"?  I'm very curious about this...

>> Personally I think AI has been oversold.

I'd like to shake your hand.  I have my own opinions about AI that I
won't (and shouldn't) flame about, but KEEP THAT DISRESPECT FLOWING !!!

-- Rich