From: hansoft AT visitweb DOT com (Hans 'the Beez' Bezemer)
Newsgroups: comp.os.msdos.djgpp,comp.lang.c
Subject: Re: Scripting language library
Date: Mon, 15 Dec 1997 22:35:53 GMT
Organization: HanSoft & Partners
Lines: 61
Message-ID: <3495ac96.1274644@news.xs4all.nl>
References: <348e5375 DOT 789931 AT news DOT xs4all DOT nl> <34910cce DOT 1337533 AT news DOT xs4all DOT nl> <66sask$2q$1 AT home DOT edu DOT stockholm DOT se> <3494eaf4 DOT 492512 AT news DOT xs4all DOT nl> <67453a$t1v$1 AT home DOT edu DOT stockholm DOT se>
Reply-To: hansoft AT visitweb DOT com
NNTP-Posting-Host: ztm06-12.dial.xs4all.nl
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

ulric AT evelin DOT edu DOT stockholm DOT se (Ulric Eriksson) wrote:

>The problem is that your solution (make the newbie learn Hans' Version
>Of C Which Looks Like Pascal) is worse than the problem (make the newbie
>learn C).
Well, that is a point for debate. I'm sure I've prevented many errors by using
these constructions (together with a standard and a list of pitfalls). First
make it work (let 'em write acceptable programs) then improve it (teach 'em
just another way to write the same thing).

>In my case, all you accomplished was to make me suspicious. I'm sure
>that's not what you had intended.
Oh no, certainly not. But never judge a book by its cover. First impressions
are usually wrong due to a lack of information (contrary to popular belief).
And RTFM.

>>You also counted the examples that are inside the comments of each and every
>>function/program. If main() doesn't return anything it is allowed IMHO, but
>>maybe somebody else has more detailled information on this. As far as 4th.c is
>>concerned, you are right. That is an error. Thank you for pointing this out to
>>me.
>void main is indeed permitted inside comments, but I still recommend
>against it, for reasons I am sure you can guess.
Sure I can. I guess that I'm the victim of these pre-ANSI compilers (following
the draft) and manuals. Man is never too old to learn. It will be fixed in the
next release.

>>A compiler is hard enough to understand as is. If EasyC makes that easier, why
>>not. If that is a reason for you to discard the package, I'll be happy to put
>>sed to work and give you a "true" C version. And of course, I'm always
>>interested in comments that can improve the quality of the package.
>It is not enough of a reason to discard the package, but it makes
>it more difficult to evaluate. For another case of warped C,
>take a look at George Carrette's Scheme interpreter SIOD.
>SIOD is written in C formatted like LISP, making it hard to read
>for a C programmer.
C'mon. There is a 1:1 relation between the keywords and the C equivalents. And
are there really good C programmers that can only write C? I'm sure you can
program Basic, Assembler, Pascal and a handful of other languages. I can
understand it "distorts" your C view (making it look "ugly"), but you can read
it. C and LISP is quite another story. Not the same family of languages. But
both Pascal and C have their roots in Algol.

>If you do clean up your code, I promise to stand up and cheer. You
>will have done your users and yourself a favour.
Yeah, that's the trouble with these issues. If you do it one way, one stands up
and cheers and another sits down and gets grumpy. Note that you don't have to
use EasyC to maintain it. You can intermix both. Not too neat, I agree, but
there is no real technical problem there. And once compiled to a library you
don't have to look at the source code ever again.

Let me put it this way: 4tH has a small, but dedicated following. If they want
me to use the usual syntax, I'll follow because the customer is always right.
As a matter of fact, you're the first one that complains.

Hans
================
"First make it work, then improve it."
Visit our website! http://visitweb.com/hansoft

*** Home of the 4tH compiler! ***