Re: A sample RosAsm macro (actually a whole set)

From: The Half A Wannabee ("The)
Date: 02/12/04 

Date: Thu, 12 Feb 2004 06:58:34 +0100

"Randall Hyde" <randyhyde@earthlink.net> wrote in message
news:3CCWb.763$tL3.468@newsread1.news.pas.earthlink.net...
>
> "The Half A Wannabee" <ShakainZulu_AT(Nightwish - She Is My Sin
> (Live).mp3)_Hotmail.com> wrote in message
> news:402aeae1$1@news.broadpark.no...
> >
> > The reason I answered, the point I was making :
> > ______> > Its _not_ about verbosity in typing at all to
me,_______________
>
> That statement is in contrast to previous posts you've made, but okay.

Not if you can read. Okey, maybe I cant write. But the point was : I have no
problem with writing long identifyers. I have however a problem with
creating a new syntax, that :

Doesnt add/convei anything more than the previous syntax, except for
verbosity. Its a bit snobbish of me perhaps. Why shouldnt people create
unneeded syntax if they want to. Please do, but dont say its inventive or
smart. Like you do, and say that it sets the language apart from any other
language in a useful way.

> Study programming language design.

I like to design my own solutions. This is an assembler forum Randy. Not
McDonalds.

> Study typography.
What is that ? You are so funny ....lol.

> Study software engineering and program layout.
You are a super geek, you know that ? I bet you need a PDF to wank off.

> Alas, it's a bit too complicated a subject to distill into a single
> post in a form that a teenager with no real-world programming
> experience can appreciate, but there are very good reasons for
> adding those extra characters.

Okey. Above is one of the reasons I am on your neck Randall. You do it all
the time. You say you cant explain it, and the reason is that I wouldnt
appreciate it. Maybe true, but I think the reason is, that you have no real
substains to what you would say. And _thats_ the reason you arent saying it.

>But just to give you one hint,
> when a statement becomes so long it's better to split it across multiple
> lines, those parentheses come in *very* handy for keeping the
> operands grouped.

So, you use them to recover from you own mess ?

> No doubt, you'd argue that you should only use
> them at that point, not when it fits on a single line.

I would not agree! ( Its by design Randall. The world is deterministic. But
so complex that it seem sometimes that people are not... )

>However,
> *consistency*
> is *very* important in a programming language.

Yes ! DUH. Of Course. But not in this case Randall. Not in this case at all.
Consistency in this case is wasteful and on thin ice with regards to
anything but personal preference. Its not based in reason. Theres no need
for it, and it offers no benefits. Why would you want to splitt assembler
syntax on a line, and holding it together with a paranthesis. Its an
"invention" that there is simply no use for. Not that thats important, but
you take it to the other extreme. you say its a feature.And argue its a good
one. Its like saying blue color is a good feature of a dishwasher. Only that
with the diffrence that blue color is not in you way.

>That's why you put
> parentheses around single parameters/operands even though, syntactically,
> they might not be required.

Thats was ALL I intended to say initially.

>That's why you put parentheses around
> instruction operands as well as subroutine parameter lists and macro
> argument lists: to keep it all consistent. Consistent not only with the
> assembly language, but consistent with what *professional* programmers
> (as opposed to teenage hackers) expect from *any* programming
> language.

Lol. So "professional" programmers expect _what_ ? Needless syntax or
consistency ? See how weak you argument is. If I am a teenager, then you
come out looking like a puppy. Talking to adult people about consistency !
While you keep pouring out versions for a text converter that you will
eventually dump, and re"create" from scratch and call it 2.0 While you are
allready dreaming about a 3.0. And according to popular myth, syntax is
broken between minor updates, and this you have the nerve to call consistent
? All the while your text converter is just a front end for other assemblers
? Where is the consistency ? The only really consistent about you is your
plug. And the way you adress newbies :

"just take a look at Webster, in the standardlibrary."

> > if D$eax does the job, then DWORD PTR [eax] is not needed. And this
makes
> > the change a logical one.
>
> if .!#%$ does the job, then why bother with D$eax?

Is this how you teach your students ? Did it ever float by at one time that
.!#%$ will not do the job ? Where is eax ? And what kind of silly argument
is this ?

> Does the phrase "readable vs. 'explosion in an alphabet soup factory'"
mean
> anything to you?

You just read that joke somewhere didnt you....

>You may convince *yourself* that D$eax is easier to read.
> Most people would probably not agree.

 Most (whom you dont know) people would _probably_ not agree ? Is this the
method of the science you use to empower you text converter ?

> > In fact all that would be need was $eax, since DWORD is assumed
"default"
> > declaration type.
>
> Uh, no it doesn't. EAX is the pointer to the object, not the value of the
> object.
> You need to learn a little more assembly language on top of programming
> language design.

Sorry, but I think you got it wrong here too. Lol. I have 4-5 weeks of
experience ..... But when I say D$eax I mean the DWORD/INTEGER/32BIT/LONGINT
at the memory location pointed to by eax ? And when I say B$eax I mean the
byte/char/8bit at the location pointed to by eax ? So since DWORDS are
considered default in RosAsm as in the declaration of 2 identical DWORD
sized data below, then one could expect the same to apply to D$eax. $eax
could implisitly mean D$eax, but for other, non machine sized variables one
would have to spesify. As the machine sized variables are most common, and
that they are usually faster than word and bytes, it wouldnt hurt.

> > [MySmallBuffer: 0]
> > [MySmallBuffer: D$ 0] <----------------- ** those
are the same. **
>
> Why? What makes the first one a dword vs. word, byte, or any other size?

By convention, agreement, between the language developer and the user. For
Dwords no need to spesify D$ (the type) as it is implisitt. W$ for a word or
B$ for a byte, must be spesified.

>
> And why is the second one somehow more readable than:
>
> MySmallBuffer db 0
> or
> MySmallBuffer:byte := 0;
>
> ???

I never said anything about that at all. I think the two last examples are
equally clear. In fact the last one is a bit clearer. But in RosAsm the
first one would be more consistent with the rest.

> Again, typing fewer characters does not make something more
> readable. e.g.,

No, simple but not too simple. I cannot see you have posed any good
arguments in this post for dec ( ebx ) which btw was the main reason for the
original post.

>
> agn tpng fwr chrctrs ds nt mk smthng mr rdbl

Never said that Randall. Wasnt my point at all. But I guess, even on 3. time
I say it you still dont take it.

>
> Cheers,
> Randy Hyde
>
>

Relevant Pages

  • Re: Learning a language
    ... >The reason to reject your reason is simple: ... "That's just how the language works?" ... >But if the design were consistent, the users would find it much easier ... Eastern Iowa Esperanto League, the Esperanto League for North America or even ...
    (sci.lang)
  • Re: primitive recursive: obsolete?
    ... all true sentences of arithmetic in the language of PA. ... We know the set is consistent in the sense that we PROVE that it ... Then that depends on a given axiomatization. ... theorem of ThN" makes no sense. ...
    (sci.logic)
  • Re: SQL
    ... >> Dmitry A. Kazakov wrote: ... but OO has no consistent "right way". ... Ideally the query language would be extendable if a DBA etc. wanted to ...
    (comp.object)
  • Re: FOL and AST
    ... about the cardinality of the language. ... completeness, not soundness -- specifically, for proving that every ... consistent set S of sentences of a given language L has a model (i.e., ... In a typical proof using Henkin's method, the cardinality is not ...
    (sci.logic)
  • Re: FOL and AST
    ... When I asked about which version of completeness elsewhere in the ... about the cardinality of the language. ... consistent set S of sentences of a given language L has a model (i.e., ...
    (sci.logic)