From: Herbert Kleebauer on
Phil Carmody wrote:
> Hans-Peter Diettrich <DrDiettrich1(a)aol.com> writes:
> > Herbert Kleebauer wrote:

> > The assignment is the last operation in the whole expression (lowest
> > precedence), so that the --j will occur
>
> just after Scott Nudds flies out of your nose and hits your monitor.
>
> > most probably just before the
> > assignment, after evaluation of the RHS expression, and during
> > evaluation of the LHS.
>
> Please take your off-topic

How can a content related reply be off-topic? This is only
possible if the original posting also is off-topic, and that
posting was from ala.

> incorrect

You should never use such accusation without providing a proof.

> nonsense

Even if something is incorrect, why does it that make a nonsense?

> to comp.lang.c.

Most probably he is from comp.lang.c

> When you've got the flaming you deserve, hopefully you will not
> post it anywhere else again.

Seems you have got many flaming from there. But I suppose this
is because of your inability to make logical arguments and not
because of the superior membership of that group.
From: jukka on
On Jul 21, 7:19 pm, Herbert Kleebauer <k...(a)unibwm.de> wrote:
> "T.M. Sommers" wrote:
> > Rod Pemberton wrote:
> > >>>case 0x05: {la[--j] = la[j] / la[j+1]; break;}
> > > I.e., assuming LALR(1) and applying precedence, the above is:
> > > case 0x05:
> > > {
> > > --j;
> > > la[j] = la[j] / la[j+1];
> > > break;
> > > }
>
> > > It appears that DJGPP (GCC) works properly while OpenWatcom v1.3 fails. I'd
> > > guess that most K&R style compilers work.
>
> > The origial line invokes undefined behavior; anything the
> > compiler does is "proper".
>
> Then you have a very bizarre definition of "proper".

The original code does invoke undefined behaviour, there is nothing
"proper" about that even if it is relying on some compiler specific
detail. Write correct C and move on.

From: Evenbit on
On Jul 20, 4:59 am, Herbert Kleebauer <k...(a)unibwm.de> wrote:
> Evenbit wrote:
> > On Jul 19, 6:59 am, Herbert Kleebauer <k...(a)unibwm.de> wrote:
> > > crashed?????
>
> > I got that to work and ./lindela runs but when I try to feed it an
> > file from /EXAMPLE/LINUX it bombs with what I assume to be some German
> > expletives.
>
> I ignore warnings of a C compiler but not a crash of my programs.
> Can you give me a little bit more information which source you used
> and which error message you got.

Well, I have tried a number of variations, even copied the source to
the same directory:

/EXAMPLE/LINUX/PAR_ENV.MAC
PAR_ENV
par_env
DOS2UNIX.MAC
DOS2UNIX
dos2unix.mac
dos2unix
etc...

Always I get:
Quelltextdatei konnte nicht geoeffnet werden in Zeile 1

>
> > Is this thing PD, GPL, or what?
>
> It's an ASCII string, much to short and ugly written to deserve any license.
> All I wanted was a debugger/monitor program for DOS and because I only
> found this awful Intel syntax assemblers I first had to write an
> assembler to be able to write the debugger/disassembler. So there was no
> time to make a well designed program, it's only a quick hack written
> on an Atari ST (that's the reason why you still have to enter the
> source file name after you start the program because the Atari ST didn't
> have a command line interface where you could enter parameters. But
> in DOS/Windows you can use a batch to start it: echo %1|windela ).

Then I deduce that Daniela is a Public Domain example of how to write
a "quick hack" assembler in non-conforming C as a student's "first-
written-program" exercise.

>
> > Do you mind if I tuck it somewhere
>
> I don't think it's of interest for anyone, but you can do with it what ever
> you like.

Thanks.

>
> > "safe" on the 'net so it'd be available long after your server "powers
> > down" for the last time?
>
> Hey, "mistress" is up and online now for 13 years (maybe one of the
> oldest, still running Linux installations) and I hope she will do it
> a few more years.

Well, if the "powers-that-be" can make Beth "disappear"... we just
cannot count on anything these days...

Nathan.

From: Herbert Kleebauer on
Evenbit wrote:

> Well, I have tried a number of variations, even copied the source to
> the same directory:
>
> /EXAMPLE/LINUX/PAR_ENV.MAC
> PAR_ENV
> par_env
> DOS2UNIX.MAC
> DOS2UNIX
> dos2unix.mac
> dos2unix
> etc...

It doesn't matter where lindela is located as long as it is
somewhere in the $PATH (most Linux add ~/bin to the path).
If you start lindela by typing lindela (or ./lindela if
it's in the current directory) at the command prompt,
you should get:

*****************************************************
* *
* DDDD AAA N N III EEEEE L AAA *
* D D A A NN N I E L A A *
* D D AAAAA N N N I EEEE L AAAAA *
* D D A A N NN I E L A A *
* DDDD A A N N III EEEEE LLLLL A A *
* *
* Der *
* ANdere *
* IntEL 486 Version 0.36 *
* Assembler 05.02.07 *
* *
* *
*****************************************************



Filename (ohne .mac):


Now enter the source file name without the extension .mac,
for example: dos2unix

(The extension for the source files must be lower case .mac).



> Always I get:
> Quelltextdatei konnte nicht geoeffnet werden in Zeile 1

(source file couldn't be opened in line 1)


Then the file dos2unix.mac isn't in your current directory.
From: Rod Pemberton on

"Phil Carmody" <thefatphil_demunged(a)yahoo.co.uk> wrote in message
news:87ejj1ubnn.fsf(a)nonospaz.fatphil.org...
> "Rod Pemberton" <do_not_have(a)nowhere.cmm> writes:
> > "santosh" <santosh.k83(a)gmail.com> wrote:
> > > case 0x05:
> > > {
> > > la[j-1] = la[j] / la[j+1];
> > > j--;
> > > break;
> > > }
> ...
>
> > I.e., assuming LALR(1) and applying precedence, the above is:
>
> Parsing/precedence and order of evaluation are utterly utterly utterly
> unrelated.
>
> And by "the above" do you mean "the original", not "the above", as
> the above is certainly nothing like the below?
>
> > case 0x05:
> > {
> > --j;
> > la[j] = la[j] / la[j+1];
> > break;
> > }
> >
> > It appears that DJGPP (GCC) works properly
>
> Nope. There is no properly.
>

I like how you just drop "I.e., assuming LALR(1) and applying precedence,
the above is:" right out the conversation. It appears that you can't carry
a thought to the remainder of the post... It reminds me of other
conversations I've had with you where you just randomly become hostile and
ignore all other inputs,.er., truths.

Snipping the historical 1-7 as both you an T.M. Sommers did, also distorts
my entire post... To which, it appears that five or six others who never
read it responded...

> Take this to c.l.c and have a bunfight, which you will lose, there.

I'm completely aware that this is the historical context of how things are
parsed in C and not in compliance with the standards. There is no reason
for me to post to c.l.c. If you hadn't snipped 1-7, maybe you'd have read
5: "5) Although not required by the standards...".

Phil, your posts seem to become hostile, and your comprehension low on
weekends. Lay off the sauce...

> Stop trying to confuse a.l.a readers.

The point was to help Santosh, who while looking at Keeblauer's extremely
simple C code, came to the wrong conclusion about it's meaning.

If he had learned K&R C from an LALR(1) C compiler, he'd know the correct
interpretation. He didn't. In part, this is due to c.l.c. regulars
insisting that only ISO abstracted C is correct. Many versions of C existed
prior to the first ANSI C. Ritchie's "DMR" recursive descent C. Johnson's
YACC based LALR(1) C: PCC. K&R C. etc.


Rod Pemberton

First  |  Prev  |  Next  |  Last
Pages: 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Prev: masm linking from console
Next: NASM HelloWorld - DOS