«

Developed by Denis richi.............

Op Sun, 21 Dec 2008 08:02:02 -0800 (PST) schreef :

> Developed by Denis richi.............

Your spelling of his name is not even middle level.
--
Coos

"Jujitsu Lizard" <> writes:
[...]
> If you can't consistently write correct code, you SHOULD be using another
> language.

Do you know anyone who *can* consistently write correct code?
I certainly can't; I make mistakes.

--
Keith Thompson (The_Other_Keith) kst- <http://www.ghoti.net/~kst>
Nokia
"We must do something. This is something. Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"

Jujitsu Lizard wrote:
>
> If you can't consistently write correct code, you SHOULD be using another
> language.
>
Which self correcting language would you recommend?

--
Ian Collins

Jujitsu Lizard wrote:
> If you can't consistently write correct code, you SHOULD be using another
> language.

....which implies that C should not be used by anyone.


August

Richard Heathfield wrote:
> August Karlstrom said:
>
>> Jujitsu Lizard wrote:
>>> If you can't consistently write correct code, you SHOULD be using
>>> another language.
>> ...which implies that C should not be used by anyone.
>
> Since anyone who can't consistently write correct code should not be
> using /any/ programming language, we can draw the further
> conclusion that nobody should be programming *at all*. Whilst I
> don't actually agree with this rather extreme position, it would
> certainly be tempting to use Sturgeon's Law to sack 90% of the
> programmers out there, and then sit back and wait for the fabulous
> stuff that the remaining 10% will eventually produce.

The remaining 10% will be swamped wasting their time and skills taking
care of the routine tasks that the other 90% used to take care of. They
won't have much time to spare to do anything "fabulous" (they will have
some time to spare).

On Dec 23, 9:02*am, "Jujitsu Lizard" <jujitsu.liz...@gmail.com> wrote:

> In C you can do marvelous things like:
>
> a)Continuing to use a block of memory after it has been deallocated.
>
> b)Returning a pointer to an automatic variable, which is then used after the
> stack frame has expired.
>
> c)Allowing an integer array index to roll over backwards, setting up to
> corrupt (for example), unrelated memory or the stack frame.


All of those can be caught with:

gcc -Wall -fmudflap -lmudflap

Ian Collins <ian-> writes:
> Jujitsu Lizard wrote:
>>
>> If you can't consistently write correct code, you SHOULD be using another
>> language.

Note to Keith - 'consistently' doesn't mean 'always'.
I'm sure Keith and Ian et al. do consistently write
correct code. I'd like to think I do too. But no-one
is perfect.

> Which self correcting language would you recommend?

LOGO. You might not get what you initially intended, but at
least you can *always* call it "art".

Phil
--
I tried the Vista speech recognition by running the tutorial. I was
amazed, it was awesome, recognised every word I said. Then I said the
wrong word ... and it typed the right one. It was actually just
detecting a sound and printing the expected word! -- pbhj on /.

On Dec 22, 2:23*pm, "Jujitsu Lizard" <jujitsu.liz...@gmail.com> wrote:
> That would be Dennis Ritchie:
>
> http://en.wikipedia.org/wiki/Dennis_Ritchie
>
> C in my opinion is "just right".
>
> The reason is that if you go too high-level, there are performance
> implications.
>
> C is, plus or minus a little, a structured assembler.
>
> C gives you enough rope to hang yourself (no array bounds checking, etc.),
> but on the other hand you can write those things yourself if you so desire
> and you really won't take a performance hit over a "high-level" lanugage, as
> your compiled code will do about the same things that the compiled
> "high-level" language would do.
>
> If you can't consistently write correct code, you SHOULD be using another
> language.
>
> --
> Jujitsu Lizard (jujitsu.liz...@gmail.com)

I'm a big fan of C + datastructure tools. I find C + datadraw to be
higher performance than C alone, while more productive for EDA
applications than C++. The problems with higher level languages is
they make a lot of assumptions about the target application. They
assume we can afford some loss in performance, and that we mostly need
static inheritance rather than the ability to dynamically extend
objects that already exist in a database. Both assumptions are wrong
for most EDA programs. They also do simple one-class implementation
inheritance which isn't even capable of doing a decent performance
linked-list, because it optimally involves adding functionality to two
classes rather than just one. Graphs, the heart of most EDA
applications, are a PITA to describe generically in most high-level
languages, though Java does a better job than C++, IMO. Then there's
garbage collection. Modern languages assume they can help us by
eliminating the need to call "delete" on an object, but we have to
remove it from all database relationships anyway. They saved me one
line of code, but I still have to write my own destructors from
scratch.

There's a reason most new EDA applications are still written in C, and
it's not that programmers are stupid.

writes:
[...]
> I'm a big fan of C + datastructure tools. I find C + datadraw to be
> higher performance than C alone, while more productive for EDA
> applications than C++.
[...]
>
> There's a reason most new EDA applications are still written in C, and
> it's not that programmers are stupid.

That might make more sense if you told us what "EDA" means.

--
Keith Thompson (The_Other_Keith) kst- <http://www.ghoti.net/~kst>
Nokia
"We must do something. This is something. Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"