On the development of C

[Tony]

Yes.

You have someone defending or supporting you. Cool.

I do not "believe" that. I have implemented it in lcc-win.
This is a small difference, but crucial in my opinion.

Everytime you say "lcc-win", you sound like a marketeer.

The problem with C++ is that they have gone into the complexity
trip so far that this reaches absurd proportions.

It's your "job": make the better mousetrap. (I suggest that C-compatibility
is baggage for the most part).

The problem with C is

There is no ONE problem with C. Start anew. Hard for you because you know C.
You don't need C. You know about what you know. It's not about letting go,
it's about getting their grimy hands off of yours.

What I propose is not going into the extremes. That's why
I am attacked from all sides

You are not attacked from any side nor is that desireable. You live in a
world of aggression and have brainwashed into suppression and conformance.
Do you feel the need to conform to improve? Do you feel the need to war to
improve?

Tony

 

[Tony]

It's not 2010 yet. The committee has plenty of time to do whatever
it's going to do with gets().

Just the break that JN needs! Sink them before they get a chance to swim JN!

(Time is on your side).

Tony

 

[Tony]

Of course!

The future is very long...

They can debate that until 2020 or maybe 2030 who knows?

They could even wait until the clock turns around the 32 bits
in 2038...

I don't think you can do it. You're romantic. (Read, someone will make a
movie about you many years after you die. Starving artist?)

Tony

 

[Tony]

Have you ever tried to persuade any other compiler maintainers to adopt
your implementation? That would be the first step to getting them
standardised.

Thank you for stating the problem: "compiler maintainers". Time to stop that
oppression.

As we've argued many times, a team or individual is free to pick the
subset of the language they are happy with.

That's not true. You throw about the term "team", well, I know you. I ain't
"your team" bitch. Go **** someone else over (in which case I may choose to
war against you). Don't even get me started (you owe back taxes).

One team (of DSP programmers) I'm currently mentoring has done just that.
They have opted for C++

Noted: you are LIGHTWEIGHT "leader".

because they initially want RAII and function overloading.

Oh, but you asserted to be the HEAVYWEIGHT "Leader" huh... It's lookin bad
for you babe.

Once they are more familiar with the language understand the benefits of
other features, they will use more of them.

Wrong answer. Listen up. Anyone can go across any domains of knowledge and
politic their way into position. Personally, I have found nothing else but
this one key fact. Give GM money? Use your dick and walk the fence without
stuttering in speech?

Go **** yourself.

2009: Black man becomes president of irrelevancy. Gives money to Pharoah.

There we agree!

Are you even dummer? You don't start from political rust. You only recognize
steel.

You get "attacked" for refusing to consider alternatives.

He is not attacked. No one is attacking him. He WANTS to be "attacked" in
political effort to assert his "product".

Tony

 

[Walter Banks]

I do not "believe" that. I have implemented it in lcc-win.
This is a small difference, but crucial in my opinion.

The problem with C++ is that they have gone into the complexity
trip so far that this reaches absurd proportions.

The problem with C is that it has stangnated also into absurd
proportions like keeping "gets()" in the standard in 2010!

What I propose is not going into the extremes. That's why
I am attacked from all sides

Ranting about the standards and not seeking to change them
is a huge dis-service to lcc-win customers.

Standard changes were never meant to be easy, they
are meant to follow a process that debates the merits of
changes. The change process is both formal and
transparent.

Regards,

 

[jacob navia]

When 31 bits are set, ctime() will give us "Tue Jan 19 03:14:07 2038".

Then, one second later it will give time zero. This is called "wrapping
around" and this is what I was speaking about.

 

[Harald van Dijk]

Then, one second later it will give time zero. This is called "wrapping
around" and this is what I was speaking about.

So time_t on your system is a 31-bit unsigned integer type?

 

[James Kuyper]

Dik T. Winter wrote:
....

Everybody is apparently missing the point. The part with double quotes is
from an earlier article by Nick Keighley that Tony copied. I do not know
any newsreader that will put >'s automatically when you copy and paste from
any other source than the article you are replying to.

In Thunderbird, if you select Edit/Paste as Quotation from the main
menu, the '>'s are in fact automatically added.

 

[Keith Thompson]

Then, one second later it will give time zero. This is called "wrapping
around" and this is what I was speaking about.

Assuming the typical Unix representation for time_t (a 32-bit signed
integer representing the number of seconds since 1970-01-01 00:00:00
UTC), the last representable time is Tue Jan 19 03:14:07 2038. One
second after, the time would be displayed as Fri Dec 13 20:45:52 1901
-- assuming that signed overflow behavior is the typical
2's-complement wraparound. (All times are UTC.)

Of course, neither the representation for time_t nor the behavior of
signed integers on wraparound is specified by the standard.

<SEMI-OT>
Some have suggested that the solution is to switch from signed to
unsigned. That would let us represent times up to Sun Feb 7 06:28:15
2106. But the cost would be that we could no longer represent times
before 1970. We have nearly 3 decades to transition from 32-bit
time_t to 64-bit time_t. I've already used systems where time_t is 64
bits, and C requires support for 64-bit integer types anyway.
Considering how much things have changed in the last 3 decades, I
think that stealing the sign bit to buy an extra 68 years would be
short-sighted.
</SEMI-OT>

 

[Dik T. Winter]

>
> Assuming the typical Unix representation for time_t (a 32-bit signed
> integer representing the number of seconds since 1970-01-01 00:00:00
> UTC), the last representable time is Tue Jan 19 03:14:07 2038. One
> second after, the time would be displayed as Fri Dec 13 20:45:52 1901
> -- assuming that signed overflow behavior is the typical
> 2's-complement wraparound. (All times are UTC.)

Linux.x86_64/nissiros.cst/ksh: cat time.c
#include <stdio.h>
#include <time.h>

int main(void) {
printf("%d\n", sizeof(time_t));
}
Linux.x86_64/nissiros.cst/ksh: a.out
8
Linux.x86_64/nissiros.cst/ksh:

So, when do have to worry on that system?

 

[Dik T. Winter]

> James Kuyper said:
>
>
> It's the same in KNode.

Well, not in the newsreader I use. Moreover it is pretty difficult; when
I want to include text from an earlier message I just go to the news
spool directory, find the relevant article and copy the text in my editor.

 

[Stephen Sprunk]

Linux.x86_64/nissiros.cst/ksh: cat time.c
#include <stdio.h>
#include <time.h>

int main(void) {
printf("%d\n", sizeof(time_t));
}

Technically, that should be "%zu\n".

Linux.x86_64/nissiros.cst/ksh: a.out
8

Not surprising; most "64-bit" systems use a 64-bit time_t these days.
There's even some discussion of switching 32-bit systems to that as
well, but AFAIK no mainstream ones have yet -- but there's still 29
years left to get it done.

Linux.x86_64/nissiros.cst/ksh:

So, when do have to worry on that system?

Just shy of 300 billion years from now, by which time we'll certainly
using a different timekeeping system or at least epoch. Then again,
that kind of thinking is how we got into the Y2k mess...

S

 

[Guest]

Thank you for stating the problem: "compiler maintainers". Time to stop that
oppression.


That's not true. You throw about the term "team", well, I know you. I ain't
"your team" bitch.

<snip drivel>

don't post when drunk

 

[Bartc]

<snip drivel>

don't post when drunk

He seems to have been getting drunk with Larry Gates.

 

[James Kuyper]

Well, not in the newsreader I use. Moreover it is pretty difficult; when
I want to include text from an earlier message I just go to the news
spool directory, find the relevant article and copy the text in my editor.

There's another approach that I've also used, particularly if I only
have Google Groups available. If I'm combining text from two or more
different e-mail or usenet messages, I'll select Reply for each message,
popping up a separate window for each message. Then I choose one of the
reply windows to be the actual reply, and copy and paste from the other
reply windows into that one, with the '>' attached. I then discard the
other reply windows without hitting 'Send'.

Is this approach feasible with your newsreader?

 

[James Kuyper]

Technically, that should be "%zu\n".


Not surprising; most "64-bit" systems use a 64-bit time_t these days.
There's even some discussion of switching 32-bit systems to that as
well, but AFAIK no mainstream ones have yet -- but there's still 29
years left to get it done.

Not really. People do often work with future times; consider, for
instance, printing out a payment schedule for a 30 year fixed-rate
mortgage. The time to start the transition is now, while the problems
are still relatively infrequent.

....

Just shy of 300 billion years from now, by which time we'll certainly
using a different timekeeping system or at least epoch. Then again,
that kind of thinking is how we got into the Y2k mess...

The Y10K problem will come first ... unless there's a change of epoch
before then, which seems pretty likely even on that "short" time scale.

 

[Dik T. Winter]

> Dik T. Winter wrote: ....
>
> There's another approach that I've also used, particularly if I only
> have Google Groups available. If I'm combining text from two or more
> different e-mail or usenet messages, I'll select Reply for each message,
> popping up a separate window for each message. Then I choose one of the
> reply windows to be the actual reply, and copy and paste from the other
> reply windows into that one, with the '>' attached. I then discard the
> other reply windows without hitting 'Send'.
>
> Is this approach feasible with your newsreader?

No. It does not pop up windows, it is simply in a single text based terminal
window.

 

[Tony]

"don't post when drunk"

At least I'm not drinking and driving!

 

[lovecreatesbeauty]

jacob navia said:


Simple, fast, portable, stable. C99 breaks three of these (although
hopefully two of them can eventually be fixed), but all four
advantages do still apply to C90.

Is four the exact number?

C is maybe also powerful, flexible, small ...

I want to know the answers to the following questions for a long time.
Can they be C FAQs?

1. Where do these advantages of C show in the language itself?

2. Why is C better than another languages (eg. D++) in all the
aspects?

 

[lovecreatesbeauty]

jacob navia said:


Simple, fast, portable, stable. C99 breaks three of these (although
hopefully two of them can eventually be fixed), but all four
advantages do still apply to C90.

You say eg. fast, but how fast is it? Why is it so fast? Is there
another language faster than C?

And Do it have disadvantages?