A good compiler

H

Harald van =?UTF-8?B?RMSzaw==?=

Chris said:
Where is "locally"?

As suggested by my headers, I live in the Netherlands. Two Dutch online
shops I've bought books from before, Bol and Selexyz if any other Dutch
folks are interested, offer the printed version jacob navia mentioned for
53 and 60 Euros respectively.
 
K

Keith Thompson

Chris Hills said:
You are mistaken.
The published C99 (+ published TC1 and TC2 ) is the standard (for what
good it does :-( .

N1124 is a document it has no standing whatsoever until it is
published. Then it becomes a different document. The problem is that
changes can and do happen right up to the point of it hitting the
presses. Some may only be typos but they can be critical I have seen
a document "issued" and then pulled again before actually being
published so some corrections can be made.

Quoting from the N docs because you don't want to spend the money on
an official document is no excuse (I know you have official docs,
Keith so that comment wasn't aimed at you)
[...]

N1124 is intended to incorporate C99, TC1, and TC2. For the purpose
of posting quotes here, it's good enough for me; I haven't heard of
any inconsistencies in the 2+ years since N1124 came out.

If you know of any inconsistencies, please let us know; otherwise, I
fail to see the point of your "You are mistaken" remark above.
There's nothing magical about an "official" standard that makes errors
impossible.

(Expect me to be a bit terse for the next few days, until this huge
bandage comes off my left hand.)
 
K

Keith Thompson

jacob navia said:
Some people just do not understand that not everything is free
and open source. The standards are sold for a small amount of
money, they are just not free.

I'd be a lot more impressed by that if the money went to the people
who did the actual work. There are language standards that are freely
available (Ada, for one), and IMHO those communities benefit
considerably. For one thing, they don't have these arguments.

And for my purposes, I find N1124 (free) more useful than the C
standard (for which I did pay).
 
K

Keith Thompson

jacob navia said:
Harald van Dijk wrote: [...]
Well, I'm quoting from the drafts because even though I would be
quite happy
to purchase a copy of the standard for the price ANSI charges for it, I am
not able to do so: payment by credit card is simply not an option for me
(that's assuming ANSI doesn't sell only to folks from the US; I'm not sure
about that), and all other methods I am aware of would cost me over 100
Euros extra. If you can tell me how I can obtain the official standard for
a reasonable price (which may be higher than ANSI's, but more than four
times as high is ridiculous), please let me know, I would be very
interested. If you cannot tell me how to obtain the official standard for a
reasonable price, I will continue to rely on the drafts. How you feel about
that is up to you; I don't and won't feel bad about it.

"The C Standard" Incorporating Technical Corrigendum No 1"
Wiley and Sons Ltd.
ISBN 0-470-84573-2
[...]

Alas, you can't grep dead trees, and you can't copy-and-paste any
passage more than once.

Are they going to publish a version with TC2?
 
C

CBFalconer

Keith said:
.... snip ...

And for my purposes, I find N1124 (free) more useful than the C
standard (for which I did pay).

And, for mine, N869 is most useful, because it is available in text
format, searchable by grep and the whole panoply of text utilities.
 
C

Chris Hills

Keith Thompson said:
I'd be a lot more impressed by that if the money went to the people
who did the actual work.

Why? We know what the deal is when we started. None of us get paid for
any standards work. In some areas you actually have to pay to join the
organisation that does the standardisation.

It is business.
There are language standards that are freely
available (Ada, for one), and IMHO those communities benefit
considerably. For one thing, they don't have these arguments.

No standards are free. Ada cost a LOT of money the difference was the US
Government footed the bill. Several millions of USD. I am sure that if
you are prepared to put up several millions of USD the C standard would
be free.

This is more easy to do now pdf. Is the norm rather than having to
physically print a book**.
And for my purposes, I find N1124 (free) more useful than the C
standard (for which I did pay).

Fair enough but if you want to quote chapter and verse use the standard
for the numbers.


*** The digital revolution has cost many jobs.
In the closest city to me there used to be 6 professional photographic
film houses. All bar one have gone. The 1 hour kiosks have been
replaced with digital ones but many still print at home.

With books almost all the manuals for non consumer items (except
computers) are now PDF, Many printers have lost work. Those that have
not moved to Digital are loosing work fast.

Type setting and printing of standards etc costs a lot of money for a
comparatively small market. I know I have seen he figures for some.

However even for pdf standards there is a cost in setting up the
document and proofing it etc. It all coast.

Like RedHat Linux the initial input is free the bit you pay for is
turning it into a polished product.
 
C

Chris Hills

Keith Thompson said:
jacob navia said:
Harald van D0 >[...]
Well, I'm quoting from the drafts because even though I would be
quite happy
to purchase a copy of the standard for the price ANSI charges for it, I am
not able to do so: payment by credit card is simply not an option for me
(that's assuming ANSI doesn't sell only to folks from the US; I'm not sure
about that), and all other methods I am aware of would cost me over 100
Euros extra. If you can tell me how I can obtain the official standard for
a reasonable price (which may be higher than ANSI's, but more than four
times as high is ridiculous), please let me know, I would be very
interested. If you cannot tell me how to obtain the official standard for a
reasonable price, I will continue to rely on the drafts. How you feel about
that is up to you; I don't and won't feel bad about it.

"The C Standard" Incorporating Technical Corrigendum No 1"
Wiley and Sons Ltd.
ISBN 0-470-84573-2
[...]

Alas, you can't grep dead trees, and you can't copy-and-paste any
passage more than once.

Are they going to publish a version with TC2?

I though we were going to get a C0* with TC1 and TC2 rolled in but it
hasn't happened. I think the problem is that C99 *still* has not been
taken up by the industry and they are looking where to go next.

There has been some discussion of dropping things from the next version.
It was suggested that anything not implemented by more than two
compilers could be dropped. However it was still very early stages of
a "what if" sort of conversation at the WG15 meeting in April.
 
C

Chris Hills

Keith Thompson said:
Chris Hills said:
You are mistaken.
The published C99 (+ published TC1 and TC2 ) is the standard (for what
good it does :-( .

N1124 is a document it has no standing whatsoever until it is
published. Then it becomes a different document. The problem is that
changes can and do happen right up to the point of it hitting the
presses. Some may only be typos but they can be critical I have seen
a document "issued" and then pulled again before actually being
published so some corrections can be made.

Quoting from the N docs because you don't want to spend the money on
an official document is no excuse (I know you have official docs,
Keith so that comment wasn't aimed at you)
[...]

N1124 is intended to incorporate C99, TC1, and TC2. For the purpose
of posting quotes here, it's good enough for me; I haven't heard of
any inconsistencies in the 2+ years since N1124 came out.

If you know of any inconsistencies, please let us know; otherwise, I
fail to see the point of your "You are mistaken" remark above.
There's nothing magical about an "official" standard that makes errors
impossible.

True but it is the Official standard.

If you are prepared to quote from non official sources you can't
complain of people discussing "c-like" languages here. You can't have
it both ways.

Incidentally if you (an the little band of net nannies) stopped all this
OT nonsense there would be a lot better signal to noise on this NG even
with the threads you consider OT.

More to the point you would be able to explain standard C to the
heretics who thing that MS VC or Gcc are standard. Besides as you have
just pointed out even the standard has mistakes in it... so "standard C"
is not perfect.
(Expect me to be a bit terse for the next few days, until this huge
bandage comes off my left hand.)

Best wishes.... I trust it is nothing life threatening and will heal
fully
 
J

jacob navia

Chris said:
Why? We know what the deal is when we started. None of us get paid for
any standards work. In some areas you actually have to pay to join the
organisation that does the standardisation.

The French standards committee (AFNOR) asks for
10 000 Euros to start doing anything.

Then you have to pay all the travel expenses.

I wanted to propose some of my extensions but this is greatly beyond
what I can pay...

It is business.

Let's say it is an expensive business :)
 
K

Keith Thompson

Chris Hills said:
True but it is the Official standard.
Yeah...

If you are prepared to quote from non official sources you can't
complain of people discussing "c-like" languages here. You can't have
it both ways.

Nonsense. This newsgroup discusses the C language. N1124 is a good
description of (a version of) that language. If quotations from K&R
(both editions) are topical here, surely quotations from N1124 are as
well. And if anyone points out an inconsistency betwen an N1124
passage that I quote and the corresponding text in the Offical
C99+TC1+TC2, I'll be glad to acknowledge it (after I verify it; I have
copies of all 4 documents).

This is not comp.std.c, which emphasizes the C standard *document*
rather than the language it defines. We quote from the standard and
other sources to support points about the language.

Some people here quote from N869, which is (as you know) a pre-C99
draft. If there are significant differences between N869 and C99,
they'll most likely be noticed and pointed out. The two are mostly
consistent; cases where they differ are significant, but not the end
of the world.

None of this implies that compiler-specifi extensions such as qfloat,
statement expressions, operator overloading, and arithmetic on void
pointers are topical.

You've complained (in some cases rightly) about discussions of gcc
here. My impression is that a lot of those discussions are about how
to make gcc behave as a conforming C compiler. I see fewer
discussions about gcc-specific extensions here than about
lcc-win32-specific extensions, and both are equally off-topic.
(Proposals to standardize such extensions are topical in comp.std.c.)
Incidentally if you (an the little band of net nannies) stopped all
this OT nonsense there would be a lot better signal to noise on this
NG even with the threads you consider OT.

Obviously I disagree. But if you'll present a coherent statement
about what you think should be considered topical here, I'll take your
position more seriously. Until then ...
More to the point you would be able to explain standard C to the
heretics who thing that MS VC or Gcc are standard. Besides as you
have just pointed out even the standard has mistakes in it... so
"standard C" is not perfect.

Never said it was.

[...]
 
R

Richard Heathfield

Chris Hills said:
Right! Back to the fight...

Please, Chris, if it's a fight you want, can you take it to rec.boxing
or something? Or if you are just here to disagree with people, then
please say so explicitly.

At the moment, I have you down as "clueful person who nevertheless
doesn't understand why clc defines topicality the way it does", or
perhaps "clueful person who does understand but doesn't agree"
(although I think that's less likely, TBH).

But if in reality you are "clueful person who enjoys indulging in
argument for argument's sake", then there's nothing wrong with that and
I can see how it might have its appeal, but frankly I have better
things to do than to argue the toss with people just as a way of
passing the time.

So if you're this latter kind, please fess up so that I can protect my
time usage accordingly.
 
K

Keith Thompson

Richard Heathfield said:
Chris Hills said:

Please, Chris, if it's a fight you want, can you take it to rec.boxing
or something? Or if you are just here to disagree with people, then
please say so explicitly.
[...]

I think he was joking.
 
C

Chris Hills

Richard Heathfield said:
Chris Hills said:


Please, Chris, if it's a fight you want, can you take it to rec.boxing
or something? Or if you are just here to disagree with people, then
please say so explicitly.

Please quote correctly . You missed out the smileys
 
C

Chris Hills

Keith Thompson said:
Richard Heathfield said:
Chris Hills said:

Please, Chris, if it's a fight you want, can you take it to rec.boxing
or something? Or if you are just here to disagree with people, then
please say so explicitly.
[...]

I think he was joking.
I was.
 
R

Richard Heathfield

Chris Hills said:
Keith Thompson said:
Richard Heathfield said:
Chris Hills said:
Right! Back to the fight...

Please, Chris, if it's a fight you want, can you take it to
rec.boxing or something? Or if you are just here to disagree with
people, then please say so explicitly.
[...]

I think he was joking.
I was.

Okay, fairysnuff... I apologise for misreading you.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,774
Messages
2,569,596
Members
45,140
Latest member
SweetcalmCBDreview
Top