Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.c > #103347 > unrolled thread

Is it legal to download K&R from this site ?

Back to article view | Back to comp.lang.c

Contents

  Is it legal to download K&R from this site ? MehdiAmini <atorrses@gmail.com> - 2017-02-07 14:09 +0330
    Re: Is it legal to download K&R from this site ? David Brown <david.brown@hesbynett.no> - 2017-02-07 12:17 +0100
      Re: Is it legal to download K&R from this site ? MehdiAmini <atorrses@gmail.com> - 2017-02-07 19:44 +0330
        Re: Is it legal to download K&R from this site ? David Brown <david.brown@hesbynett.no> - 2017-02-08 09:47 +0100
          Re: Is it legal to download K&R from this site ? void <nathan.gothier@gmail.com> - 2017-02-11 04:25 -0800
            Re: Is it legal to download K&R from this site ? "James R. Kuyper" <jameskuyper@verizon.net> - 2017-02-11 10:18 -0500
              Re: Is it legal to download K&R from this site ? void <nathan.gothier@gmail.com> - 2017-02-11 10:02 -0800
                Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-11 10:32 -0800
                  Re: Is it legal to download K&R from this site ? void <nathan.gothier@gmail.com> - 2017-02-11 11:17 -0800
                    Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-11 11:46 -0800
                      Re: Is it legal to download K&R from this site ? void <nathan.gothier@gmail.com> - 2017-02-11 12:25 -0800
                        Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-11 13:29 -0800
                          Re: Is it legal to download K&R from this site ? Melzzzzz <Melzzzzz@zzzzz.com> - 2017-02-11 22:59 +0000
                          Re: Is it legal to download K&R from this site ? Tim Prince <tprince@computer.org> - 2017-02-11 18:14 -0500
                          Re: Is it legal to download K&R from this site ? GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-02-12 04:08 +0100
                    Re: Is it legal to download K&R from this site ? Flippie Flink <flip@flink.not> - 2017-02-11 23:44 +0100
                Re: Is it legal to download K&R from this site ? "James R. Kuyper" <jameskuyper@verizon.net> - 2017-02-11 13:36 -0500
                  Re: Is it legal to download K&R from this site ? Tim Prince <tprince@computer.org> - 2017-02-11 18:21 -0500
                Re: Is it legal to download K&R from this site ? David Brown <david.brown@hesbynett.no> - 2017-02-13 13:56 +0100
                  Re: Is it legal to download K&R from this site ? GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-02-13 14:06 +0100
                    Re: Is it legal to download K&R from this site ? David Brown <david.brown@hesbynett.no> - 2017-02-13 15:05 +0100
                    Re: Is it legal to download K&R from this site ? "James R. Kuyper" <jameskuyper@verizon.net> - 2017-02-13 10:38 -0500
              Re: Is it legal to download K&R from this site ? Robert Wessel <robertwessel2@yahoo.com> - 2017-02-12 01:38 -0600
                Re: Is it legal to download K&R from this site ? "James R. Kuyper" <jameskuyper@verizon.net> - 2017-02-12 09:47 -0500
              Re: Is it legal to download K&R from this site ? supercat@casperkitty.com - 2017-02-12 14:15 -0800
              Re: Is it legal to download K&R from this site ? raltbos@xs4all.nl (Richard Bos) - 2017-02-14 12:25 +0000
                Re: Is it legal to download K&R from this site ? James Kuyper <jameskuyper@verizon.net> - 2017-02-14 11:55 -0500
                  Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-14 09:30 -0800
                Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-14 09:06 -0800
                  Re: Is it legal to download K&R from this site ? raltbos@xs4all.nl (Richard Bos) - 2017-02-15 21:04 +0000
            Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-11 08:58 -0800
            Re: Is it legal to download K&R from this site ? raltbos@xs4all.nl (Richard Bos) - 2017-02-14 12:23 +0000
              Re: Is it legal to download K&R from this site ? GOTHIER Nathan <nathan.gothier@gmail.com> - 2017-02-14 14:08 +0100
            Re: Is it legal to download K&R from this site ? Robert Latest <boblatest@yahoo.com> - 2017-02-19 11:13 +0000
              Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-19 07:36 -0800
                Re: Is it legal to download K&R from this site ? Tim Rentsch <txr@alumni.caltech.edu> - 2017-02-19 10:18 -0800
        Re: Is it legal to download K&R from this site ? Tim Rentsch <txr@alumni.caltech.edu> - 2017-02-09 08:15 -0800
    Re: Is it legal to download K&R from this site ? Keith Thompson <kst-u@mib.org> - 2017-02-07 08:35 -0800
    Re: Is it legal to download K&R from this site ? Real Troll <real.troll@trolls.com> - 2017-02-12 00:01 -0400
      Re: Is it legal to download K&R from this site ? MehdiAmini <atorrses@gmail.com> - 2017-02-12 08:32 +0330
        Re: Is it legal to download K&R from this site ? Real Troll <real.troll@trolls.com> - 2017-02-12 01:55 -0400
        Re: Is it legal to download K&R from this site ? "James R. Kuyper" <jameskuyper@verizon.net> - 2017-02-12 09:50 -0500
          Re: Is it legal to download K&R from this site ? Real Troll <real.troll@trolls.com> - 2017-02-12 14:45 -0400
    Re: Is it legal to download K&R from this site ? "Rick C. Hodgin" <rick.c.hodgin@gmail.com> - 2017-02-14 09:33 -0800

Page 1 of 3  [1] 2 3  Next page →

#103347 — Is it legal to download K&R from this site ?

Hi,

Can I download the book from the provided link ?


http://caryney.us/the_c_programming_language_second_edition.pdf

-- 
Farewell.

[toc] | [next] | [standalone]

#103348

On 07/02/17 11:39, MehdiAmini wrote:
> Hi,
> 
> Can I download the book from the provided link ?
> 
> 
> http://caryney.us/the_c_programming_language_second_edition.pdf
> 

The book is old, but still on sale both as physical copies and ebook
formats.  It is still often recommended as an introduction to C, and
also as a fine example of good technical writing, though it does not
cover more modern C standards (C99 and C11).

Since it is popular to buy, I do not expect there to be any legal way to
get a free download and you should be wary of any site offering it.

[toc] | [prev] | [next] | [standalone]

#103372

On 02/07/2017 02:47 PM, David Brown wrote:
 > On 07/02/17 11:39, MehdiAmini wrote:
 >> Hi,
 >>
 >> Can I download the book from the provided link ?
 >>
 >>
 >> http://caryney.us/the_c_programming_language_second_edition.pdf
 >>
 >
 > The book is old, but still on sale both as physical copies and ebook
 > formats.  It is still often recommended as an introduction to C, and
 > also as a fine example of good technical writing, though it does not
 > cover more modern C standards (C99 and C11).

I like C90 better than C99 or C11.

 >
 > Since it is popular to buy, I do not expect there to be any legal way to
 > get a free download and you should be wary of any site offering it.
 >

So maybe for now I have to stick to latest C89 Draft.


-- 
Farewell.

[toc] | [prev] | [next] | [standalone]

#103432

On 07/02/17 17:14, MehdiAmini wrote:
> On 02/07/2017 02:47 PM, David Brown wrote:
>> On 07/02/17 11:39, MehdiAmini wrote:
>>> Hi,
>>>
>>> Can I download the book from the provided link ?
>>>
>>>
>>> http://caryney.us/the_c_programming_language_second_edition.pdf
>>>
>>
>> The book is old, but still on sale both as physical copies and ebook
>> formats.  It is still often recommended as an introduction to C, and
>> also as a fine example of good technical writing, though it does not
>> cover more modern C standards (C99 and C11).
> 
> I like C90 better than C99 or C11.

Why?  What are you trying to achieve here?  If you are learning C, then
you need to learn the C that people use - which is mostly C99, with a
small amount of C11 in some cases, and a small amount of C90 for code
that needs maximal portability or for maintaining old code.

> 
>>
>> Since it is popular to buy, I do not expect there to be any legal way to
>> get a free download and you should be wary of any site offering it.
>>
> 
> So maybe for now I have to stick to latest C89 Draft.
> 

Do you mean the draft C89 standard?

[toc] | [prev] | [next] | [standalone]

#103676

On Wednesday, February 8, 2017 at 9:47:50 AM UTC+1, David Brown wrote:
> On 07/02/17 17:14, MehdiAmini wrote:
> > On 02/07/2017 02:47 PM, David Brown wrote:
> >> On 07/02/17 11:39, MehdiAmini wrote:
> >>> Hi,
> >>>
> >>> Can I download the book from the provided link ?
> >>>
> >>>
> >>> http://caryney.us/the_c_programming_language_second_edition.pdf
> >>>
> >>
> >> The book is old, but still on sale both as physical copies and ebook
> >> formats.  It is still often recommended as an introduction to C, and
> >> also as a fine example of good technical writing, though it does not
> >> cover more modern C standards (C99 and C11).
> > 
> > I like C90 better than C99 or C11.
> 
> Why?  What are you trying to achieve here?  If you are learning C, then
> you need to learn the C that people use - which is mostly C99, with a
> small amount of C11 in some cases, and a small amount of C90 for code
> that needs maximal portability or for maintaining old code.

C90 is fully implemented by most popular compilers while C99 and C11 don't. Thus K&R is the only common denomiator for C programmers and by far cheaper than any ISO paper. Once more new features aren't outstanding...

[toc] | [prev] | [next] | [standalone]

#103679

On 02/11/2017 07:25 AM, void wrote:
> On Wednesday, February 8, 2017 at 9:47:50 AM UTC+1, David Brown wrote:
>> On 07/02/17 17:14, MehdiAmini wrote:
...
>>> I like C90 better than C99 or C11.
>>
>> Why?  What are you trying to achieve here?  If you are learning C, then
>> you need to learn the C that people use - which is mostly C99, with a
>> small amount of C11 in some cases, and a small amount of C90 for code
>> that needs maximal portability or for maintaining old code.
>
> C90 is fully implemented by most popular compilers while C99 and C11 don't. Thus K&R is the only common denomiator for C programmers and by far cheaper than any ISO paper. Once more new features aren't outstanding...

C99 has been amended by three technical corrigenda, each of which 
specified the changes to be made to C99. However, it has never been 
officially released with those changes made. n1256.pdf is an unofficial 
(but still ISO) draft of what C99 would look like with those changes, 
making it paradoxically more useful than any official version of C99. It 
contains only one known error relative to it's official base documents, 
and is absolutely free. How could K&R be cheaper than that?

n1570.pdf is almost identical to the final C2011 standard, having only 
one known error, and is also available for free.

C99 is almost completely implemented by all of the popular compilers. If 
one of those compilers isn't installed, I would pressure the SAs to 
install it. If a feature I'd like to use isn't implemented, I'd pressure 
the implementor to add that feature. I don't see a point in insisting on 
conforming to an ancient version of the standard when it's entirely 
feasible to use almost the entirety of C99 on almost all platforms.

Unless, of course, someone else requires you to write code which 
"complies" with C90, as I am, for one of the two projects I work on 
(more precisely, it must "comply" with C95 == C90 + two technical 
corrigenda).

[toc] | [prev] | [next] | [standalone]

#103688

On Saturday, February 11, 2017 at 4:18:14 PM UTC+1, James R. Kuyper wrote:
> On 02/11/2017 07:25 AM, void wrote:
> > On Wednesday, February 8, 2017 at 9:47:50 AM UTC+1, David Brown wrote:
> >> On 07/02/17 17:14, MehdiAmini wrote:
> ...
> >>> I like C90 better than C99 or C11.
> >>
> >> Why?  What are you trying to achieve here?  If you are learning C, then
> >> you need to learn the C that people use - which is mostly C99, with a
> >> small amount of C11 in some cases, and a small amount of C90 for code
> >> that needs maximal portability or for maintaining old code.
> >
> > C90 is fully implemented by most popular compilers while C99 and C11 don't. Thus K&R is the only common denomiator for C programmers and by far cheaper than any ISO paper. Once more new features aren't outstanding...
> 
> C99 has been amended by three technical corrigenda, each of which 
> specified the changes to be made to C99. However, it has never been 
> officially released with those changes made. n1256.pdf is an unofficial 
> (but still ISO) draft of what C99 would look like with those changes, 
> making it paradoxically more useful than any official version of C99. It 
> contains only one known error relative to it's official base documents, 
> and is absolutely free. How could K&R be cheaper than that?
> 
> n1570.pdf is almost identical to the final C2011 standard, having only 
> one known error, and is also available for free.
> 

Drafts can give C programmers a quick view on new features but these documents are incomplete descriptions of future standards. Of course, there are free of charge (minus your ISP provider cost) because there are working document in progress. Who would pay for incomplete products? I personally used draft of C9x and rationales (very educative to understand the C library implementation) to learn C90 wich could be enough to begin C programming. Later I bought K&R because I prefer to get the C Bible at hand when I need to clear or complete my knowledge. People should know that a physical book is a true portable document contrarily to a digital book chained to a reader with digital restrictions management.

> C99 is almost completely implemented by all of the popular compilers. If 
> one of those compilers isn't installed, I would pressure the SAs to 
> install it. If a feature I'd like to use isn't implemented, I'd pressure 
> the implementor to add that feature. I don't see a point in insisting on 
> conforming to an ancient version of the standard when it's entirely 
> feasible to use almost the entirety of C99 on almost all platforms.
> 
> Unless, of course, someone else requires you to write code which 
> "complies" with C90, as I am, for one of the two projects I work on 
> (more precisely, it must "comply" with C95 == C90 + two technical 
> corrigenda).

C99 may be almost completely implemented by popular compilers but that doesn't mean that C99 features are required by all C programmers. Actually I prefer to comply to C90 most of the time or C95 for i18n support and C99 for big integers. Actually, writing C90 code doesn't prevent you from using C99 compilers. ;-)

[toc] | [prev] | [next] | [standalone]

#103691

void <nathan.gothier@gmail.com> writes:
[...]
> Drafts can give C programmers a quick view on new features but these
> documents are incomplete descriptions of future standards. Of course,
> there are free of charge (minus your ISP provider cost) because there
> are working document in progress. Who would pay for incomplete
> products? I personally used draft of C9x and rationales (very
> educative to understand the C library implementation) to learn C90
> wich could be enough to begin C programming. Later I bought K&R
> because I prefer to get the C Bible at hand when I need to clear or
> complete my knowledge. People should know that a physical book is a
> true portable document contrarily to a digital book chained to a
> reader with digital restrictions management.

Please limit your line length to 80 characters or less, preferably
about 72.

Not all digital books have restrictions.  The PDF drafts of the
C standard, particularly n1256.pdf and n1570.pdf, have no DRM
restrictions.  My PDF copy of the ISO C11 standard, which I paid
for, does have some kind of DRM (my reader says it's password
protected, but I've never been prompted to enter a password),
and the bottom of each page says its licensed to me personally.
I'm able to copy-and-paste from it.  I did have to agree to some
terms when I bought it, but those terms are not directly enforced.

[...]

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"

[toc] | [prev] | [next] | [standalone]

#103694

On Saturday, February 11, 2017 at 7:32:09 PM UTC+1, Keith Thompson wrote:
> void writes:
> [...]
> > Drafts can give C programmers a quick view on new features but these
> > documents are incomplete descriptions of future standards. Of course,
> > there are free of charge (minus your ISP provider cost) because there
> > are working document in progress. Who would pay for incomplete
> > products? I personally used draft of C9x and rationales (very
> > educative to understand the C library implementation) to learn C90
> > wich could be enough to begin C programming. Later I bought K&R
> > because I prefer to get the C Bible at hand when I need to clear or
> > complete my knowledge. People should know that a physical book is a
> > true portable document contrarily to a digital book chained to a
> > reader with digital restrictions management.
> 
> Please limit your line length to 80 characters or less, preferably
> about 72.
> 

Sorry but I don't have a such limit reading text from Google groups. I thought NNTP is no longer supported by Google groups and since wide screens are pretty common nowadays...

> Not all digital books have restrictions.  The PDF drafts of the
> C standard, particularly n1256.pdf and n1570.pdf, have no DRM
> restrictions.  My PDF copy of the ISO C11 standard, which I paid
> for, does have some kind of DRM (my reader says it's password
> protected, but I've never been prompted to enter a password),
> and the bottom of each page says its licensed to me personally.
> I'm able to copy-and-paste from it.  I did have to agree to some
> terms when I bought it, but those terms are not directly enforced.
> 
> [...]

Almost $200.00 for C11... that's not really affordable for everybody. Can you lend it at will? I can move my K&R book everywhere and read it with the green energy of natural light. :-)

[toc] | [prev] | [next] | [standalone]

#103695

void <nathan.gothier@gmail.com> writes:
> On Saturday, February 11, 2017 at 7:32:09 PM UTC+1, Keith Thompson wrote:
>> void writes:
>> [...]
>> > Drafts can give C programmers a quick view on new features but these
>> > documents are incomplete descriptions of future standards. Of course,
>> > there are free of charge (minus your ISP provider cost) because there
>> > are working document in progress. Who would pay for incomplete
>> > products? I personally used draft of C9x and rationales (very
>> > educative to understand the C library implementation) to learn C90
>> > wich could be enough to begin C programming. Later I bought K&R
>> > because I prefer to get the C Bible at hand when I need to clear or
>> > complete my knowledge. People should know that a physical book is a
>> > true portable document contrarily to a digital book chained to a
>> > reader with digital restrictions management.
>> 
>> Please limit your line length to 80 characters or less, preferably
>> about 72.
>
> Sorry but I don't have a such limit reading text from Google groups. I
> thought NNTP is no longer supported by Google groups and since wide
> screens are pretty common nowadays...

Google Groups is, in part, a poorly implemented interface to Usenet,
which predates it by several decades.  Most of us here don't use Google
Groups; we use NNTP-based newsreaders.  I'm asking you to format your
posts to make them easier for most of us to read.

[...]

> Almost $200.00 for C11... that's not really affordable for
> everybody. Can you lend it at will? I can move my K&R book everywhere
> and read it with the green energy of natural light. :-)

I paid much less than that, but ISO doesn't consult me regarding their
pricing policies.  No, I can't lend it.  But even though I have a copy
of the official standard, I usually use N1570.

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"

[toc] | [prev] | [next] | [standalone]

#103696

On Saturday, February 11, 2017 at 8:45:55 PM UTC+1, Keith Thompson wrote:
> void writes:
> > On Saturday, February 11, 2017 at 7:32:09 PM UTC+1, Keith Thompson wrote:
> >> void writes:
> >> [...]
> >> > Drafts can give C programmers a quick view on new features but these
> >> > documents are incomplete descriptions of future standards. Of course,
> >> > there are free of charge (minus your ISP provider cost) because there
> >> > are working document in progress. Who would pay for incomplete
> >> > products? I personally used draft of C9x and rationales (very
> >> > educative to understand the C library implementation) to learn C90
> >> > wich could be enough to begin C programming. Later I bought K&R
> >> > because I prefer to get the C Bible at hand when I need to clear or
> >> > complete my knowledge. People should know that a physical book is a
> >> > true portable document contrarily to a digital book chained to a
> >> > reader with digital restrictions management.
> >> 
> >> Please limit your line length to 80 characters or less, preferably
> >> about 72.
> >
> > Sorry but I don't have a such limit reading text from Google groups. I
> > thought NNTP is no longer supported by Google groups and since wide
> > screens are pretty common nowadays...
> 
> Google Groups is, in part, a poorly implemented interface to Usenet,
> which predates it by several decades.  Most of us here don't use Google
> Groups; we use NNTP-based newsreaders.  I'm asking you to format your
> posts to make them easier for most of us to read.

Can you tell me which News server and reader do you use?

[toc] | [prev] | [next] | [standalone]

#103697

void <nathan.gothier@gmail.com> writes:
> On Saturday, February 11, 2017 at 8:45:55 PM UTC+1, Keith Thompson wrote:
[...]
>> Google Groups is, in part, a poorly implemented interface to Usenet,
>> which predates it by several decades.  Most of us here don't use Google
>> Groups; we use NNTP-based newsreaders.  I'm asking you to format your
>> posts to make them easier for most of us to read.
>
> Can you tell me which News server and reader do you use?

My news server is news.eternal-september.org, which is free.  My news
reader is Gnus, which runs as a mode under Emacs.  If you don't like
Emacs, you won't like Gnus.  Thunderbird is another popular option.

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Working, but not speaking, for JetHead Development, Inc.
"We must do something.  This is something.  Therefore, we must do this."
    -- Antony Jay and Jonathan Lynn, "Yes Minister"

[toc] | [prev] | [next] | [standalone]

#103704

On 2017-02-11, Keith Thompson <kst-u@mib.org> wrote:
> void <nathan.gothier@gmail.com> writes:
>> On Saturday, February 11, 2017 at 8:45:55 PM UTC+1, Keith Thompson wrote:
> [...]
>>> Google Groups is, in part, a poorly implemented interface to Usenet,
>>> which predates it by several decades.  Most of us here don't use Google
>>> Groups; we use NNTP-based newsreaders.  I'm asking you to format your
>>> posts to make them easier for most of us to read.
>>
>> Can you tell me which News server and reader do you use?
>
> My news server is news.eternal-september.org, which is free.  My news
> reader is Gnus, which runs as a mode under Emacs.  If you don't like
> Emacs, you won't like Gnus.  Thunderbird is another popular option.
>
slrn/vim here ;)

-- 
press any key to continue or any other to quit...

[toc] | [prev] | [next] | [standalone]

#103706

On 2/11/2017 4:29 PM, Keith Thompson wrote:
> void <nathan.gothier@gmail.com> writes:
>> On Saturday, February 11, 2017 at 8:45:55 PM UTC+1, Keith Thompson wrote:
> [...]
>>> Google Groups is, in part, a poorly implemented interface to Usenet,
>>> which predates it by several decades.  Most of us here don't use Google
>>> Groups; we use NNTP-based newsreaders.  I'm asking you to format your
>>> posts to make them easier for most of us to read.
>>
>> Can you tell me which News server and reader do you use?
> 
> My news server is news.eternal-september.org, which is free.  My news
> reader is Gnus, which runs as a mode under Emacs.  If you don't like
> Emacs, you won't like Gnus.  Thunderbird is another popular option.
> 
My main complaint about tbird is that I don't find an equivalent for
Android (or, if I give up and go with Apple) phones. I've found no
incentive to drop the very competitively priced news.individual.net,
works well with all my linux and Windows boxes.

[toc] | [prev] | [next] | [standalone]

#103711

On Sat, 11 Feb 2017 13:29:04 -0800
Keith Thompson <kst-u@mib.org> wrote:

> My news server is news.eternal-september.org, which is free.  My news
> reader is Gnus, which runs as a mode under Emacs.  If you don't like
> Emacs, you won't like Gnus.  Thunderbird is another popular option.

Thanks for the info. I'm currently testing Sylpheed connected to news.eternal-september.org and it seems to work correctly.

[toc] | [prev] | [next] | [standalone]

#103701

On 11-02-17 20:17, void wrote:
>> Please limit your line length to 80 characters or less, preferably
>> about 72.
>>
>
> Sorry but I don't have a such limit reading text from Google groups. I
> thought NNTP is no longer supported by Google groups and since wide
> screens are pretty common nowadays...

It's not about screen width, but about readability. Newspapers and magazines
use narrow columns too for the same reason.

-- 
Flinke Flip

[toc] | [prev] | [next] | [standalone]

#103692

On 02/11/2017 01:02 PM, void wrote:
> On Saturday, February 11, 2017 at 4:18:14 PM UTC+1, James R. Kuyper wrote:
...
>> C99 has been amended by three technical corrigenda, each of which
>> specified the changes to be made to C99. However, it has never been
>> officially released with those changes made. n1256.pdf is an unofficial
>> (but still ISO) draft of what C99 would look like with those changes,
>> making it paradoxically more useful than any official version of C99. It
>> contains only one known error relative to it's official base documents,
>> and is absolutely free. How could K&R be cheaper than that?
>>
>> n1570.pdf is almost identical to the final C2011 standard, having only
>> one known error, and is also available for free.
>>
>
> Drafts can give C programmers a quick view on new features but these
> documents are incomplete descriptions of future standards. Of course,

I don't think you properly understand exactly what n1256.pdf and 
n1570.pdf are. They contain almost exactly the same text as the actual 
standard, formatted almost exactly as it is in the actual standard. 
C2011 may have been a "future standard" at the time n1570.pdf was 
produced, but it is no longer a "future standard". At the time n1256.pdf 
was created, C99+TC1+TC2+TC3 was already the current standard. n1256.pdf 
differs from the final version of C99 only by reason of incorporating 
the changes specified in TC1, TC2, and TC3. Those changes were 
officially approved, but a version of the standard with those changes 
actually implemented has never been officially approved, and my 
understanding is that there never will be such a version. So n1256.pdf 
is not, in any sense, a description of any "future standard".

...
> C99 may be almost completely implemented by popular compilers but
> thatdoesn't mean that C99 features are required by all C programmers.

I wasn't talking about C99 features being required. There are clumsy 
workarounds for dealing with most of the situations where C99 features 
are more convenient than C90. I'm saying that for most of those 
features, compilers that support those features are sufficiently widely 
available that it is not justifiable to avoid using those features for 
the sake of portability (avoiding a feature because you just don't like 
it is a different matter entirely).
If there's any C99 feature that you like, you should feel free to use it 
without worrying about the possibility that your code might need to be 
ported to a platform for which there is no compiler that implements that 
feature. Only worry about such issues if and when they come up. And if 
they do come up, consider the possibility of delaying the port until 
after you've successfully pressured someone to actually implement the 
feature for that platform.

> Actually I prefer to comply to C90 most of the time or C95 for i18n
> support and C99 for big integers. Actually, writing C90 code doesn't
> prevent you from using C99 compilers. ;-)

I said nothing to suggest that it did; but it is in fact the case that 
some code that was correct according to C90 is no longer correct 
according to the latest version of the standard. This includes any code 
that relies upon implicit int, uses gets(), or uses an identifier that 
matches one of the newer keywords. However, I will concede that most 
code which avoids those pitfalls should compile just fine under C99.

[toc] | [prev] | [next] | [standalone]

#103707

On 2/11/2017 1:36 PM, James R. Kuyper wrote:
> On 02/11/2017 01:02 PM, void wrote:
>> On Saturday, February 11, 2017 at 4:18:14 PM UTC+1, James R. Kuyper
>> wrote:
> ...
>>> C99 has been amended by three technical corrigenda, each of which
>>> specified the changes to be made to C99. However, it has never been
>>> officially released with those changes made. n1256.pdf is an unofficial
>>> (but still ISO) draft of what C99 would look like with those changes,
>>> making it paradoxically more useful than any official version of C99. It
>>> contains only one known error relative to it's official base documents,
>>> and is absolutely free. How could K&R be cheaper than that?
>>>
>>> n1570.pdf is almost identical to the final C2011 standard, having only
>>> one known error, and is also available for free.
>>>
>>
>> Drafts can give C programmers a quick view on new features but these
>> documents are incomplete descriptions of future standards. Of course,
> 
> I don't think you properly understand exactly what n1256.pdf and
> n1570.pdf are. They contain almost exactly the same text as the actual
> standard, formatted almost exactly as it is in the actual standard.
> C2011 may have been a "future standard" at the time n1570.pdf was
> produced, but it is no longer a "future standard". At the time n1256.pdf
> was created, C99+TC1+TC2+TC3 was already the current standard. n1256.pdf
> differs from the final version of C99 only by reason of incorporating
> the changes specified in TC1, TC2, and TC3. Those changes were
> officially approved, but a version of the standard with those changes
> actually implemented has never been officially approved, and my
> understanding is that there never will be such a version. So n1256.pdf
> is not, in any sense, a description of any "future standard".
> 
> ...
>> C99 may be almost completely implemented by popular compilers but
>> thatdoesn't mean that C99 features are required by all C programmers.
> 
> I wasn't talking about C99 features being required. There are clumsy
> workarounds for dealing with most of the situations where C99 features
> are more convenient than C90. I'm saying that for most of those
> features, compilers that support those features are sufficiently widely
> available that it is not justifiable to avoid using those features for
> the sake of portability (avoiding a feature because you just don't like
> it is a different matter entirely).
> If there's any C99 feature that you like, you should feel free to use it
> without worrying about the possibility that your code might need to be
> ported to a platform for which there is no compiler that implements that
> feature. Only worry about such issues if and when they come up. And if
> they do come up, consider the possibility of delaying the port until
> after you've successfully pressured someone to actually implement the
> feature for that platform.
> 
>> Actually I prefer to comply to C90 most of the time or C95 for i18n
>> support and C99 for big integers. Actually, writing C90 code doesn't
>> prevent you from using C99 compilers. ;-)
> 
> I said nothing to suggest that it did; but it is in fact the case that
> some code that was correct according to C90 is no longer correct
> according to the latest version of the standard. This includes any code
> that relies upon implicit int, uses gets(), or uses an identifier that
> matches one of the newer keywords. However, I will concede that most
> code which avoids those pitfalls should compile just fine under C99.
> 
I'm still reluctant to use MSVC as a C compiler, even though VS2015
accepts a majority of the C99 features I use, and performs competitively
on a majority of my code.  I've not seen any documentation of the VS2015
subset of C99.  Not knowing the intended limitations detracts from its
usefulness as a C compiler.  So does the limitation to OpenMP 2.0, but
this comment will get me ostracized from usenet.

[toc] | [prev] | [next] | [standalone]

#103791

On 11/02/17 19:02, void wrote:
> On Saturday, February 11, 2017 at 4:18:14 PM UTC+1, James R. Kuyper
> wrote:
>> On 02/11/2017 07:25 AM, void wrote:
>>> On Wednesday, February 8, 2017 at 9:47:50 AM UTC+1, David Brown
>>> wrote:
>>>> On 07/02/17 17:14, MehdiAmini wrote:
>> ...
>>>>> I like C90 better than C99 or C11.
>>>> 
>>>> Why?  What are you trying to achieve here?  If you are learning
>>>> C, then you need to learn the C that people use - which is
>>>> mostly C99, with a small amount of C11 in some cases, and a
>>>> small amount of C90 for code that needs maximal portability or
>>>> for maintaining old code.
>>> 
>>> C90 is fully implemented by most popular compilers while C99 and
>>> C11 don't. Thus K&R is the only common denomiator for C
>>> programmers and by far cheaper than any ISO paper. Once more new
>>> features aren't outstanding...
>> 
>> C99 has been amended by three technical corrigenda, each of which 
>> specified the changes to be made to C99. However, it has never been
>>  officially released with those changes made. n1256.pdf is an
>> unofficial (but still ISO) draft of what C99 would look like with
>> those changes, making it paradoxically more useful than any
>> official version of C99. It contains only one known error relative
>> to it's official base documents, and is absolutely free. How could
>> K&R be cheaper than that?
>> 
>> n1570.pdf is almost identical to the final C2011 standard, having
>> only one known error, and is also available for free.
>> 
> 
> Drafts can give C programmers a quick view on new features but these
> documents are incomplete descriptions of future standards.

As James said, the drafts of the C standards (C99 and C11) are /almost
identical/ to the final versions.  They are not "incomplete
descriptions", nor are they for "future standards".  In the case of
N1570, there is /one/ minor difference between the freely available
"draft" and the expensive final release - out of some 700 pages.

And these are not "future standards", but real, existing and implemented
standards.  Most serious C compilers implement practically all of C99
(with the major exception of MSVC, though I believe it is better than it
used to be).  Newer versions of gcc and clang support practically all of
C11.  Some parts of the standards are dependent on the library rather
than the compiler.

> Of course,
> there are free of charge (minus your ISP provider cost) because there
> are working document in progress. Who would pay for incomplete
> products? I personally used draft of C9x and rationales (very
> educative to understand the C library implementation) to learn C90
> wich could be enough to begin C programming. Later I bought K&R
> because I prefer to get the C Bible at hand when I need to clear or
> complete my knowledge. People should know that a physical book is a
> true portable document contrarily to a digital book chained to a
> reader with digital restrictions management.

The draft standards have no DRM.  To my knowledge, the pdf files are
completely unrestricted and follow the ISO pdf standards, and work fine
with a wide variety of pdf readers.

> 
>> C99 is almost completely implemented by all of the popular
>> compilers. If one of those compilers isn't installed, I would
>> pressure the SAs to install it. If a feature I'd like to use isn't
>> implemented, I'd pressure the implementor to add that feature. I
>> don't see a point in insisting on conforming to an ancient version
>> of the standard when it's entirely feasible to use almost the
>> entirety of C99 on almost all platforms.
>> 
>> Unless, of course, someone else requires you to write code which 
>> "complies" with C90, as I am, for one of the two projects I work on
>>  (more precisely, it must "comply" with C95 == C90 + two technical
>>  corrigenda).
> 
> C99 may be almost completely implemented by popular compilers but
> that doesn't mean that C99 features are required by all C
> programmers. Actually I prefer to comply to C90 most of the time or
> C95 for i18n support and C99 for big integers. Actually, writing C90
> code doesn't prevent you from using C99 compilers. ;-)
> 

I don't use every feature of C90 in my programming - does that mean I
should use K&R C instead?  Your "argument" here is just silly.

[toc] | [prev] | [next] | [standalone]

#103792

On Mon, 13 Feb 2017 13:56:23 +0100
David Brown <david.brown@hesbynett.no> wrote:

> I don't use every feature of C90 in my programming - does that mean I
> should use K&R C instead?  Your "argument" here is just silly.

The latest K&R C is C90. If you don't use all C90 features you can use
pre-standard C but without garantee to be portable that's all. ;-)

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | comp.lang.c

csiph-web