Path: csiph.com!eternal-september.org!feeder.eternal-september.org!reader01.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: "...the value becomes,indeterminate each time the declaration is reached"
Date: Thu, 07 Feb 2019 10:33:39 -0800
Organization: None to speak of
Lines: 50
Message-ID: <ln7eebv470.fsf@kst-u.example.com>
References: <q2q79p$1j7$1@dont-email.me> <5c51d4fb.236542547@news.xmission.com> <q2sq30$3u8$1@dont-email.me> <q2vcq9$gem$1@dont-email.me> <JmI4E.228608$IF1.51786@fx03.am4> <20190201132441.0b099e5f38fdb72371bc44f0@g{oogle}mail.com> <XLW4E.338937$dp2.170071@fx18.am4> <20190204120834.7d3f84f25052b6a8315f693a@g{oogle}mail.com> <q3939g$iup$1@dont-email.me> <HhW5E.684944$Fi6.335600@fx12.am4> <q3a9fq$t2$1@dont-email.me> <sW36E.455967$kE1.110872@fx20.am4> <q3bvfe$u96$1@dont-email.me> <ybh6E.591739$Y32.265491@fx19.am4> <q3cbch$73t$1@dont-email.me> <nQk6E.362552$dp2.170508@fx18.am4> <gbu653Ffhg9U7@mid.individual.net> <4c59dca0-2449-4234-a935-b6d604088caf@googlegroups.com> <lnbm3quedw.fsf@kst-u.example.com> <84b47866-600b-48d2-85b6-1538f19b00df@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: reader02.eternal-september.org; posting-host="f376dcb2b82f2c4f0e13c019c6c4653e"; logging-data="8122"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+jSIzYQUUdDvHAUJ8B2kSB"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:klbMSUVd1iunwFl98L47eGeru5U= sha1:0ZU79FwYkSojzHiWCD8rP/F6rCk=
Xref: csiph.com comp.lang.c:140739

Malcolm McLean <malcolm.arthur.mclean@gmail.com> writes:
> On Tuesday, 5 February 2019 21:14:28 UTC, Keith Thompson  wrote:
>> Malcolm McLean <malcolm.arthur.mclean@gmail.com> writes:
>> > On Tuesday, 5 February 2019 19:23:26 UTC, Ian Collins  wrote:
>> >> You are being willfully obtuse, const in C++ is a significant 
>> >> improvement over const in C.
>> >> 
>> > It's significantly more complex.
>> >
>> > It's no longer true, as it is in C, that a correct program cannot
>> > be turned into an incorrect program by const-qualifying or const-
>> > dequalifying identifiers.
>> 
>> Can you clarify what you mean by that?
>> 
>> I can turn this correct program:
>> 
>> int main(void) {
>>     int n = 10;
>>     n = 20;
>> }
>> 
>> into an incorrect program by const-qualifying n.
>> 
> What I (actually Dennis Ritchie) meant is this.
>
> We have 
>
> double total_salaries(const Employee *emp, int N);
>
> The function works, and it doesn't corrupt the employee array.
> Now we can take the const away, and the function must still work
> correctly. We can add it again, and it still works correctly.

If I have a call like this:

    const Employee staff[42];
    double cost = total_salaries(staff, 42);

then removing the "const" from the parameter declaration makes the
program incorrect.

I think that if you remove *all* occurrences of "const" from a correct
program, the result is still a correct program (though removing "const"
makes it easier to make the program incorrect).

-- 
Keith Thompson (The_Other_Keith) kst-u@mib.org  <http://www.ghoti.net/~kst>
Will write code for food.
void Void(void) { Void(); } /* The recursive call of the void */