Path: csiph.com!news.swapon.de!eternal-september.org!feeder.eternal-september.org!mx02.eternal-september.org!.POSTED!not-for-mail
From: Keith Thompson <kst-u@mib.org>
Newsgroups: comp.lang.c
Subject: Re: C pointer semantics with dynamically allocated struct
Date: Sat, 17 Oct 2015 14:50:02 -0700
Organization: None to speak of
Lines: 24
Message-ID: <lnio652mid.fsf@kst-u.example.com>
References: <mqdg0t$rv2$1@speranza.aioe.org> <561e2e49.3744281@news.xs4all.nl> <lna8rla28s.fsf@kst-u.example.com> <mvmpj0$86a$1@speranza.aioe.org> <537dc231-7dc2-4ecb-8d55-4d847c9e8e0c@googlegroups.com> <lnlhb46m0h.fsf@kst-u.example.com> <29db6d81-16ae-4f18-9a3a-ac290b823144@googlegroups.com> <lnd1wf7t37.fsf@kst-u.example.com> <839bc0ff-6090-4171-956b-10cd0db49129@googlegroups.com> <ln8u737njv.fsf@kst-u.example.com> <a6485b2d-c8df-429c-a5bf-3eafe0885966@googlegroups.com> <mvs22t$osc$1@speranza.aioe.org> <256ffb24-2a28-490d-88a3-07655412daa7@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Injection-Info: mx02.eternal-september.org; posting-host="945944de09706c9b4e29b53c9d2efdc2"; logging-data="12841"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+8JvwgAeBA80LMP4I2+xXT"
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux)
Cancel-Lock: sha1:Ej+omsCIFeeEOeHzMVM/cTbOLN4= sha1:fnIVIzY6flEhHIsD3GV7hC2KVzA=
Xref: csiph.com comp.lang.c:73158

supercat@casperkitty.com writes:
> On Friday, October 16, 2015 at 6:43:38 PM UTC-5, glen herrmannsfeldt wrote:
>> I wonder why signed char exists at all.  I suspect that there is some
>> system that makes it hard to process unsigned char data. That is, where
>> char at sign extended on promotion to larger types. With extra
>> instructions, (presumably used for actual (unsigned char), it does
>> do the conversion. That saves code space and execution time, for a
>> rarely occuring (on ASCII systems) negative char values.
>
> There are many situations where 8-bit signed values are useful.  Further, I
> believe INT_MAX is supposed to be at least equal to CHAR_MAX, which would
> imply that on systems where sizeof (int) is 1, "char" would have to be
> signed.

There is no such requirement.  Systems with CHAR_MAX > INT_MAX would
have some difficulties with stdio (in particular a program couldn't
distinguish between EOF and a valid character without checking the
results of feof() and ferror()), but they're not prohibited.

-- 
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"