From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,FREEMAIL_FROM
	autolearn=ham autolearn_force=no version=3.4.4
X-Google-Thread: 103376,5e54ec0ce937978
X-Google-Attributes: gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news1.google.com!news1.google.com!news2.google.com!news.glorb.com!newsfeed.news2me.com!newsfeed.icl.net!newsfeed.fjserv.net!peer-uk.news.demon.net!kibo.news.demon.net!mutlu.news.demon.net!news.demon.co.uk!demon!not-for-mail
From: Simon Wright <simon.j.wright@mac.com>
Newsgroups: comp.lang.ada
Subject: Re: String literals and wide_string literals - how?
Date: Sun, 22 Apr 2007 22:18:16 +0100
Organization: Pushface
Message-ID: <m28xckds6f.fsf@mac.com>
References: <1177063665.093083.241580@e65g2000hsc.googlegroups.com>
 <wccbqhjw0we.fsf@shell01.TheWorld.com> <vSaWh.5939$n_.5043@attbi_s21>
 <m2lkgmdpz3.fsf@mac.com> <u5yWh.7935$n_.7293@attbi_s21>
 <m2d51weo9p.fsf@mac.com> <2oPWh.9599$n_.6781@attbi_s21>
NNTP-Posting-Host: pogner.demon.co.uk
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Trace: news.demon.co.uk 1177276700 3734 62.49.19.209 (22 Apr 2007 21:18:20
 GMT)
X-Complaints-To: abuse@demon.net
NNTP-Posting-Date: Sun, 22 Apr 2007 21:18:20 +0000 (UTC)
Cancel-Lock: sha1:RaPKuvK1te2k5wAF+wwd7npt/Y4=
User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.95 (darwin)
Xref: g2news1.google.com comp.lang.ada:15225
Date: 2007-04-22T22:18:16+01:00
List-Id: <comp.lang.ada>

"Jeffrey R. Carter" <jrcarter@acm.org> writes:

> Simon Wright wrote:
>>
>> It may not be useful but in some applications it's essential to know.
>>
>> And it seems to me that a compiler that thought 2#00110011# was
>> acceptable probably wouldn't itself be acceptable.
>
> Why not? It adheres to the well defined and portable semantics Duff
> mentioned.

It just seems wilfully perverse, that's all. What other bizarre
decisions would the designers have made? Why choose software/hardware
that does that if there's a more reasonable alternative?

> My point was that these kinds of arrays are promoted as a better way
> to access individual bits than logical operations and shifts. But
> unless you have some definition of which bit "Y (Y'First)" accesses,
> they're not useful for that because they're extremely compiler
> dependent, while logical operations and shifts are portable across
> compilers and targets. Some sort of correspondence with the bit
> numbers used in record representation clauses would be OK.

I don't quite see the last sentence -- typically the bit numbering
depends on the endianness, very annoying -- not sure whether new Ada
sorts that one out.

For the rest, I quite agree; _much_ easier to map the C condition 
"(x & (1<<2))" to "(X and 2#00000100#) /= 0".