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=-0.3 required=5.0 tests=BAYES_00,
	REPLYTO_WITHOUT_TO_CC autolearn=no autolearn_force=no version=3.4.4
Path: 
 eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!news.eternal-september.org!news.eternal-september.org!feeder.eternal-september.org!goblin1!goblin.stu.neva.ru!news.astraweb.com!border6.a.newsrouter.astraweb.com!cyclone03.ams2.highwinds-media.com!news.highwinds-media.com!voer-me.highwinds-media.com!post02.fr7!fx13.fr7.POSTED!not-for-mail
Message-ID: <Cq5U8ICcvtySFwRj@ada-augusta.demon.co.uk>
From: Mike H <postmaster@ada-augusta.demon.co.uk>
Reply-To: Mike Hopkins <postmaster@ada-augusta.demon.co.uk>
Newsgroups: comp.lang.ada
Subject: Re: Pretty printing the pretty print.
References: <jh5A$9EEzrQSFwlw@ada-augusta.demon.co.uk>
 <a0ec7acb-416a-4962-b653-dd2adf70dfd6@googlegroups.com>
 <klBB+XGkevQSFwEz@ada-augusta.demon.co.uk>
 <da27e87a-7341-4b83-89f7-f017bb74de18@googlegroups.com>
 <ICL6XpEJdYRSFwbW@ada-augusta.demon.co.uk> <20140104121332.22310cd2@vostro>
MIME-Version: 1.0
Content-Type: text/plain;charset=us-ascii
User-Agent: Turnpike/6.07-M (<sJ2$+n7g77P7n5diSa7NqJ5a21>)
NNTP-Posting-Host: 83.104.138.185
X-Complaints-To: abuse@demon.net
X-Trace: 1389026755 83.104.138.185 (Mon, 06 Jan 2014 16:45:55 UTC)
NNTP-Posting-Date: Mon, 06 Jan 2014 16:45:55 UTC
Date: Mon, 6 Jan 2014 16:37:48 +0000
X-Received-Body-CRC: 3806746329
X-Received-Bytes: 2669
Xref: news.eternal-september.org comp.lang.ada:18132
Date: 2014-01-06T16:37:48+00:00
List-Id: <comp.lang.ada>

In message <20140104121332.22310cd2@vostro>, Oliver Kleinke
<oliver.kleinke@c-01a.de> writes
>Do you by chance remember the task?
As far as I remember the original "specification" was a requirement to
split an array at an arbitrary point and re-join the two parts (first
and last) in the reverse order (last then first). Spotting that what was
actually required was a circular shift changed the game somewhat. By
some odd coincidence, before Christmas I stumbled across the following
(see below). It seems to fit the bill. I also seem to remember that
turning it into a generic was not as sweet and clean as I expected.

function Shift_right_circular
     (A      : in Array_type;
      Places : in Natural) return Array_type is

      P : Natural := Places rem A'LENGTH;

      -- Array will be split into two slices (I1 .. I2) & (I3 .. I4) and
      -- re-made as (I3 .. I4) & (I1 .. I2). The values of I2 and I3 are
      -- derived from their position numbers, P2 and P3.
      P1 : Natural := Index_type'POS (A'FIRST); --  position number of
I1
      P2 : Natural := P1 + P - 1;                        --  position
number of I2
      P3 : Natural := P1 + P;                             --  position
number of I3
      --
      -- I1 is A'FIRST
      I2 : Index_type := Index_type'VAL(P2);
      I3 : Index_type := Index_type'VAL(P3);
      -- I4 is A'LAST

begin

      if Places = 0 then
         -- no shifting to be done
         return A;
      else
         return A (I3 .. A'LAST) & A (A'FIRST .. I2);
      end if;

end Shift_right_circular;


-- 
Mike
Swim? Naturally at Severn Vale
<http://www.severnvalesc.org/>