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 autolearn=ham
	autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,bffcdbd805329ff8
X-Google-Attributes: gid103376,public
X-Google-ArrivalTime: 2001-08-30 00:08:05 PST
Path: 
 archiver1.google.com!newsfeed.google.com!newsfeed.stanford.edu!nntp.cs.ubc.ca!newsfeed.direct.ca!look.ca!newshub2.rdc1.sfba.home.com!news.home.com!news1.rdc1.sfba.home.com.POSTED!not-for-mail
From: tmoran@acm.org
Newsgroups: comp.lang.ada
Subject: Re: Ada and Internet stuff
References: <dZbj7.2852$T4.26368@www.newsranger.com>
X-Newsreader: Tom's custom newsreader
Message-ID: <oDlj7.30207$sa.15327442@news1.rdc1.sfba.home.com>
Date: Thu, 30 Aug 2001 07:08:04 GMT
NNTP-Posting-Host: 24.7.82.199
X-Complaints-To: abuse@home.net
X-Trace: news1.rdc1.sfba.home.com 999155284 24.7.82.199 (Thu,
 30 Aug 2001 00:08:04 PDT)
NNTP-Posting-Date: Thu, 30 Aug 2001 00:08:04 PDT
Organization: Excite@Home - The Leader in Broadband http://home.com/faster
Xref: archiver1.google.com comp.lang.ada:12585
Date: 2001-08-30T07:08:04+00:00
List-Id: <comp.lang.ada>

> >of input and encodes it in 4 output text characters.  But the code
> >to automagically adjust to Stream_Elements that aren't 8 bits wide
> >would, I suspect, be substantally less clear, and rarely used, so
>
> It only takes about 2 declarations to do this, and the resulting constants can
  For true generality, the obvious is something along the lines:
Finding the GCD of 24 and any number from 1 .. 64 takes at most 5
iterations.  So
a1 : constant := 24;
b1 : constant := Stream_Element'size;
...
a5 :=
b5 :=
will give the GCD.
LCM : constant := 24*Stream_Element'size/GCD;
-- LCM is the least common multiple of 24 bits and Stream_Element'size
type Encoded_24 is record
  a,b,c,d:Six_Bits;
end record;
for Encoding_Block use record
  a at 0 range 0 .. 5;
  b at 0 range 6 .. 11;
  c at 0 range 12 .. 17;
  d at 0 range 18 .. 23;
end record;
for Encoding_Block'size use 24;

type Encodable_Group is array(1 .. LCM/24) of Encodable_Block;
for Encodable_Group'size use LCM;

subtype To_Encode is Stream_Element_Array(1 .. LCM/Stream_Element'size);
for To_Encode'size use LCM;

function SE_To_Group is new Ada.Unchecked_Conversion
  (Source => To_Encode,
   Target => Encodable_Group);

The_Group : Encodable_Group;

Then encoding proceeds in chunks by collecting a slice,
si := Source'first;
while si <= Source'last loop
  The_Group := SE_To_Group(Source(si .. Stream_Element_Offset'min(
                                           Source'last,
                                           si+To_Encode'length-1))
                           & end_case_padding);
  for i in Encodable_Group'range loop
    Emit(Translation_Character(The_Group(i).a);
    exit when this is the last actual datum excluding padding;
    Emit(Translation_Character(The_Group(i).b);
    exit when this is the last actual datum excluding padding;
    Emit(Translation_Character(The_Group(i).c);
    exit when this is the last actual datum excluding padding;
    Emit(Translation_Character(The_Group(i).d);
    exit when this is the last actual datum excluding padding;
  end loop;
  si := si+To_Encode'length;
end loop;

That's large and obscure.  Is there a better way?