File: srfi-60.html

package info (click to toggle)
drscheme 1%3A352-6
links: PTS
area: main
in suites: etch, etch-m68k
size: 71,608 kB
ctags: 55,284
sloc: ansic: 278,966; cpp: 63,318; sh: 32,265; lisp: 14,530; asm: 7,327; makefile: 4,846; pascal: 4,363; perl: 2,920; java: 1,632; yacc: 755; lex: 258; sed: 93; xml: 12
file content (884 lines) | stat: -rw-r--r-- 27,046 bytes
parent folder | download | duplicates (3)
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<HTML>
  <HEAD>
    <title>SRFI 60: Integers as Bits</title>
  </HEAD>

  <BODY>

<H1>Title</H1>

SRFI 60: Integers as Bits

<H1>Author</H1>

Aubrey Jaffer

<H1>Status</H1> 

This SRFI is currently in ``final'' status. To see an explanation of each
status that a SRFI can hold, see <A
href="http://srfi.schemers.org/srfi-process.html">here</A>. You can access
previous messages via <A
href="http://srfi.schemers.org/srfi-60/mail-archive/maillist.html">the
archive of the mailing list</A>.

<P> 
<UL>
  <LI>Received: <A HREF="http://srfi.schemers.org/cgi-bin/viewcvs.cgi/*checkout*/srfi/srfi-60/srfi-60.html?rev=1.1">2005/01/03</A></LI>
  <LI>Draft: 2005/01/03 - 2005/03/03</LI>
  <LI>Revised: <A HREF="http://srfi.schemers.org/cgi-bin/viewcvs.cgi/*checkout*/srfi/srfi-60/srfi-60.html?rev=1.2">2005/01/10</A></LI>
  <LI>Revised: <A HREF="http://srfi.schemers.org/cgi-bin/viewcvs.cgi/*checkout*/srfi/srfi-60/srfi-60.html?rev=1.4">2005/01/27</A></LI>
  <LI>Revised: <A HREF="http://srfi.schemers.org/cgi-bin/viewcvs.cgi/*checkout*/srfi/srfi-60/srfi-60.html?rev=1.5">2005/01/29</A></LI>
  <LI>Final: 2005/03/08</LI>
</UL>

<H1>Abstract</H1>

Treating integers as two's-complement strings of bits is an arcane but
important domain of computer science.  It is used for:

<UL>
<LI>hashing;

<LI>Galois-field[2] calculations of error-detecting and
    error-correcting codes;

<LI>cryptography and ciphers;

<LI>pseudo-random number generation;

<LI>register-transfer-level modeling of digital logic designs;

<LI>Fast-Fourier transforms;

<LI>packing and unpacking numbers in persistant data structures;

<LI>space-filling curves with applications to dimension reduction and
    sparse multi-dimensional database indexes; and

<LI>generating approximate seed values for root-finders and
    transcendental function algorithms.

</UL>
<P>

<H1>Rationale</H1>

This proposal describes the
<A HREF="http://swiss.csail.mit.edu/~jaffer/SLIB">SLIB</A> module
<A HREF="http://swiss.csail.mit.edu/~jaffer/slib_5.html#SEC88"><TT>logical</TT></A>,
which has been used for those purposes listed above.
<P>
The discussions of the withdrawn
<A HREF="http://srfi.schemers.org/srfi-33/">SRFI-33: "Integer
Bitwise-operation Library"</A> seemed to founder on consistency of
procedure names and arity; and on perceived competition with the
boolean arrays of SRFI-47.
<P>
I have implemented both logical number operations and boolean arrays;
and have not been conflicted as to their application.  I used boolean
arrays to construct very fast indexes for database tables having
millions of records.  To avoid running out of RAM, creation of megabit
arrays should be explicit; so the boolean array procedures put their
results into a passed array.  In contrast, these procedures are purely
functional.
<P>

<H3>Bits and Complements</H3>

A bit-index in these descriptions is nonnegative with the least
significant bit at index 0.
A positive integer has a finite number of "1" bits.
A negative integer has a finite number of "0" bits.
<P>
The reference implementation is written using only Scheme integer
operations.  Thus the only exposure of the underlying representation
is the ranges of fixnums.
<P>
The <DFN>complement</DFN> describes the representation of negative
integers.  With one's-complement fixnums, the range of integers is
-(2<SUP><I>n</I></SUP>) to 2<SUP><I>n</I></SUP>, and there are two
possible representations of 0.  With two's-complement fixnums, the
range of integers is -(2<SUP><I>n</I></SUP>+1) to
2<SUP><I>n</I></SUP>.
<P>
Since we treat integers as having two's-complement negations,
the two's-complement of an integer is simply its negation.
The one's-complement of an integer is computed by lognot:

<PRE>
(define (lognot n) (- -1 n))
</PRE>
<P>

<H3>Bitwise Operations and Integer Properties</H3>

The <TT>logior</TT>, <TT>logxor</TT>, <TT>logand</TT>,
<TT>lognot</TT>, <TT>logtest</TT>, <TT>logbit?</TT> (logbitp),
<TT>ash</TT>, <TT>logcount</TT>, and <TT>integer-length</TT>
procedures are from Common-Lisp.  <TT>Logior</TT>, <TT>logxor</TT>,
and <TT>logand</TT> have been extended to accept any arity.
Opportunities to use an <I>n</I>-ary version of <TT>logtest</TT> have
not been frequent enough to justify its extension.
<P>
In the <DFN>Bitwise Operations</DFN>, rather than striving for
orthogonal completeness, I have concentrated on a nearly minimal set
of bitwise logical functions sufficient to support the uses listed
above.
<P>
Although any two of <TT>logior</TT>, <TT>logxor</TT>, and
<TT>logand</TT> (in combination with <TT>lognot</TT>) are sufficient
to generate all the two-input logic functions, having these three
means that any nontrivial two-input logical function can be
synthesized using just one of these two-input primaries with zero or
one calls to <TT>lognot</TT>.
<P>
<TT>bitwise-if</TT> is what SRFI-33 calls <TT>bitwise-merge</TT>.
<P>
The SRFI-33 aliases: <TT>bitwise-ior</TT>, <TT>bitwise-xor</TT>,
<TT>bitwise-and</TT>, <TT>bitwise-not</TT>, <TT>bitwise-merge</TT>,
<TT>any-bits-set?</TT>, and <TT>bit-count</TT> are also provided.
<P>
<TT>log2-binary-factors</TT> (alias <TT>first-set-bit</TT>) is a
useful function which is simple but non-obvious:

<PRE>
(define (log2-binary-factors n)
  (+ -1 (integer-length (logand n (- n)))))
</PRE>

<H3>Bit Within Word and Field of Bits</H3>

The <DFN>Bit Within Word</DFN> and <DFN>Field of Bits</DFN> procedures
are used for modeling digital logic and accessing binary data
structures in software.
<P>
I have changed to <TT>copy-bit-field</TT> argument order to be
consistent with the other <DFN>Field of Bits</DFN> procedures: the
<VAR>start</VAR> and <VAR>end</VAR> index arguments are last.
This makes them analogous to the argument order to <TT>substring</TT>
and SRFI-47 arrays, which took their cue from <TT>substring</TT>.
<P>
These <VAR>start</VAR> and <VAR>end</VAR> index arguments are not
compatible with SRFI-33's <VAR>size</VAR> and <VAR>position</VAR>
arguments (occurring first) in its <TT>bit-field</TT> procedures.
Both define <TT>copy-bit-field</TT>; the arguments and purposes being
incompatible.
<P>
A procedure in slib/logical.scm, <TT>logical:rotate</TT>, rotated a
given number of low-order bits by a given number of bits.  This
function was quite servicable, but I could not name it adequately.  I
have replaced it with <TT>rotate-bit-field</TT> with the addition of a
<VAR>start</VAR> argument.  This new function rotates a given field
(from positions <VAR>start</VAR> to <VAR>end</VAR>) within an integer;
leaving the rest unchanged.
<P>
Another problematic name was <TT>logical:ones</TT>, which generated an
integer with the least significant <VAR>k</VAR> bits set.  Calls to
<TT>bit-field</TT> could have replaced its uses .  But the definition
was so short that I just replaced its uses with:

<PRE>
(lognot (ash -1 <VAR>k</VAR>))
</PRE>
<P>
The <TT>bit-reverse</TT> procedure was then the only one which took a
<VAR>width</VAR> argument.  So I replaced it with
<TT>reverse-bit-field</TT>.
<P>
The <DFN>Lamination</DFN> and <DFN>Gray-code</DFN> functions were
moved to
<A HREF="http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/slib/slib/phil-spc.scm?rev=HEAD&amp;content-type=text/vnd.viewcvs-markup">slib/phil-spc.scm</A>
<P>

<H3>Bits as Booleans</H3>

<DFN>Bits as Booleans</DFN> provides the procedures to convert between
integers and lists of booleans.  There is no comparable facility in
SRFI-33.
<P>


<H1>Specification</H1>



<H3><A NAME="SEC97">Bitwise Operations</A></H3>

<P>
<DL>
<DT><U>Function:</U> <B>logand</B> <I>n1 ...</I>
<DD><A NAME="IDX487"></A>
<DT><U>Function:</U> <B>bitwise-and</B> <I>n1 ...</I>
<DD><A NAME="IDX488"></A>
Returns the integer which is the bit-wise AND of the integer
arguments.


<P>
Example:

<PRE>
(number-&#62;string (logand #b1100 #b1010) 2)
    => "1000"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>logior</B> <I>n1 ...</I>
<DD><A NAME="IDX489"></A>
<DT><U>Function:</U> <B>bitwise-ior</B> <I>n1 ...</I>
<DD><A NAME="IDX490"></A>
Returns the integer which is the bit-wise OR of the integer arguments.


<P>
Example:

<PRE>
(number-&#62;string (logior #b1100 #b1010) 2)
    => "1110"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>logxor</B> <I>n1 ...</I>
<DD><A NAME="IDX491"></A>
<DT><U>Function:</U> <B>bitwise-xor</B> <I>n1 ...</I>
<DD><A NAME="IDX492"></A>
Returns the integer which is the bit-wise XOR of the integer
arguments.


<P>
Example:

<PRE>
(number-&#62;string (logxor #b1100 #b1010) 2)
    => "110"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>lognot</B> <I>n</I>
<DD><A NAME="IDX493"></A>
<DT><U>Function:</U> <B>bitwise-not</B> <I>n</I>
<DD><A NAME="IDX494"></A>
Returns the integer which is the one's-complement of the integer argument.


<P>
Example:

<PRE>
(number-&#62;string (lognot #b10000000) 2)
    => "-10000001"
(number-&#62;string (lognot #b0) 2)
    => "-1"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>bitwise-if</B> <I>mask n0 n1</I>
<DD><A NAME="IDX495"></A>
<DT><U>Function:</U> <B>bitwise-merge</B> <I>mask n0 n1</I>
<DD><A NAME="IDX496"></A>
Returns an integer composed of some bits from integer <VAR>n0</VAR> and some
from integer <VAR>n1</VAR>.  A bit of the result is taken from <VAR>n0</VAR> if the
corresponding bit of integer <VAR>mask</VAR> is 1 and from <VAR>n1</VAR> if that bit
of <VAR>mask</VAR> is 0.
</DL>


<P>
<DL>
<DT><U>Function:</U> <B>logtest</B> <I>j k</I>
<DD><A NAME="IDX497"></A>
<DT><U>Function:</U> <B>any-bits-set?</B> <I>j k</I>
<DD><A NAME="IDX498"></A>

<PRE>
(logtest j k) == (not (zero? (logand j k)))

(logtest #b0100 #b1011) => #f
(logtest #b0100 #b0111) => #t
</PRE>

</DL>



<H3><A NAME="SEC98">Integer Properties</A></H3>

<P>
<DL>
<DT><U>Function:</U> <B>logcount</B> <I>n</I>
<DD><A NAME="IDX499"></A>
<DT><U>Function:</U> <B>bit-count</B> <I>n</I>
<DD><A NAME="IDX500"></A>
Returns the number of bits in integer <VAR>n</VAR>.  If integer is positive,
the 1-bits in its binary representation are counted.  If negative, the
0-bits in its two's-complement binary representation are counted.  If 0,
0 is returned.


<P>
Example:

<PRE>
(logcount #b10101010)
    => 4
(logcount 0)
    => 0
(logcount -2)
    => 1
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>integer-length</B> <I>n</I>
<DD><A NAME="IDX501"></A>
Returns the number of bits neccessary to represent <VAR>n</VAR>.


<P>
Example:

<PRE>
(integer-length #b10101010)
    => 8
(integer-length 0)
    => 0
(integer-length #b1111)
    => 4
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>log2-binary-factors</B> <I>n</I>
<DD><A NAME="IDX502"></A>
<DT><U>Function:</U> <B>first-set-bit</B> <I>n</I>
<DD><A NAME="IDX503"></A>
Returns the number of factors of two of integer <VAR>n</VAR>.  This value
is also the bit-index of the least-significant <SAMP>`1'</SAMP> bit in
<VAR>n</VAR>.



<PRE>
(require 'printf)
(do ((idx 0 (+ 1 idx)))
      ((&#62; idx 16))
    (printf "%s(%3d) ==&#62; %-5d %s(%2d) ==&#62; %-5d\n"
            'log2-binary-factors
            (- idx) (log2-binary-factors (- idx))
            'log2-binary-factors
            idx (log2-binary-factors idx)))
-|
log2-binary-factors(  0) ==&#62; -1    log2-binary-factors( 0) ==&#62; -1   
log2-binary-factors( -1) ==&#62; 0     log2-binary-factors( 1) ==&#62; 0    
log2-binary-factors( -2) ==&#62; 1     log2-binary-factors( 2) ==&#62; 1    
log2-binary-factors( -3) ==&#62; 0     log2-binary-factors( 3) ==&#62; 0    
log2-binary-factors( -4) ==&#62; 2     log2-binary-factors( 4) ==&#62; 2    
log2-binary-factors( -5) ==&#62; 0     log2-binary-factors( 5) ==&#62; 0    
log2-binary-factors( -6) ==&#62; 1     log2-binary-factors( 6) ==&#62; 1    
log2-binary-factors( -7) ==&#62; 0     log2-binary-factors( 7) ==&#62; 0    
log2-binary-factors( -8) ==&#62; 3     log2-binary-factors( 8) ==&#62; 3    
log2-binary-factors( -9) ==&#62; 0     log2-binary-factors( 9) ==&#62; 0    
log2-binary-factors(-10) ==&#62; 1     log2-binary-factors(10) ==&#62; 1    
log2-binary-factors(-11) ==&#62; 0     log2-binary-factors(11) ==&#62; 0    
log2-binary-factors(-12) ==&#62; 2     log2-binary-factors(12) ==&#62; 2    
log2-binary-factors(-13) ==&#62; 0     log2-binary-factors(13) ==&#62; 0    
log2-binary-factors(-14) ==&#62; 1     log2-binary-factors(14) ==&#62; 1    
log2-binary-factors(-15) ==&#62; 0     log2-binary-factors(15) ==&#62; 0    
log2-binary-factors(-16) ==&#62; 4     log2-binary-factors(16) ==&#62; 4    
</PRE>

</DL>



<H3><A NAME="SEC99">Bit Within Word</A></H3>

<P>
<DL>
<DT><U>Function:</U> <B>logbit?</B> <I>index n</I>
<DD><A NAME="IDX504"></A>
<DT><U>Function:</U> <B>bit-set?</B> <I>index n</I>
<DD><A NAME="IDX505"></A>

<PRE>
(logbit? index n) == (logtest (expt 2 index) n)

(logbit? 0 #b1101) => #t
(logbit? 1 #b1101) => #f
(logbit? 2 #b1101) => #t
(logbit? 3 #b1101) => #t
(logbit? 4 #b1101) => #f
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>copy-bit</B> <I>index from bit</I>
<DD><A NAME="IDX506"></A>
Returns an integer the same as <VAR>from</VAR> except in the <VAR>index</VAR>th bit,
which is 1 if <VAR>bit</VAR> is <CODE>#t</CODE> and 0 if <VAR>bit</VAR> is <CODE>#f</CODE>.


<P>
Example:

<PRE>
(number-&#62;string (copy-bit 0 0 #t) 2)       => "1"
(number-&#62;string (copy-bit 2 0 #t) 2)       => "100"
(number-&#62;string (copy-bit 2 #b1111 #f) 2)  => "1011"
</PRE>

</DL>



<H3><A NAME="SEC100">Field of Bits</A></H3>

<P>
<DL>
<DT><U>Function:</U> <B>bit-field</B> <I>n start end</I>
<DD><A NAME="IDX507"></A>
Returns the integer composed of the <VAR>start</VAR> (inclusive) through
<VAR>end</VAR> (exclusive) bits of <VAR>n</VAR>.  The <VAR>start</VAR>th bit becomes
the 0-th bit in the result.


<P>
Example:

<PRE>
(number-&#62;string (bit-field #b1101101010 0 4) 2)
    => "1010"
(number-&#62;string (bit-field #b1101101010 4 9) 2)
    => "10110"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>copy-bit-field</B> <I>to from start end</I>
<DD><A NAME="IDX508"></A>
Returns an integer the same as <VAR>to</VAR> except possibly in the
<VAR>start</VAR> (inclusive) through <VAR>end</VAR> (exclusive) bits, which are
the same as those of <VAR>from</VAR>.  The 0-th bit of <VAR>from</VAR> becomes the
<VAR>start</VAR>th bit of the result.


<P>
Example:

<PRE>
(number-&#62;string (copy-bit-field #b1101101010 0 0 4) 2)
    => "1101100000"
(number-&#62;string (copy-bit-field #b1101101010 -1 0 4) 2)
    => "1101101111"
(number-&#62;string (copy-bit-field #b110100100010000 -1 5 9) 2)
    => "110100111110000"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>ash</B> <I>n count</I>
<DD><A NAME="IDX509"></A>
<DT><U>Function:</U> <B>arithmetic-shift</B> <I>n count</I>
<DD><A NAME="IDX510"></A>
Returns an integer equivalent to
<CODE>(inexact-&#62;exact (floor (* <VAR>n</VAR> (expt 2 <VAR>count</VAR>))))</CODE>.


<P>
Example:

<PRE>
(number-&#62;string (ash #b1 3) 2)
    => "1000"
(number-&#62;string (ash #b1010 -1) 2)
    => "101"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>rotate-bit-field</B> <I>n count start end</I>
<DD><A NAME="IDX511"></A>
Returns <VAR>n</VAR> with the bit-field from <VAR>start</VAR> to <VAR>end</VAR>
cyclically permuted by <VAR>count</VAR> bits towards high-order.


<P>
Example:

<PRE>
(number-&#62;string (rotate-bit-field #b0100 3 0 4) 2)
    => "10"
(number-&#62;string (rotate-bit-field #b0100 -1 0 4) 2)
    => "10"
(number-&#62;string (rotate-bit-field #b110100100010000 -1 5 9) 2)
    => "110100010010000"
(number-&#62;string (rotate-bit-field #b110100100010000 1 5 9) 2)
    => "110100000110000"
</PRE>

</DL>

<P>
<DL>
<DT><U>Function:</U> <B>reverse-bit-field</B> <I>n start end</I>
<DD><A NAME="IDX512"></A>
Returns <VAR>n</VAR> with the order of bits <VAR>start</VAR> to <VAR>end</VAR>
reversed.



<PRE>
(number-&#62;string (reverse-bit-field #xa7 0 8) 16)
    => "e5"
</PRE>

</DL>



<H3><A NAME="SEC101">Bits as Booleans</A></H3>

<P>
<DL>
<DT><U>Function:</U> <B>integer-&#62;list</B> <I>k len</I>
<DD><A NAME="IDX513"></A>
<DT><U>Function:</U> <B>integer-&#62;list</B> <I>k</I>
<DD><A NAME="IDX514"></A>
<CODE>integer-&#62;list</CODE> returns a list of <VAR>len</VAR> booleans corresponding
to each bit of the given integer.  #t is coded for each 1; #f for 0.
The <VAR>len</VAR> argument defaults to <CODE>(integer-length <VAR>k</VAR>)</CODE>.


<P>
<DT><U>Function:</U> <B>list-&#62;integer</B> <I>list</I>
<DD><A NAME="IDX515"></A>
<CODE>list-&#62;integer</CODE> returns an integer formed from the booleans in the
list <VAR>list</VAR>, which must be a list of booleans.  A 1 bit is coded for
each #t; a 0 bit for #f.


<P>
<CODE>integer-&#62;list</CODE> and <CODE>list-&#62;integer</CODE> are inverses so far as
<CODE>equal?</CODE> is concerned.
</DL>


<P>
<DL>
<DT><U>Function:</U> <B>booleans-&#62;integer</B> <I>bool1 ...</I>
<DD><A NAME="IDX516"></A>
Returns the integer coded by the <VAR>bool1</VAR> ... arguments.
</DL>



<H1>Implementation</H1>

<A HREF="http://savannah.gnu.org/cgi-bin/viewcvs/*checkout*/slib/slib/logical.scm?rev=HEAD&amp;content-type=text/vnd.viewcvs-markup">slib/logical.scm</A>
implements the integers-as-bits procedures for R4RS or R5RS compliant
Scheme implementations.
<P>
<PRE>
;;;; "logical.scm", bit access and operations for integers for Scheme
;;; Copyright (C) 1991, 1993, 2001, 2003, 2005 Aubrey Jaffer
;
;Permission to copy this software, to modify it, to redistribute it,
;to distribute modified versions, and to use it for any purpose is
;granted, subject to the following restrictions and understandings.
;
;1.  Any copy made of this software must include this copyright notice
;in full.
;
;2.  I have made no warranty or representation that the operation of
;this software will be error-free, and I am under no obligation to
;provide any services, by way of maintenance, update, or otherwise.
;
;3.  In conjunction with products arising from the use of this
;material, there shall be no use of my name in any advertising,
;promotional, or sales literature without prior written consent in
;each case.

(define logical:boole-xor
 '#(#(0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15)
    #(1 0 3 2 5 4 7 6 9 8 11 10 13 12 15 14)
    #(2 3 0 1 6 7 4 5 10 11 8 9 14 15 12 13)
    #(3 2 1 0 7 6 5 4 11 10 9 8 15 14 13 12)
    #(4 5 6 7 0 1 2 3 12 13 14 15 8 9 10 11)
    #(5 4 7 6 1 0 3 2 13 12 15 14 9 8 11 10)
    #(6 7 4 5 2 3 0 1 14 15 12 13 10 11 8 9)
    #(7 6 5 4 3 2 1 0 15 14 13 12 11 10 9 8)
    #(8 9 10 11 12 13 14 15 0 1 2 3 4 5 6 7)
    #(9 8 11 10 13 12 15 14 1 0 3 2 5 4 7 6)
    #(10 11 8 9 14 15 12 13 2 3 0 1 6 7 4 5)
    #(11 10 9 8 15 14 13 12 3 2 1 0 7 6 5 4)
    #(12 13 14 15 8 9 10 11 4 5 6 7 0 1 2 3)
    #(13 12 15 14 9 8 11 10 5 4 7 6 1 0 3 2)
    #(14 15 12 13 10 11 8 9 6 7 4 5 2 3 0 1)
    #(15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0)))

(define logical:boole-and
 '#(#(0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0)
    #(0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1)
    #(0 0 2 2 0 0 2 2 0 0 2 2 0 0 2 2)
    #(0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3)
    #(0 0 0 0 4 4 4 4 0 0 0 0 4 4 4 4)
    #(0 1 0 1 4 5 4 5 0 1 0 1 4 5 4 5)
    #(0 0 2 2 4 4 6 6 0 0 2 2 4 4 6 6)
    #(0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7)
    #(0 0 0 0 0 0 0 0 8 8 8 8 8 8 8 8)
    #(0 1 0 1 0 1 0 1 8 9 8 9 8 9 8 9)
    #(0 0 2 2 0 0 2 2 8 8 10 10 8 8 10 10)
    #(0 1 2 3 0 1 2 3 8 9 10 11 8 9 10 11)
    #(0 0 0 0 4 4 4 4 8 8 8 8 12 12 12 12)
    #(0 1 0 1 4 5 4 5 8 9 8 9 12 13 12 13)
    #(0 0 2 2 4 4 6 6 8 8 10 10 12 12 14 14)
    #(0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15)))

(define (logical:ash-4 x)
  (if (negative? x)
      (+ -1 (quotient (+ 1 x) 16))
      (quotient x 16)))

(define (logical:reduce op4 ident)
  (lambda args
    (do ((res ident (op4 res (car rgs) 1 0))
         (rgs args (cdr rgs)))
        ((null? rgs) res))))

;@
(define logand
  (letrec
      ((lgand
        (lambda (n2 n1 scl acc)
          (cond ((= n1 n2) (+ acc (* scl n1)))
                ((zero? n2) acc)
                ((zero? n1) acc)
                (else (lgand (logical:ash-4 n2)
                             (logical:ash-4 n1)
                             (* 16 scl)
                             (+ (* (vector-ref (vector-ref logical:boole-and
                                                           (modulo n1 16))
                                               (modulo n2 16))
                                   scl)
                                acc)))))))
    (logical:reduce lgand -1)))
;@
(define logior
  (letrec
      ((lgior
        (lambda (n2 n1 scl acc)
          (cond ((= n1 n2) (+ acc (* scl n1)))
                ((zero? n2) (+ acc (* scl n1)))
                ((zero? n1) (+ acc (* scl n2)))
                (else (lgior (logical:ash-4 n2)
                             (logical:ash-4 n1)
                             (* 16 scl)
                             (+ (* (- 15 (vector-ref
                                          (vector-ref logical:boole-and
                                                      (- 15 (modulo n1 16)))
                                          (- 15 (modulo n2 16))))
                                   scl)
                                acc)))))))
    (logical:reduce lgior 0)))
;@
(define logxor
  (letrec
      ((lgxor
        (lambda (n2 n1 scl acc)
          (cond ((= n1 n2) acc)
                ((zero? n2) (+ acc (* scl n1)))
                ((zero? n1) (+ acc (* scl n2)))
                (else (lgxor (logical:ash-4 n2)
                             (logical:ash-4 n1)
                             (* 16 scl)
                             (+ (* (vector-ref (vector-ref logical:boole-xor
                                                           (modulo n1 16))
                                               (modulo n2 16))
                                   scl)
                                acc)))))))
    (logical:reduce lgxor 0)))
;@
(define (lognot n) (- -1 n))
;@
(define (logtest n1 n2)
  (not (zero? (logand n1 n2))))
;@
(define (logbit? index n)
  (logtest (expt 2 index) n))
;@
(define (copy-bit index to bool)
  (if bool
      (logior to (arithmetic-shift 1 index))
      (logand to (lognot (arithmetic-shift 1 index)))))
;@
(define (bitwise-if mask n0 n1)
  (logior (logand mask n0)
          (logand (lognot mask) n1)))
;@
(define (bit-field n start end)
  (logand (lognot (ash -1 (- end start)))
          (arithmetic-shift n (- start))))
;@
(define (copy-bit-field to from start end)
  (bitwise-if (arithmetic-shift (lognot (ash -1 (- end start))) start)
              (arithmetic-shift from start)
              to))
;@
(define (rotate-bit-field n count start end)
  (define width (- end start))
  (set! count (modulo count width))
  (let ((mask (lognot (ash -1 width))))
    (define zn (logand mask (arithmetic-shift n (- start))))
    (logior (arithmetic-shift
             (logior (logand mask (arithmetic-shift zn count))
                     (arithmetic-shift zn (- count width)))
             start)
            (logand (lognot (ash mask start)) n))))
;@
(define (arithmetic-shift n count)
  (if (negative? count)
      (let ((k (expt 2 (- count))))
        (if (negative? n)
            (+ -1 (quotient (+ 1 n) k))
            (quotient n k)))
      (* (expt 2 count) n)))
;@
(define integer-length
  (letrec ((intlen (lambda (n tot)
                     (case n
                       ((0 -1) (+ 0 tot))
                       ((1 -2) (+ 1 tot))
                       ((2 3 -3 -4) (+ 2 tot))
                       ((4 5 6 7 -5 -6 -7 -8) (+ 3 tot))
                       (else (intlen (logical:ash-4 n) (+ 4 tot)))))))
    (lambda (n) (intlen n 0))))
;@
(define logcount
  (letrec ((logcnt (lambda (n tot)
                     (if (zero? n)
                         tot
                         (logcnt (quotient n 16)
                                 (+ (vector-ref
                                     '#(0 1 1 2 1 2 2 3 1 2 2 3 2 3 3 4)
                                     (modulo n 16))
                                    tot))))))
    (lambda (n)
      (cond ((negative? n) (logcnt (lognot n) 0))
            ((positive? n) (logcnt n 0))
            (else 0)))))
;@
(define (log2-binary-factors n)
  (+ -1 (integer-length (logand n (- n)))))

(define (bit-reverse k n)
  (do ((m (if (negative? n) (lognot n) n) (arithmetic-shift m -1))
       (k (+ -1 k) (+ -1 k))
       (rvs 0 (logior (arithmetic-shift rvs 1) (logand 1 m))))
      ((negative? k) (if (negative? n) (lognot rvs) rvs))))
;@
(define (reverse-bit-field n start end)
  (define width (- end start))
  (let ((mask (lognot (ash -1 width))))
    (define zn (logand mask (arithmetic-shift n (- start))))
    (logior (arithmetic-shift (bit-reverse width zn) start)
            (logand (lognot (ash mask start)) n))))
;@
(define (integer-&gt;list k . len)
  (if (null? len)
      (do ((k k (arithmetic-shift k -1))
           (lst '() (cons (odd? k) lst)))
          ((&lt;= k 0) lst))
      (do ((idx (+ -1 (car len)) (+ -1 idx))
           (k k (arithmetic-shift k -1))
           (lst '() (cons (odd? k) lst)))
          ((negative? idx) lst))))
;@
(define (list-&gt;integer bools)
  (do ((bs bools (cdr bs))
       (acc 0 (+ acc acc (if (car bs) 1 0))))
      ((null? bs) acc)))
(define (booleans-&gt;integer . bools)
  (list-&gt;integer bools))

;;;;@ SRFI-60 aliases
(define ash arithmetic-shift)
(define bitwise-ior logior)
(define bitwise-xor logxor)
(define bitwise-and logand)
(define bitwise-not lognot)
(define bit-count logcount)
(define bit-set?   logbit?)
(define any-bits-set? logtest)
(define first-set-bit log2-binary-factors)
(define bitwise-merge bitwise-if)

;;; Legacy
;;(define (logical:rotate k count len) (rotate-bit-field k count 0 len))
;;(define (logical:ones deg) (lognot (ash -1 deg)))
;;(define integer-expt expt)            ; legacy name
</PRE>


<H1>Copyright</H1>
<p>Copyright (C) Aubrey Jaffer (2004, 2005). All Rights Reserved.</p>

<p>
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
</p>
<p>
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
</p>
<p>
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
</p>


    <HR>
    <ADDRESS>Editor: <A HREF="mailto:srfi-editors@srfi.schemers.org">David Van Horn</A></ADDRESS>
<!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
<!-- hhmts start -->
Last modified: Sat Jan 29 13:16:05 EST 2005
<!-- hhmts end -->
  </BODY>
</HTML>