\documentclass{article}
\usepackage{open-axiom}
\begin{document}
\title{src/algebra aggcat.spad}
\author{Michael Monagan, Manuel Bronstein, Richard Jenks, Stephen Watt}
\maketitle

\begin{abstract}
\end{abstract}

\tableofcontents
\eject

\section{category AGG Aggregate}

<<category AGG Aggregate>>=
import Type
import Boolean
import NonNegativeInteger
)abbrev category AGG Aggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ The notion of aggregate serves to model any data structure aggregate,
++ designating any collection of objects,
++ with heterogenous or homogeneous members,
++ with a finite or infinite number
++ of members, explicitly or implicitly represented.
++ An aggregate can in principle
++ represent everything from a string of characters to abstract sets such
++ as "the set of x satisfying relation {\em r(x)}"
++ An attribute \spadatt{finiteAggregate} is used to assert that a domain
++ has a finite number of elements.
Aggregate: Category == Type with
   eq?: (%,%) -> Boolean
     ++ eq?(u,v) tests if u and v are same objects.
   copy: % -> %
     ++ copy(u) returns a top-level (non-recursive) copy of u.
     ++ Note: for collections, \axiom{copy(u) == [x for x in u]}.
   empty: () -> %
     ++ empty()$D creates an aggregate of type D with 0 elements.
     ++ Note: The {\em $D} can be dropped if understood by context,
     ++ e.g. \axiom{u: D := empty()}.
   empty?: % -> Boolean
     ++ empty?(u) tests if u has 0 elements.
   less?: (%,NonNegativeInteger) -> Boolean
     ++ less?(u,n) tests if u has less than n elements.
   more?: (%,NonNegativeInteger) -> Boolean
     ++ more?(u,n) tests if u has greater than n elements.
   size?: (%,NonNegativeInteger) -> Boolean
     ++ size?(u,n) tests if u has exactly n elements.
   sample: constant -> %    ++ sample yields a value of type %
   if % has finiteAggregate then
     #: % -> NonNegativeInteger     ++ # u returns the number of items in u.
 add
  eq?(a,b) == %peq(a,b)$Foreign(Builtin)
  sample() == empty()
  if % has finiteAggregate then
    empty? a   == #a = 0
    less?(a,n) == #a < n
    more?(a,n) == #a > n
    size?(a,n) == #a = n

@

\section{category HOAGG HomogeneousAggregate}

<<category HOAGG HomogeneousAggregate>>=
import Boolean
import OutputForm
import SetCategory
import Aggregate
import CoercibleTo OutputForm
import Evalable
)abbrev category HOAGG HomogeneousAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991, May 1995
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A homogeneous aggregate is an aggregate of elements all of the
++ same type.
++ In the current system, all aggregates are homogeneous.
++ Two attributes characterize classes of aggregates.
++ Aggregates from domains with attribute \spadatt{finiteAggregate}
++ have a finite number of members.
++ Those with attribute \spadatt{shallowlyMutable} allow an element
++ to be modified or updated without changing its overall value.
HomogeneousAggregate(S:Type): Category == Aggregate with
   if S has CoercibleTo(OutputForm) then CoercibleTo(OutputForm)
   if S has BasicType then BasicType
   if S has SetCategory then SetCategory
   if S has SetCategory then
      if S has Evalable S then Evalable S
   map	   : (S->S,%) -> %
     ++ map(f,u) returns a copy of u with each element x replaced by f(x).
     ++ For collections, \axiom{map(f,u) = [f(x) for x in u]}.
   if % has shallowlyMutable then
     map!: (S->S,%) -> %
	++ map!(f,u) destructively replaces each element x of u by \axiom{f(x)}.
   if % has finiteAggregate then
      any?: (S->Boolean,%) -> Boolean
	++ any?(p,u) tests if \axiom{p(x)} is true for any element x of u.
	++ Note: for collections,
	++ \axiom{any?(p,u) = reduce(or,map(f,u),false,true)}.
      every?: (S->Boolean,%) -> Boolean
	++ every?(f,u) tests if p(x) is true for all elements x of u.
	++ Note: for collections,
	++ \axiom{every?(p,u) = reduce(and,map(f,u),true,false)}.
      count: (S->Boolean,%) -> NonNegativeInteger
	++ count(p,u) returns the number of elements x in u
	++ such that \axiom{p(x)} is true. For collections,
	++ \axiom{count(p,u) = reduce(+,[1 for x in u | p(x)],0)}.
      parts: % -> List S
	++ parts(u) returns a list of the consecutive elements of u.
	++ For collections, \axiom{parts([x,y,...,z]) = (x,y,...,z)}.
      members: % -> List S
	++ members(u) returns a list of the consecutive elements of u.
	++ For collections, \axiom{parts([x,y,...,z]) = (x,y,...,z)}.
      if S has SetCategory then
	count: (S,%) -> NonNegativeInteger
	  ++ count(x,u) returns the number of occurrences of x in u.
	  ++ For collections, \axiom{count(x,u) = reduce(+,[x=y for y in u],0)}.
	member?: (S,%) -> Boolean
	  ++ member?(x,u) tests if x is a member of u.
	  ++ For collections,
	  ++ \axiom{member?(x,u) = reduce(or,[x=y for y in u],false)}.
  add
   if S has Evalable S then
     eval(u:%,l:List Equation S):% == map(eval(#1,l),u)

   if % has finiteAggregate then
     #c			  == # parts c
     any?(f, c)		  == or/[f x for x in parts c]
     every?(f, c)	  == and/[f x for x in parts c]
     count(f:S -> Boolean, c:%) == +/[1 for x in parts c | f x]
     members x		  == parts x

     if S has BasicType then
       x = y ==
         size?(x, #y) and (and/[a = b for a in parts x for b in parts y])

     if S has SetCategory then
       count(s:S, x:%) == count(s = #1, x)
       member?(e, c)   == any?(e = #1,c)

     if S has CoercibleTo(OutputForm) then
       coerce(x:%):OutputForm ==
	 bracket
	    commaSeparate [a::OutputForm for a in parts x]$List(OutputForm)

@

\section{category CLAGG Collection}

<<category CLAGG Collection>>=
import Boolean
import HomogeneousAggregate
)abbrev category CLAGG Collection
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A collection is a homogeneous aggregate which can built from
++ list of members. The operation used to build the aggregate is
++ generically named \spadfun{construct}. However, each collection
++ provides its own special function with the same name as the
++ data type, except with an initial lower case letter, e.g.
++ \spadfun{list} for \spadtype{List},
++ \spadfun{flexibleArray} for \spadtype{FlexibleArray}, and so on.
Collection(S:Type): Category == HomogeneousAggregate(S) with
   construct: List S -> %
     ++ \axiom{construct(x,y,...,z)} returns the collection of elements \axiom{x,y,...,z}
     ++ ordered as given. Equivalently written as \axiom{[x,y,...,z]$D}, where
     ++ D is the domain. D may be omitted for those of type List.
   find: (S->Boolean, %) -> Union(S, "failed")
     ++ find(p,u) returns the first x in u such that \axiom{p(x)} is true, and
     ++ "failed" otherwise.
   if % has finiteAggregate then
      reduce: ((S,S)->S,%) -> S
	++ reduce(f,u) reduces the binary operation f across u. For example,
	++ if u is \axiom{[x,y,...,z]} then \axiom{reduce(f,u)} returns \axiom{f(..f(f(x,y),...),z)}.
	++ Note: if u has one element x, \axiom{reduce(f,u)} returns x.
	++ Error: if u is empty.
      reduce: ((S,S)->S,%,S) -> S
	++ reduce(f,u,x) reduces the binary operation f across u, where x is
	++ the identity operation of f.
	++ Same as \axiom{reduce(f,u)} if u has 2 or more elements.
	++ Returns \axiom{f(x,y)} if u has one element y,
	++ x if u is empty.
	++ For example, \axiom{reduce(+,u,0)} returns the
	++ sum of the elements of u.
      remove: (S->Boolean,%) -> %
	++ remove(p,u) returns a copy of u removing all elements x such that
	++ \axiom{p(x)} is true.
	++ Note: \axiom{remove(p,u) == [x for x in u | not p(x)]}.
      select: (S->Boolean,%) -> %
	++ select(p,u) returns a copy of u containing only those elements such
	++ \axiom{p(x)} is true.
	++ Note: \axiom{select(p,u) == [x for x in u | p(x)]}.
      if S has SetCategory then
	reduce: ((S,S)->S,%,S,S) -> S
	  ++ reduce(f,u,x,z) reduces the binary operation f across u, stopping
	  ++ when an "absorbing element" z is encountered.
	  ++ As for \axiom{reduce(f,u,x)}, x is the identity operation of f.
	  ++ Same as \axiom{reduce(f,u,x)} when u contains no element z.
	  ++ Thus the third argument x is returned when u is empty.
	remove: (S,%) -> %
	  ++ remove(x,u) returns a copy of u with all
	  ++ elements \axiom{y = x} removed.
	  ++ Note: \axiom{remove(y,c) == [x for x in c | x ~= y]}.
	removeDuplicates: % -> %
	  ++ removeDuplicates(u) returns a copy of u with all duplicates removed.
   if S has ConvertibleTo InputForm then ConvertibleTo InputForm
 add
   if % has finiteAggregate then
     #c			  == # parts c
     count(f:S -> Boolean, c:%) == +/[1 for x in parts c | f x]
     any?(f, c)		  == or/[f x for x in parts c]
     every?(f, c)	  == and/[f x for x in parts c]
     find(f:S -> Boolean, c:%) == find(f, parts c)
     reduce(f:(S,S)->S, x:%) == reduce(f, parts x)
     reduce(f:(S,S)->S, x:%, s:S) == reduce(f, parts x, s)
     remove(f:S->Boolean, x:%) ==
       construct remove(f, parts x)
     select(f:S->Boolean, x:%) ==
       construct select(f, parts x)

     if S has SetCategory then
       remove(s:S, x:%) == remove(#1 = s, x)
       reduce(f:(S,S)->S, x:%, s1:S, s2:S) == reduce(f, parts x, s1, s2)
       removeDuplicates(x) == construct removeDuplicates parts x

@

\section{category BGAGG BagAggregate}

<<category BGAGG BagAggregate>>=
import HomogeneousAggregate
import List
)abbrev category BGAGG BagAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A bag aggregate is an aggregate for which one can insert and extract objects,
++ and where the order in which objects are inserted determines the order
++ of extraction.
++ Examples of bags are stacks, queues, and dequeues.
BagAggregate(S:Type): Category == HomogeneousAggregate S with
   shallowlyMutable
     ++ shallowlyMutable means that elements of bags may be destructively changed.
   bag: List S -> %
     ++ bag([x,y,...,z]) creates a bag with elements x,y,...,z.
   extract!: % -> S
     ++ extract!(u) destructively removes a (random) item from bag u.
   insert!: (S,%) -> %
     ++ insert!(x,u) inserts item x into bag u.
   inspect: % -> S
     ++ inspect(u) returns an (random) element from a bag.
 add
   bag(l) ==
     x:=empty()
     for s in l repeat x:=insert!(s,x)
     x

@

\section{category SKAGG StackAggregate}

<<category SKAGG StackAggregate>>=
import NonNegativeInteger
import BagAggregate
)abbrev category SKAGG StackAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A stack is a bag where the last item inserted is the first item extracted.
StackAggregate(S:Type): Category == BagAggregate S with
   finiteAggregate
   push!: (S,%) -> S
     ++ push!(x,s) pushes x onto stack s, i.e. destructively changing s
     ++ so as to have a new first (top) element x.
     ++ Afterwards, pop!(s) produces x and pop!(s) produces the original s.
   pop!: % -> S
     ++ pop!(s) returns the top element x, destructively removing x from s.
     ++ Note: Use \axiom{top(s)} to obtain x without removing it from s.
     ++ Error: if s is empty.
   top: % -> S
     ++ top(s) returns the top element x from s; s remains unchanged.
     ++ Note: Use \axiom{pop!(s)} to obtain x and remove it from s.
   depth: % -> NonNegativeInteger
     ++ depth(s) returns the number of elements of stack s.
     ++ Note: \axiom{depth(s) = #s}.


@

\section{category QUAGG QueueAggregate}

<<category QUAGG QueueAggregate>>=
import NonNegativeInteger
import BagAggregate
)abbrev category QUAGG QueueAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A queue is a bag where the first item inserted is the first item extracted.
QueueAggregate(S:Type): Category == BagAggregate S with
   finiteAggregate
   enqueue!: (S, %) -> S
     ++ enqueue!(x,q) inserts x into the queue q at the back end.
   dequeue!: % -> S
     ++ dequeue! s destructively extracts the first (top) element from queue q.
     ++ The element previously second in the queue becomes the first element.
     ++ Error: if q is empty.
   rotate!: % -> %
     ++ rotate! q rotates queue q so that the element at the front of
     ++ the queue goes to the back of the queue.
     ++ Note: rotate! q is equivalent to enqueue!(dequeue!(q)).
   length: % -> NonNegativeInteger
     ++ length(q) returns the number of elements in the queue.
     ++ Note: \axiom{length(q) = #q}.
   front: % -> S
     ++ front(q) returns the element at the front of the queue.
     ++ The queue q is unchanged by this operation.
     ++ Error: if q is empty.
   back: % -> S
     ++ back(q) returns the element at the back of the queue.
     ++ The queue q is unchanged by this operation.
     ++ Error: if q is empty.

@

\section{category DQAGG DequeueAggregate}

<<category DQAGG DequeueAggregate>>=
import StackAggregate
import QueueAggregate
)abbrev category DQAGG DequeueAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A dequeue is a doubly ended stack, that is, a bag where first items
++ inserted are the first items extracted, at either the front or the back end
++ of the data structure.
DequeueAggregate(S:Type):
 Category == Join(StackAggregate S,QueueAggregate S) with
   dequeue: () -> %
     ++ dequeue()$D creates an empty dequeue of type D.
   dequeue: List S -> %
     ++ dequeue([x,y,...,z]) creates a dequeue with first (top or front)
     ++ element x, second element y,...,and last (bottom or back) element z.
   height: % -> NonNegativeInteger
     ++ height(d) returns the number of elements in dequeue d.
     ++ Note: \axiom{height(d) = # d}.
   top!: % -> S
     ++ top!(d) returns the element at the top (front) of the dequeue.
   bottom!: % -> S
     ++ bottom!(d) returns the element at the bottom (back) of the dequeue.
   insertTop!: (S,%) -> S
     ++ insertTop!(x,d) destructively inserts x into the dequeue d, that is,
     ++ at the top (front) of the dequeue.
     ++ The element previously at the top of the dequeue becomes the
     ++ second in the dequeue, and so on.
   insertBottom!: (S,%) -> S
     ++ insertBottom!(x,d) destructively inserts x into the dequeue d
     ++ at the bottom (back) of the dequeue.
   extractTop!: % -> S
     ++ extractTop!(d) destructively extracts the top (front) element
     ++ from the dequeue d.
     ++ Error: if d is empty.
   extractBottom!: % -> S
     ++ extractBottom!(d) destructively extracts the bottom (back) element
     ++ from the dequeue d.
     ++ Error: if d is empty.
   reverse!: % -> %
     ++ reverse!(d) destructively replaces d by its reverse dequeue, i.e.
     ++ the top (front) element is now the bottom (back) element, and so on.

@

\section{category PRQAGG PriorityQueueAggregate}

<<category PRQAGG PriorityQueueAggregate>>=
import BagAggregate
)abbrev category PRQAGG PriorityQueueAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A priority queue is a bag of items from an ordered set where the item
++ extracted is always the maximum element.
PriorityQueueAggregate(S:OrderedSet): Category == BagAggregate S with
   finiteAggregate
   max: % -> S
     ++ max(q) returns the maximum element of priority queue q.
   merge: (%,%) -> %
     ++ merge(q1,q2) returns combines priority queues q1 and q2 to return
     ++ a single priority queue q.
   merge!: (%,%) -> %
     ++ merge!(q,q1) destructively changes priority queue q to include the
     ++ values from priority queue q1.

@

\section{category DIOPS DictionaryOperations}
<<category DIOPS DictionaryOperations>>=
import Boolean
import Collection
import BagAggregate
)abbrev category DIOPS DictionaryOperations
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ This category is a collection of operations common to both
++ categories \spadtype{Dictionary} and \spadtype{MultiDictionary}
DictionaryOperations(S:SetCategory): Category ==
  Join(BagAggregate S, Collection(S)) with
   dictionary: () -> %
     ++ dictionary()$D creates an empty dictionary of type D.
   dictionary: List S -> %
     ++ dictionary([x,y,...,z]) creates a dictionary consisting of
     ++ entries \axiom{x,y,...,z}.
-- insert: (S,%) -> S		      ++ insert an entry
-- member?: (S,%) -> Boolean		      ++ search for an entry
-- remove!: (S,%,NonNegativeInteger) -> %
--   ++ remove!(x,d,n) destructively changes dictionary d by removing
--   ++ up to n entries y such that \axiom{y = x}.
-- remove!: (S->Boolean,%,NonNegativeInteger) -> %
--   ++ remove!(p,d,n) destructively changes dictionary d by removing
--   ++ up to n entries x such that \axiom{p(x)} is true.
   if % has finiteAggregate then
     remove!: (S,%) -> %
       ++ remove!(x,d) destructively changes dictionary d by removing
       ++ all entries y such that \axiom{y = x}.
     remove!: (S->Boolean,%) -> %
       ++ remove!(p,d) destructively changes dictionary d by removeing
       ++ all entries x such that \axiom{p(x)} is true.
     select!: (S->Boolean,%) -> %
       ++ select!(p,d) destructively changes dictionary d by removing
       ++ all entries x such that \axiom{p(x)} is not true.
 add
   construct l == dictionary l
   dictionary() == empty()
   if % has finiteAggregate then
     copy d == dictionary parts d
     coerce(s:%):OutputForm ==
       prefix("dictionary"@String :: OutputForm,
				      [x::OutputForm for x in parts s])

@

\section{category DIAGG Dictionary}

<<category DIAGG Dictionary>>=
import Boolean
import DictionaryOperations
)abbrev category DIAGG Dictionary
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A dictionary is an aggregate in which entries can be inserted,
++ searched for and removed. Duplicates are thrown away on insertion.
++ This category models the usual notion of dictionary which involves
++ large amounts of data where copying is impractical.
++ Principal operations are thus destructive (non-copying) ones.
Dictionary(S:SetCategory): Category ==
 DictionaryOperations S add
   dictionary l ==
     d := dictionary()
     for x in l repeat insert!(x, d)
     d

   if % has finiteAggregate then
    -- remove(f:S->Boolean,t:%)  == remove!(f, copy t)
    -- select(f, t)	   == select!(f, copy t)
     select!(f, t)	 == remove!(not f #1, t)

     --extract! d ==
     --	 empty? d => error "empty dictionary"
     --	 remove!(x := first parts d, d, 1)
     --	 x

     s = t ==
       eq?(s,t) => true
       #s ~= #t => false
       and/[member?(x, t) for x in parts s]

     remove!(f:S->Boolean, t:%) ==
       for m in parts t repeat if f m then remove!(m, t)
       t

@

\section{category MDAGG MultiDictionary}

<<category MDAGG MultiDictionary>>=
import NonNegativeInteger
import DictionaryOperations
)abbrev category MDAGG MultiDictionary
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A multi-dictionary is a dictionary which may contain duplicates.
++ As for any dictionary, its size is assumed large so that
++ copying (non-destructive) operations are generally to be avoided.
MultiDictionary(S:SetCategory): Category == DictionaryOperations S with
-- count: (S,%) -> NonNegativeInteger		       ++ multiplicity count
   insert!: (S,%,NonNegativeInteger) -> %
     ++ insert!(x,d,n) destructively inserts n copies of x into dictionary d.
-- remove!: (S,%,NonNegativeInteger) -> %
--   ++ remove!(x,d,n) destructively removes (up to) n copies of x from
--   ++ dictionary d.
   removeDuplicates!: % -> %
     ++ removeDuplicates!(d) destructively removes any duplicate values
     ++ in dictionary d.
   duplicates: % -> List Record(entry:S,count:NonNegativeInteger)
     ++ duplicates(d) returns a list of values which have duplicates in d
--   ++ duplicates(d) returns a list of		     ++ duplicates iterator
-- to become duplicates: % -> Iterator(D,D)

@

\section{category SETAGG SetAggregate}

<<category SETAGG SetAggregate>>=
import SetCategory
import Collection
)abbrev category SETAGG SetAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: 14 Oct, 1993 by RSS
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A set category lists a collection of set-theoretic operations
++ useful for both finite sets and multisets.
++ Note however that finite sets are distinct from multisets.
++ Although the operations defined for set categories are
++ common to both, the relationship between the two cannot
++ be described by inclusion or inheritance.
SetAggregate(S:SetCategory):
  Category == Join(SetCategory, Collection(S)) with
   partiallyOrderedSet
   part?         : (%, %) -> Boolean
     ++ s < t returns true if all elements of set aggregate s are also
     ++ elements of set aggregate t.
   brace       : () -> %
     ++ brace()$D (otherwise written {}$D)
     ++ creates an empty set aggregate of type D.
     ++ This form is considered obsolete. Use \axiomFun{set} instead.
   brace       : List S -> %
     ++ brace([x,y,...,z]) 
     ++ creates a set aggregate containing items x,y,...,z.
     ++ This form is considered obsolete. Use \axiomFun{set} instead.
   set	       : () -> %
     ++ set()$D creates an empty set aggregate of type D.
   set	       : List S -> %
     ++ set([x,y,...,z]) creates a set aggregate containing items x,y,...,z.
   intersect: (%, %) -> %
     ++ intersect(u,v) returns the set aggregate w consisting of
     ++ elements common to both set aggregates u and v.
     ++ Note: equivalent to the notation (not currently supported)
     ++ {x for x in u | member?(x,v)}.
   difference  : (%, %) -> %
     ++ difference(u,v) returns the set aggregate w consisting of
     ++ elements in set aggregate u but not in set aggregate v.
     ++ If u and v have no elements in common, \axiom{difference(u,v)}
     ++ returns a copy of u.
     ++ Note: equivalent to the notation (not currently supported)
     ++ \axiom{{x for x in u | not member?(x,v)}}.
   difference  : (%, S) -> %
     ++ difference(u,x) returns the set aggregate u with element x removed.
     ++ If u does not contain x, a copy of u is returned.
     ++ Note: \axiom{difference(s, x) = difference(s, {x})}.
   symmetricDifference : (%, %) -> %
     ++ symmetricDifference(u,v) returns the set aggregate of elements x which
     ++ are members of set aggregate u or set aggregate v but not both.
     ++ If u and v have no elements in common, \axiom{symmetricDifference(u,v)}
     ++ returns a copy of u.
     ++ Note: \axiom{symmetricDifference(u,v) = union(difference(u,v),difference(v,u))}
   subset?     : (%, %) -> Boolean
     ++ subset?(u,v) tests if u is a subset of v.
     ++ Note: equivalent to
     ++ \axiom{reduce(and,{member?(x,v) for x in u},true,false)}.
   union       : (%, %) -> %
     ++ union(u,v) returns the set aggregate of elements which are members
     ++ of either set aggregate u or v.
   union       : (%, S) -> %
     ++ union(u,x) returns the set aggregate u with the element x added.
     ++ If u already contains x, \axiom{union(u,x)} returns a copy of u.
   union       : (S, %) -> %
     ++ union(x,u) returns the set aggregate u with the element x added.
     ++ If u already contains x, \axiom{union(x,u)} returns a copy of u.
 add
  symmetricDifference(x, y)    == union(difference(x, y), difference(y, x))
  union(s:%, x:S) == union(s, {x})
  union(x:S, s:%) == union(s, {x})
  difference(s:%, x:S) == difference(s, {x})

@

\section{category FSAGG FiniteSetAggregate}

<<category FSAGG FiniteSetAggregate>>=
import Dictionary
import SetAggregate
)abbrev category FSAGG FiniteSetAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: 14 Oct, 1993 by RSS
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A finite-set aggregate models the notion of a finite set, that is,
++ a collection of elements characterized by membership, but not
++ by order or multiplicity.
++ See \spadtype{Set} for an example.
FiniteSetAggregate(S:SetCategory): Category ==
  Join(Dictionary S, SetAggregate S) with
    finiteAggregate
    cardinality: % -> NonNegativeInteger
      ++ cardinality(u) returns the number of elements of u.
      ++ Note: \axiom{cardinality(u) = #u}.
    if S has Finite then
      Finite
      complement: % -> %
	++ complement(u) returns the complement of the set u,
	++ i.e. the set of all values not in u.
      universe: () -> %
	++ universe()$D returns the universal set for finite set aggregate D.
    if S has OrderedSet then
      max: % -> S
	++ max(u) returns the largest element of aggregate u.
      min: % -> S
	++ min(u) returns the smallest element of aggregate u.

 add
   part?(s,t)	   == #s < #t and s = intersect(s,t)
   s = t	   == #s = #t and empty? difference(s,t)
   brace l	   == construct l
   set	 l	   == construct l
   cardinality s   == #s
   construct l	   == (s := set(); for x in l repeat insert!(x,s); s)
   count(x:S, s:%) == (member?(x, s) => 1; 0)
   subset?(s, t)   == #s < #t and (and/[member?(x, t) for x in parts s])

   coerce(s:%):OutputForm ==
     brace [x::OutputForm for x in parts s]$List(OutputForm)

   intersect(s, t) ==
     i := {}
     for x in parts s | member?(x, t) repeat insert!(x, i)
     i

   difference(s:%, t:%) ==
     m := copy s
     for x in parts t repeat remove!(x, m)
     m

   symmetricDifference(s, t) ==
     d := copy s
     for x in parts t repeat
       if member?(x, s) then remove!(x, d) else insert!(x, d)
     d

   union(s:%, t:%) ==
      u := copy s
      for x in parts t repeat insert!(x, u)
      u

   if S has Finite then
     universe()	  == {index(i::PositiveInteger) for i in 1..size()$S}
     complement s == difference(universe(), s )
     size()	  == 2 ** size()$S
     index i	 == {index(j::PositiveInteger)$S for j in 1..size()$S | bit?(i-1,j-1)}
     random()	  == index((random()$Integer rem (size()$% + 1))::PositiveInteger)

     lookup s ==
       n:PositiveInteger := 1
       for x in parts s repeat n := n + 2 ** ((lookup(x) - 1)::NonNegativeInteger)
       n

   if S has OrderedSet then
     max s ==
       empty?(l := parts s) => error "Empty set"
       reduce("max", l)

     min s ==
       empty?(l := parts s) => error "Empty set"
       reduce("min", l)

@

\section{category MSETAGG MultisetAggregate}

<<category MSETAGG MultisetAggregate>>=
import MultiDictionary
import SetAggregate
)abbrev category MSETAGG MultisetAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A multi-set aggregate is a set which keeps track of the multiplicity
++ of its elements.
MultisetAggregate(S:SetCategory):
 Category == Join(MultiDictionary S, SetAggregate S)

@

\section{category OMSAGG OrderedMultisetAggregate}

<<category OMSAGG OrderedMultisetAggregate>>=
import MultisetAggregate
import PriorityQueueAggregate
import List
)abbrev category OMSAGG OrderedMultisetAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ An ordered-multiset aggregate is a multiset built over an ordered set S
++ so that the relative sizes of its entries can be assessed.
++ These aggregates serve as models for priority queues.
OrderedMultisetAggregate(S:OrderedSet): Category ==
   Join(MultisetAggregate S,PriorityQueueAggregate S) with
   -- max: % -> S		      ++ smallest entry in the set
   -- duplicates: % -> List Record(entry:S,count:NonNegativeInteger)
        ++ to become an in order iterator
   -- parts: % -> List S	      ++ in order iterator
      min: % -> S
	++ min(u) returns the smallest entry in the multiset aggregate u.

@

\section{category KDAGG KeyedDictionary}

<<category KDAGG KeyedDictionary>>=
import Dictionary
import List
)abbrev category KDAGG KeyedDictionary
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A keyed dictionary is a dictionary of key-entry pairs for which there is
++ a unique entry for each key.
KeyedDictionary(Key:SetCategory, Entry:SetCategory): Category ==
  Dictionary Record(key:Key,entry:Entry) with
   key?: (Key, %) -> Boolean
     ++ key?(k,t) tests if k is a key in table t.
   keys: % -> List Key
     ++ keys(t) returns the list the keys in table t.
   -- to become keys: % -> Key* and keys: % -> Iterator(Entry,Entry)
   remove!: (Key, %) -> Union(Entry,"failed")
     ++ remove!(k,t) searches the table t for the key k removing
     ++ (and return) the entry if there.
     ++ If t has no such key, \axiom{remove!(k,t)} returns "failed".
   search: (Key, %) -> Union(Entry,"failed")
     ++ search(k,t) searches the table t for the key k,
     ++ returning the entry stored in t for key k.
     ++ If t has no such key, \axiom{search(k,t)} returns "failed".
 add
   key?(k, t) == search(k, t) case Entry

   member?(p, t) ==
     r := search(p.key, t)
     r case Entry and r::Entry = p.entry

   if % has finiteAggregate then
     keys t == [x.key for x in parts t]

@

\section{category ELTAB Eltable}

<<category ELTAB Eltable>>=
import Type
import SetCategory
)abbrev category ELTAB Eltable
++ Author: Michael Monagan; revised by Manuel Bronstein
++ Date Created: August 87 through August 88
++ Date Last Updated: April 25, 2010
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ An eltable over domains \spad{S} and \spad{T} is a structure which
++ can be viewed as a function from \spad{S} to \spad{T}.
++ Examples of eltable structures range from data structures, e.g. those
++ of type \spadtype{List}, to algebraic structures, e.g. \spadtype{Polynomial}.
Eltable(S: Type, T: Type): Category == Type with
  elt : (%, S) -> T
     ++ \spad{elt(u,s)} (also written: \spad{u.s}) returns the value
     ++ of \spad{u} at \spad{s}.
     ++ Error: if \spad{u} is not defined at \spad{s}.

@

\section{category ELTAGG EltableAggregate}

<<category ELTAGG EltableAggregate>>=
import Type
import SetCategory
)abbrev category ELTAGG EltableAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ An eltable aggregate is one which can be viewed as a function.
++ For example, the list \axiom{[1,7,4]} can applied to 0,1, and 2 respectively
++ will return the integers 1,7, and 4; thus this list may be viewed
++ as mapping 0 to 1, 1 to 7 and 2 to 4. In general, an aggregate
++ can map members of a domain {\em Dom} to an image domain {\em Im}.
EltableAggregate(Dom:SetCategory, Im:Type): Category ==
  Eltable(Dom, Im) with
    elt : (%, Dom, Im) -> Im
       ++ elt(u, x, y) applies u to x if x is in the domain of u,
       ++ and returns y otherwise.
       ++ For example, if u is a polynomial in \axiom{x} over the rationals,
       ++ \axiom{elt(u,n,0)} may define the coefficient of \axiom{x}
       ++ to the power n, returning 0 when n is out of range.
    qelt: (%, Dom) -> Im
       ++ qelt(u, x) applies \axiom{u} to \axiom{x} without checking whether
       ++ \axiom{x} is in the domain of \axiom{u}.  If \axiom{x} is not in the
       ++ domain of \axiom{u} a memory-access violation may occur.  If a check
       ++ on whether \axiom{x} is in the domain of \axiom{u} is required, use
       ++ the function \axiom{elt}.
    if % has shallowlyMutable then
       setelt : (%, Dom, Im) -> Im
	   ++ setelt(u,x,y) sets the image of x to be y under u,
	   ++ assuming x is in the domain of u.
	   ++ Error: if x is not in the domain of u.
	   -- this function will soon be renamed as setelt!.
       qsetelt!: (%, Dom, Im) -> Im
	   ++ qsetelt!(u,x,y) sets the image of \axiom{x} to be \axiom{y} under
           ++ \axiom{u}, without checking that \axiom{x} is in the domain of
           ++ \axiom{u}.
           ++ If such a check is required use the function \axiom{setelt}.
 add
  qelt(a, x) == elt(a, x)
  if % has shallowlyMutable then
    qsetelt!(a, x, y) == (a.x := y)

@

\section{category IXAGG IndexedAggregate}

<<category IXAGG IndexedAggregate>>=
import Type
import SetCategory
import HomogeneousAggregate
import EltableAggregate
import List
)abbrev category IXAGG IndexedAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ An indexed aggregate is a many-to-one mapping of indices to entries.
++ For example, a one-dimensional-array is an indexed aggregate where
++ the index is an integer.  Also, a table is an indexed aggregate
++ where the indices and entries may have any type.
IndexedAggregate(Index: SetCategory, Entry: Type): Category ==
  Join(HomogeneousAggregate(Entry), EltableAggregate(Index, Entry)) with
   entries: % -> List Entry
      ++ entries(u) returns a list of all the entries of aggregate u
      ++ in no assumed order.
      -- to become entries: % -> Entry* and entries: % -> Iterator(Entry,Entry)
   index?: (Index,%) -> Boolean
      ++ index?(i,u) tests if i is an index of aggregate u.
   indices: % -> List Index
      ++ indices(u) returns a list of indices of aggregate u in no
      ++ particular order.
      -- to become indices: % -> Index* and indices: % -> Iterator(Index,Index).
-- map: ((Entry,Entry)->Entry,%,%,Entry) -> %
--    ++ exists c = map(f,a,b,x), i:Index where
--    ++    c.i = f(a(i,x),b(i,x)) | index?(i,a) or index?(i,b)
   if Entry has SetCategory and % has finiteAggregate then
      entry?: (Entry,%) -> Boolean
	++ entry?(x,u) tests if x equals \axiom{u . i} for some index i.
   if Index has OrderedSet then
      maxIndex: % -> Index
	++ maxIndex(u) returns the maximum index i of aggregate u.
	++ Note: in general,
	++ \axiom{maxIndex(u) = reduce(max,[i for i in indices u])};
	++ if u is a list, \axiom{maxIndex(u) = #u}.
      minIndex: % -> Index
	++ minIndex(u) returns the minimum index i of aggregate u.
	++ Note: in general,
	++ \axiom{minIndex(a) = reduce(min,[i for i in indices a])};
	++ for lists, \axiom{minIndex(a) = 1}.
      first   : % -> Entry
	++ first(u) returns the first element x of u.
	++ Note: for collections, \axiom{first([x,y,...,z]) = x}.
	++ Error: if u is empty.

   if % has shallowlyMutable then
      fill!: (%,Entry) -> %
	++ fill!(u,x) replaces each entry in aggregate u by x.
	++ The modified u is returned as value.
      swap!: (%,Index,Index) -> Void
	++ swap!(u,i,j) interchanges elements i and j of aggregate u.
	++ No meaningful value is returned.
 add
  elt(a, i, x) == (index?(i, a) => qelt(a, i); x)

  if % has finiteAggregate then
    entries x == parts x
    if Entry has SetCategory then
      entry?(x, a) == member?(x, a)

  if Index has OrderedSet then
    maxIndex a == "max"/indices(a)
    minIndex a == "min"/indices(a)
    first a    == a minIndex a

  if % has shallowlyMutable then
    map(f, a) == map!(f, copy a)

    map!(f, a) ==
      for i in indices a repeat qsetelt!(a, i, f qelt(a, i))
      a

    fill!(a, x) ==
      for i in indices a repeat qsetelt!(a, i, x)
      a

    swap!(a, i, j) ==
      t := a.i
      qsetelt!(a, i, a.j)
      qsetelt!(a, j, t)

@

\section{category TBAGG TableAggregate}

<<category TBAGG TableAggregate>>=
import SetCategory
import KeyedDictionary
import IndexedAggregate
import Boolean
import OutputForm
import List
)abbrev category TBAGG TableAggregate
++ Author: Michael Monagan, Stephen Watt; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A table aggregate is a model of a table, i.e. a discrete many-to-one
++ mapping from keys to entries.
TableAggregate(Key:SetCategory, Entry:SetCategory): Category ==
  Join(KeyedDictionary(Key,Entry),IndexedAggregate(Key,Entry)) with
   setelt: (%,Key,Entry) -> Entry	   -- setelt! later
     ++ setelt(t,k,e) (also written \axiom{t.k := e}) is equivalent
     ++ to \axiom{(insert([k,e],t); e)}.
   table: () -> %
     ++ table()$T creates an empty table of type T.
   table: List Record(key:Key,entry:Entry) -> %
     ++ table([x,y,...,z]) creates a table consisting of entries
     ++ \axiom{x,y,...,z}.
   -- to become table: Record(key:Key,entry:Entry)* -> %
   map: ((Entry, Entry) -> Entry, %, %) -> %
     ++ map(fn,t1,t2) creates a new table t from given tables t1 and t2 with
     ++ elements fn(x,y) where x and y are corresponding elements from t1
     ++ and t2 respectively.
 add
   table()	       == empty()
   table l	       == dictionary l
-- empty()	       == dictionary()

   insert!(p, t)      == (t(p.key) := p.entry; t)
   indices t	       == keys t

   coerce(t:%):OutputForm ==
     prefix("table"::OutputForm,
		    [k::OutputForm = (t.k)::OutputForm for k in keys t])

   elt(t, k) ==
      (r := search(k, t)) case Entry => r::Entry
      error "key not in table"

   elt(t, k, e) ==
      (r := search(k, t)) case Entry => r::Entry
      e

   map!(f: Entry->Entry, t: %) ==
      for k in keys t repeat t.k := f t.k
      t

   map(f:(Entry, Entry) -> Entry, s:%, t:%) ==
      z := table()
      for k in keys s | key?(k, t) repeat z.k := f(s.k, t.k)
      z

-- map(f, s, t, x) ==
--    z := table()
--    for k in keys s repeat z.k := f(s.k, t(k, x))
--    for k in keys t | not key?(k, s) repeat z.k := f(t.k, x)
--    z

   if % has finiteAggregate then
     parts(t:%):List Record(key:Key,entry:Entry)	     == [[k, t.k] for k in keys t]
     parts(t:%):List Entry   == [t.k for k in keys t]
     entries(t:%):List Entry == parts(t)

     s:% = t:% ==
       eq?(s,t) => true
       #s ~= #t => false
       for k in keys s repeat
	 (e := search(k, t)) case "failed" or (e::Entry) ~= s.k => 
            return false
       true

     map(f: Record(key:Key,entry:Entry)->Record(key:Key,entry:Entry), t: %): % ==
       z := table()
       for k in keys t repeat
	 ke: Record(key:Key,entry:Entry) := f [k, t.k]
	 z ke.key := ke.entry
       z
     map!(f: Record(key:Key,entry:Entry)->Record(key:Key,entry:Entry), t: %): % ==
       lke: List Record(key:Key,entry:Entry) := nil()
       for k in keys t repeat
	 lke := cons(f [k, remove!(k,t)::Entry], lke)
       for ke in lke repeat
	 t ke.key := ke.entry
       t

     inspect(t: %): Record(key:Key,entry:Entry) ==
       ks := keys t
       empty? ks => error "Cannot extract from an empty aggregate"
       [first ks, t first ks]

     find(f: Record(key:Key,entry:Entry)->Boolean, t:%): Union(Record(key:Key,entry:Entry), "failed") ==
       for ke in parts(t)@List(Record(key:Key,entry:Entry)) repeat if f ke then return ke
       "failed"

     index?(k: Key, t: %): Boolean ==
       search(k,t) case Entry

     remove!(x:Record(key:Key,entry:Entry), t:%) ==
       if member?(x, t) then remove!(x.key, t)
       t
     extract!(t: %): Record(key:Key,entry:Entry) ==
       k: Record(key:Key,entry:Entry) := inspect t
       remove!(k.key, t)
       k

     any?(f: Entry->Boolean, t: %): Boolean ==
       for k in keys t | f t k repeat return true
       false
     every?(f: Entry->Boolean, t: %): Boolean ==
       for k in keys t | not f t k repeat return false
       true
     count(f: Entry->Boolean, t: %): NonNegativeInteger ==
       tally: NonNegativeInteger := 0
       for k in keys t | f t k repeat tally := tally + 1
       tally

@

\section{category RCAGG RecursiveAggregate}

<<category RCAGG RecursiveAggregate>>=
import Type
import SetCategory
import List
import Boolean
)abbrev category RCAGG RecursiveAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A recursive aggregate over a type S is a model for a
++ a directed graph containing values of type S.
++ Recursively, a recursive aggregate is a {\em node}
++ consisting of a \spadfun{value} from S and 0 or more \spadfun{children}
++ which are recursive aggregates.
++ A node with no children is called a \spadfun{leaf} node.
++ A recursive aggregate may be cyclic for which some operations as noted
++ may go into an infinite loop.
RecursiveAggregate(S:Type): Category == HomogeneousAggregate(S) with
   children: % -> List %
     ++ children(u) returns a list of the children of aggregate u.
   -- should be % -> %* and also needs children: % -> Iterator(S,S)
   nodes: % -> List %
     ++ nodes(u) returns a list of all of the nodes of aggregate u.
   -- to become % -> %* and also nodes: % -> Iterator(S,S)
   leaf?: % -> Boolean
     ++ leaf?(u) tests if u is a terminal node.
   value: % -> S
     ++ value(u) returns the value of the node u.
   elt: (%,"value") -> S
     ++ elt(u,"value") (also written: \axiom{a. value}) is
     ++ equivalent to \axiom{value(a)}.
   cyclic?: % -> Boolean
     ++ cyclic?(u) tests if u has a cycle.
   leaves: % -> List S
     ++ leaves(t) returns the list of values in obtained by visiting the
     ++ nodes of tree \axiom{t} in left-to-right order.
   distance: (%,%) -> Integer
     ++ distance(u,v) returns the path length (an integer) from node u to v.
   if S has SetCategory then
      child?: (%,%) -> Boolean
	++ child?(u,v) tests if node u is a child of node v.
      node?: (%,%) -> Boolean
	++ node?(u,v) tests if node u is contained in node v
	++ (either as a child, a child of a child, etc.).
   if % has shallowlyMutable then
      setchildren!: (%,List %)->%
	++ setchildren!(u,v) replaces the current children of node u
	++ with the members of v in left-to-right order.
      setelt: (%,"value",S) -> S
	++ setelt(a,"value",x) (also written \axiom{a . value := x})
	++ is equivalent to \axiom{setvalue!(a,x)}
      setvalue!: (%,S) -> S
	++ setvalue!(u,x) sets the value of node u to x.
 add
   elt(x,"value") == value x
   if % has shallowlyMutable then
     setelt(x,"value",y) == setvalue!(x,y)
   if S has SetCategory then
     child?(x,l) == member?(x,children(l))

@

\section{category BRAGG BinaryRecursiveAggregate}

<<category BRAGG BinaryRecursiveAggregate>>=
import Type
import RecursiveAggregate
)abbrev category BRAGG BinaryRecursiveAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A binary-recursive aggregate has 0, 1 or 2 children and
++ serves as a model for a binary tree or a doubly-linked aggregate structure
BinaryRecursiveAggregate(S:Type):Category == RecursiveAggregate S with
   -- needs preorder, inorder and postorder iterators
   left: % -> %
     ++ left(u) returns the left child.
   elt: (%,"left") -> %
     ++ elt(u,"left") (also written: \axiom{a . left}) is
     ++ equivalent to \axiom{left(a)}.
   right: % -> %
     ++ right(a) returns the right child.
   elt: (%,"right") -> %
     ++ elt(a,"right") (also written: \axiom{a . right})
     ++ is equivalent to \axiom{right(a)}.
   if % has shallowlyMutable then
      setelt: (%,"left",%) -> %
	++ setelt(a,"left",b) (also written \axiom{a . left := b}) is equivalent
	++ to \axiom{setleft!(a,b)}.
      setleft!: (%,%) -> %
	 ++ setleft!(a,b) sets the left child of \axiom{a} to be b.
      setelt: (%,"right",%) -> %
	 ++ setelt(a,"right",b) (also written \axiom{b . right := b})
	 ++ is equivalent to \axiom{setright!(a,b)}.
      setright!: (%,%) -> %
	 ++ setright!(a,x) sets the right child of t to be x.
 add
   cycleMax ==> 1000

   elt(x,"left")  == left x
   elt(x,"right") == right x
   leaf? x == empty? x or empty? left x and empty? right x
   leaves t ==
     empty? t => empty()$List(S)
     leaf? t => [value t]
     concat(leaves left t,leaves right t)
   nodes x ==
     l := empty()$List(%)
     empty? x => l
     concat(nodes left x,concat([x],nodes right x))
   children x ==
     l := empty()$List(%)
     empty? x => l
     empty? left x  => [right x]
     empty? right x => [left x]
     [left x, right x]
   if % has SetAggregate(S) and S has SetCategory then
     node?(u,v) ==
       empty? v => false
       u = v => true
       for y in children v repeat node?(u,y) => return true
       false
     x = y ==
       empty?(x) => empty?(y)
       empty?(y) => false
       value x = value y and left x = left y and right x = right y
     if % has finiteAggregate then
       member?(x,u) ==
	 empty? u => false
	 x = value u => true
	 member?(x,left u) or member?(x,right u)

   if S has CoercibleTo(OutputForm) then
     coerce(t:%): OutputForm ==
       empty? t =>  bracket(empty()$OutputForm)
       v := value(t):: OutputForm
       empty? left t =>
	 empty? right t => v
	 r := (right t)::OutputForm
	 bracket ["."::OutputForm, v, r]
       l := (left t)::OutputForm
       r :=
	 empty? right t => "."::OutputForm
	 (right t)::OutputForm
       bracket [l, v, r]

   if % has finiteAggregate then
     aggCount: (%,NonNegativeInteger) -> NonNegativeInteger
     #x == aggCount(x,0)
     aggCount(x,k) ==
       empty? x => 0
       k := k + 1
       k = cycleMax and cyclic? x => error "cyclic tree"
       for y in children x repeat k := aggCount(y,k)
       k

   isCycle?:  (%, List %) -> Boolean
   eqMember?: (%, List %) -> Boolean
   cyclic? x	 == not empty? x and isCycle?(x,empty()$(List %))
   isCycle?(x,acc) ==
     empty? x => false
     eqMember?(x,acc) => true
     for y in children x | not empty? y repeat
       isCycle?(y,acc) => return true
     false
   eqMember?(y,l) ==
     for x in l repeat eq?(x,y) => return true
     false
   if % has shallowlyMutable then
     setelt(x,"left",b)  == setleft!(x,b)
     setelt(x,"right",b) == setright!(x,b)

@

\section{category DLAGG DoublyLinkedAggregate}

<<category DLAGG DoublyLinkedAggregate>>=
import Type
import RecursiveAggregate
)abbrev category DLAGG DoublyLinkedAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A doubly-linked aggregate serves as a model for a doubly-linked
++ list, that is, a list which can has links to both next and previous
++ nodes and thus can be efficiently traversed in both directions.
DoublyLinkedAggregate(S:Type): Category == RecursiveAggregate S with
   last: % -> S
     ++ last(l) returns the last element of a doubly-linked aggregate l.
     ++ Error: if l is empty.
   head: % -> %
     ++ head(l) returns the first element of a doubly-linked aggregate l.
     ++ Error: if l is empty.
   tail: % -> %
     ++ tail(l) returns the doubly-linked aggregate l starting at
     ++ its second element.
     ++ Error: if l is empty.
   previous: % -> %
     ++ previous(l) returns the doubly-link list beginning with its previous
     ++ element.
     ++ Error: if l has no previous element.
     ++ Note: \axiom{next(previous(l)) = l}.
   next: % -> %
     ++ next(l) returns the doubly-linked aggregate beginning with its next
     ++ element.
     ++ Error: if l has no next element.
     ++ Note: \axiom{next(l) = rest(l)} and \axiom{previous(next(l)) = l}.
   if % has shallowlyMutable then
      concat!: (%,%) -> %
	++ concat!(u,v) destructively concatenates doubly-linked aggregate v to the end of doubly-linked aggregate u.
      setprevious!: (%,%) -> %
	++ setprevious!(u,v) destructively sets the previous node of doubly-linked aggregate u to v, returning v.
      setnext!: (%,%) -> %
	++ setnext!(u,v) destructively sets the next node of doubly-linked aggregate u to v, returning v.

@

\section{category URAGG UnaryRecursiveAggregate}

<<category URAGG UnaryRecursiveAggregate>>=
import Type
import RecursiveAggregate
import NonNegativeInteger
)abbrev category URAGG UnaryRecursiveAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A unary-recursive aggregate is a one where nodes may have either
++ 0 or 1 children.
++ This aggregate models, though not precisely, a linked
++ list possibly with a single cycle.
++ A node with one children models a non-empty list, with the
++ \spadfun{value} of the list designating the head, or \spadfun{first}, of the
++ list, and the child designating the tail, or \spadfun{rest}, of the list.
++ A node with no child then designates the empty list.
++ Since these aggregates are recursive aggregates, they may be cyclic.
UnaryRecursiveAggregate(S:Type): Category == RecursiveAggregate S with
   concat: (%,%) -> %
      ++ concat(u,v) returns an aggregate w consisting of the elements of u
      ++ followed by the elements of v.
      ++ Note: \axiom{v = rest(w,#a)}.
   concat: (S,%) -> %
      ++ concat(x,u) returns aggregate consisting of x followed by
      ++ the elements of u.
      ++ Note: if \axiom{v = concat(x,u)} then \axiom{x = first v}
      ++ and \axiom{u = rest v}.
   first: % -> S
      ++ first(u) returns the first element of u
      ++ (equivalently, the value at the current node).
   elt: (%,"first") -> S
      ++ elt(u,"first") (also written: \axiom{u . first}) is equivalent to first u.
   first: (%,NonNegativeInteger) -> %
      ++ first(u,n) returns a copy of the first n (\axiom{n >= 0}) elements of u.
   rest: % -> %
      ++ rest(u) returns an aggregate consisting of all but the first
      ++ element of u
      ++ (equivalently, the next node of u).
   elt: (%,"rest") -> %
      ++ elt(%,"rest") (also written: \axiom{u.rest}) is
      ++ equivalent to \axiom{rest u}.
   rest: (%,NonNegativeInteger) -> %
      ++ rest(u,n) returns the \axiom{n}th (n >= 0) node of u.
      ++ Note: \axiom{rest(u,0) = u}.
   last: % -> S
      ++ last(u) resturn the last element of u.
      ++ Note: for lists, \axiom{last(u) = u . (maxIndex u) = u . (# u - 1)}.
   elt: (%,"last") -> S
      ++ elt(u,"last") (also written: \axiom{u . last}) is equivalent to last u.
   last: (%,NonNegativeInteger) -> %
      ++ last(u,n) returns a copy of the last n (\axiom{n >= 0}) nodes of u.
      ++ Note: \axiom{last(u,n)} is a list of n elements.
   tail: % -> %
      ++ tail(u) returns the last node of u.
      ++ Note: if u is \axiom{shallowlyMutable},
      ++ \axiom{setrest(tail(u),v) = concat(u,v)}.
   second: % -> S
      ++ second(u) returns the second element of u.
      ++ Note: \axiom{second(u) = first(rest(u))}.
   third: % -> S
      ++ third(u) returns the third element of u.
      ++ Note: \axiom{third(u) = first(rest(rest(u)))}.
   cycleEntry: % -> %
      ++ cycleEntry(u) returns the head of a top-level cycle contained in
      ++ aggregate u, or \axiom{empty()} if none exists.
   cycleLength: % -> NonNegativeInteger
      ++ cycleLength(u) returns the length of a top-level cycle
      ++ contained  in aggregate u, or 0 is u has no such cycle.
   cycleTail: % -> %
      ++ cycleTail(u) returns the last node in the cycle, or
      ++ empty if none exists.
   if % has shallowlyMutable then
      concat!: (%,%) -> %
	++ concat!(u,v) destructively concatenates v to the end of u.
	++ Note: \axiom{concat!(u,v) = setlast!(u,v)}.
      concat!: (%,S) -> %
	++ concat!(u,x) destructively adds element x to the end of u.
	++ Note: \axiom{concat!(a,x) = setlast!(a,[x])}.
      cycleSplit!: % -> %
	++ cycleSplit!(u) splits the aggregate by dropping off the cycle.
	++ The value returned is the cycle entry, or nil if none exists.
	++ For example, if \axiom{w = concat(u,v)} is the cyclic list where v is
	++ the head of the cycle, \axiom{cycleSplit!(w)} will drop v off w thus
	++ destructively changing w to u, and returning v.
      setfirst!: (%,S) -> S
	++ setfirst!(u,x) destructively changes the first element of a to x.
      setelt: (%,"first",S) -> S
	++ setelt(u,"first",x) (also written: \axiom{u.first := x}) is
	++ equivalent to \axiom{setfirst!(u,x)}.
      setrest!: (%,%) -> %
	++ setrest!(u,v) destructively changes the rest of u to v.
      setelt: (%,"rest",%) -> %
	++ setelt(u,"rest",v) (also written: \axiom{u.rest := v}) is equivalent to
	++ \axiom{setrest!(u,v)}.
      setlast!: (%,S) -> S
	++ setlast!(u,x) destructively changes the last element of u to x.
      setelt: (%,"last",S) -> S
	++ setelt(u,"last",x) (also written: \axiom{u.last := b})
	++ is equivalent to \axiom{setlast!(u,v)}.
      split!: (%,Integer) -> %
	 ++ split!(u,n) splits u into two aggregates: \axiom{v = rest(u,n)}
	 ++ and \axiom{w = first(u,n)}, returning \axiom{v}.
	 ++ Note: afterwards \axiom{rest(u,n)} returns \axiom{empty()}.
 add
  cycleMax ==> 1000

  findCycle: % -> %

  elt(x, "first") == first x
  elt(x,  "last") == last x
  elt(x,  "rest") == rest x
  second x	  == first rest x
  third x	  == first rest rest x
  cyclic? x	  == not empty? x and not empty? findCycle x
  last x	  == first tail x

  nodes x ==
    l := empty()$List(%)
    while not empty? x repeat
      l := concat(x, l)
      x := rest x
    reverse! l

  children x ==
    l := empty()$List(%)
    empty? x => l
    concat(rest x,l)

  leaf? x == empty? x

  value x ==
    empty? x => error "value of empty object"
    first x

  less?(l, n) ==
    i := n::Integer
    while i > 0 and not empty? l repeat (l := rest l; i := i - 1)
    i > 0

  more?(l, n) ==
    i := n::Integer
    while i > 0 and not empty? l repeat (l := rest l; i := i - 1)
    zero?(i) and not empty? l

  size?(l, n) ==
    i := n::Integer
    while not empty? l and i > 0 repeat (l := rest l; i := i - 1)
    empty? l and zero? i

  #x ==
    k: NonNegativeInteger := 0
    while not empty? x repeat
      k = cycleMax and cyclic? x => error "cyclic list"
      x := rest x
      k := k + 1
    k

  tail x ==
    empty? x => error "empty list"
    y := rest x
    for k in 0.. while not empty? y repeat
      k = cycleMax and cyclic? x => error "cyclic list"
      y := rest(x := y)
    x

  findCycle x ==
    y := rest x
    while not empty? y repeat
      if eq?(x, y) then return x
      x := rest x
      y := rest y
      if empty? y then return y
      if eq?(x, y) then return y
      y := rest y
    y

  cycleTail x ==
    empty?(y := x := cycleEntry x) => x
    z := rest x
    while not eq?(x,z) repeat (y := z; z := rest z)
    y

  cycleEntry x ==
    empty? x => x
    empty?(y := findCycle x) => y
    z := rest y
    l: NonNegativeInteger := 1
    while not eq?(y,z) repeat
      z := rest z
      l := l + 1
    y := x
    for k in 1..l repeat y := rest y
    while not eq?(x,y) repeat (x := rest x; y := rest y)
    x

  cycleLength x ==
    empty? x => 0
    empty?(x := findCycle x) => 0
    y := rest x
    k: NonNegativeInteger := 1
    while not eq?(x,y) repeat
      y := rest y
      k := k + 1
    k

  rest(x, n) ==
    for i in 1..n repeat
      empty? x => error "Index out of range"
      x := rest x
    x

  if % has finiteAggregate then
    last(x, n) ==
      n > (m := #x) => error "index out of range"
      copy rest(x, (m - n)::NonNegativeInteger)

  if S has SetCategory then
    x = y ==
      eq?(x, y) => true
      for k in 0.. while not empty? x and not empty? y repeat
	k = cycleMax and cyclic? x => error "cyclic list"
	first x ~= first y => return false
	x := rest x
	y := rest y
      empty? x and empty? y

    node?(u, v) ==
      for k in 0.. while not empty? v repeat
	u = v => return true
	k = cycleMax and cyclic? v => error "cyclic list"
	v := rest v
      u=v

  if % has shallowlyMutable then
    setelt(x, "first", a) == setfirst!(x, a)
    setelt(x,  "last", a) == setlast!(x, a)
    setelt(x,  "rest", a) == setrest!(x, a)
    concat(x:%, y:%)	  == concat!(copy x, y)

    setlast!(x, s) ==
      empty? x => error "setlast: empty list"
      setfirst!(tail x, s)
      s

    setchildren!(u,lv) ==
      #lv=1 => setrest!(u, first lv)
      error "wrong number of children specified"

    setvalue!(u,s) == setfirst!(u,s)

    split!(p, n) ==
      n < 1 => error "index out of range"
      p := rest(p, (n - 1)::NonNegativeInteger)
      q := rest p
      setrest!(p, empty())
      q

    cycleSplit! x ==
      empty?(y := cycleEntry x) or eq?(x, y) => y
      z := rest x
      while not eq?(z, y) repeat (x := z; z := rest z)
      setrest!(x, empty())
      y

@

\section{category STAGG StreamAggregate}

<<category STAGG StreamAggregate>>=
import Type
import UnaryRecursiveAggregate
import LinearAggregate
import Boolean
)abbrev category STAGG StreamAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A stream aggregate is a linear aggregate which possibly has an infinite
++ number of elements. A basic domain constructor which builds stream
++ aggregates is \spadtype{Stream}. From streams, a number of infinite
++ structures such power series can be built. A stream aggregate may
++ also be infinite since it may be cyclic.
++ For example, see \spadtype{DecimalExpansion}.
StreamAggregate(S:Type): Category ==
   Join(UnaryRecursiveAggregate S, LinearAggregate S) with
      explicitlyFinite?: % -> Boolean
	++ explicitlyFinite?(s) tests if the stream has a finite
	++ number of elements, and false otherwise.
	++ Note: for many datatypes, \axiom{explicitlyFinite?(s) = not possiblyInfinite?(s)}.
      possiblyInfinite?: % -> Boolean
	++ possiblyInfinite?(s) tests if the stream s could possibly
	++ have an infinite number of elements.
	++ Note: for many datatypes, \axiom{possiblyInfinite?(s) = not explictlyFinite?(s)}.
 add
   c2: (%, %) -> S

   explicitlyFinite? x == not cyclic? x
   possiblyInfinite? x == cyclic? x
   first(x, n)	       == construct [c2(x, x := rest x) for i in 1..n]

   c2(x, r) ==
     empty? x => error "Index out of range"
     first x

   elt(x:%, i:Integer) ==
     i := i - minIndex x
     (i < 0) or empty?(x := rest(x, i::NonNegativeInteger)) => error "index out of range"
     first x

   elt(x:%, i:UniversalSegment(Integer)) ==
     l := lo(i) - minIndex x
     l < 0 => error "index out of range"
     not hasHi i => copy(rest(x, l::NonNegativeInteger))
     (h := hi(i) - minIndex x) < l => empty()
     first(rest(x, l::NonNegativeInteger), (h - l + 1)::NonNegativeInteger)

   if % has shallowlyMutable then
     concat(x:%, y:%) == concat!(copy x, y)

     concat l ==
       empty? l => empty()
       concat!(copy first l, concat rest l)

     map!(f, l) ==
       y := l
       while not empty? l repeat
	 setfirst!(l, f first l)
	 l := rest l
       y

     fill!(x, s) ==
       y := x
       while not empty? y repeat (setfirst!(y, s); y := rest y)
       x

     setelt(x:%, i:Integer, s:S) ==
      i := i - minIndex x
      (i < 0) or empty?(x := rest(x,i::NonNegativeInteger)) => error "index out of range"
      setfirst!(x, s)

     setelt(x:%, i:UniversalSegment(Integer), s:S) ==
      (l := lo(i) - minIndex x) < 0 => error "index out of range"
      h := if hasHi i then hi(i) - minIndex x else maxIndex x
      h < l => s
      y := rest(x, l::NonNegativeInteger)
      z := rest(y, (h - l + 1)::NonNegativeInteger)
      while not eq?(y, z) repeat (setfirst!(y, s); y := rest y)
      s

     concat!(x:%, y:%) ==
       empty? x => y
       setrest!(tail x, y)
       x

@

\section{category LNAGG LinearAggregate}

<<category LNAGG LinearAggregate>>=
import Type
import Collection
import IndexedAggregate
import NonNegativeInteger
import Integer
)abbrev category LNAGG LinearAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A linear aggregate is an aggregate whose elements are indexed by integers.
++ Examples of linear aggregates are strings, lists, and
++ arrays.
++ Most of the exported operations for linear aggregates are non-destructive
++ but are not always efficient for a particular aggregate.
++ For example, \spadfun{concat} of two lists needs only to copy its first
++ argument, whereas \spadfun{concat} of two arrays needs to copy both arguments.
++ Most of the operations exported here apply to infinite objects (e.g. streams)
++ as well to finite ones.
++ For finite linear aggregates, see \spadtype{FiniteLinearAggregate}.
LinearAggregate(S:Type): Category ==
  Join(IndexedAggregate(Integer, S), Collection(S),_
    Eltable(UniversalSegment Integer, %)) with
   new	 : (NonNegativeInteger,S) -> %
     ++ new(n,x) returns \axiom{fill!(new n,x)}.
   concat: (%,S) -> %
     ++ concat(u,x) returns aggregate u with additional element x at the end.
     ++ Note: for lists, \axiom{concat(u,x) == concat(u,[x])}
   concat: (S,%) -> %
     ++ concat(x,u) returns aggregate u with additional element at the front.
     ++ Note: for lists: \axiom{concat(x,u) == concat([x],u)}.
   concat: (%,%) -> %
      ++ concat(u,v) returns an aggregate consisting of the elements of u
      ++ followed by the elements of v.
      ++ Note: if \axiom{w = concat(u,v)} then \axiom{w.i = u.i for i in indices u}
      ++ and \axiom{w.(j + maxIndex u) = v.j for j in indices v}.
   concat: List % -> %
      ++ concat(u), where u is a lists of aggregates \axiom{[a,b,...,c]}, returns
      ++ a single aggregate consisting of the elements of \axiom{a}
      ++ followed by those
      ++ of b followed ... by the elements of c.
      ++ Note: \axiom{concat(a,b,...,c) = concat(a,concat(b,...,c))}.
   map: ((S,S)->S,%,%) -> %
     ++ map(f,u,v) returns a new collection w with elements \axiom{z = f(x,y)}
     ++ for corresponding elements x and y from u and v.
     ++ Note: for linear aggregates, \axiom{w.i = f(u.i,v.i)}.
   delete: (%,Integer) -> %
      ++ delete(u,i) returns a copy of u with the \axiom{i}th element deleted.
      ++ Note: for lists, \axiom{delete(a,i) == concat(a(0..i - 1),a(i + 1,..))}.
   delete: (%,UniversalSegment(Integer)) -> %
      ++ delete(u,i..j) returns a copy of u with the \axiom{i}th through
      ++ \axiom{j}th element deleted.
      ++ Note: \axiom{delete(a,i..j) = concat(a(0..i-1),a(j+1..))}.
   insert: (S,%,Integer) -> %
      ++ insert(x,u,i) returns a copy of u having x as its \axiom{i}th element.
      ++ Note: \axiom{insert(x,a,k) = concat(concat(a(0..k-1),x),a(k..))}.
   insert: (%,%,Integer) -> %
      ++ insert(v,u,k) returns a copy of u having v inserted beginning at the
      ++ \axiom{i}th element.
      ++ Note: \axiom{insert(v,u,k) = concat( u(0..k-1), v, u(k..) )}.
   if % has shallowlyMutable then setelt: (%,UniversalSegment(Integer),S) -> S
      ++ setelt(u,i..j,x) (also written: \axiom{u(i..j) := x}) destructively
      ++ replaces each element in the segment \axiom{u(i..j)} by x.
      ++ The value x is returned.
      ++ Note: u is destructively change so
      ++ that \axiom{u.k := x for k in i..j};
      ++ its length remains unchanged.
 add
  indices a	 == [i for i in minIndex a .. maxIndex a]
  index?(i, a)	 == i >= minIndex a and i <= maxIndex a
  concat(a:%, x:S)	== concat(a, new(1, x))
  concat(x:S, y:%)	== concat(new(1, x), y)
  insert(x:S, a:%, i:Integer) == insert(new(1, x), a, i)
  if % has finiteAggregate then
    maxIndex l == #l - 1 + minIndex l

--if % has shallowlyMutable then new(n, s)  == fill!(new n, s)

@

\section{category FLAGG FiniteLinearAggregate}

<<category FLAGG FiniteLinearAggregate>>=
import Type
import SetCategory
import OrderedSet
import LinearAggregate
import Boolean
import Integer
)abbrev category FLAGG FiniteLinearAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A finite linear aggregate is a linear aggregate of finite length.
++ The finite property of the aggregate adds several exports to the
++ list of exports from \spadtype{LinearAggregate} such as
++ \spadfun{reverse}, \spadfun{sort}, and so on.
FiniteLinearAggregate(S:Type): Category == LinearAggregate S with
   finiteAggregate
   merge: ((S,S)->Boolean,%,%) -> %
      ++ merge(p,a,b) returns an aggregate c which merges \axiom{a} and b.
      ++ The result is produced by examining each element x of \axiom{a} and y
      ++ of b successively. If \axiom{p(x,y)} is true, then x is inserted into
      ++ the result; otherwise y is inserted. If x is chosen, the next element
      ++ of \axiom{a} is examined, and so on. When all the elements of one
      ++ aggregate are examined, the remaining elements of the other
      ++ are appended.
      ++ For example, \axiom{merge(<,[1,3],[2,7,5])} returns \axiom{[1,2,3,7,5]}.
   reverse: % -> %
      ++ reverse(a) returns a copy of \axiom{a} with elements in reverse order.
   sort: ((S,S)->Boolean,%) -> %
      ++ sort(p,a) returns a copy of \axiom{a} sorted using total ordering predicate p.
   sorted?: ((S,S)->Boolean,%) -> Boolean
      ++ sorted?(p,a) tests if \axiom{a} is sorted according to predicate p.
   position: (S->Boolean, %) -> Integer
      ++ position(p,a) returns the index i of the first x in \axiom{a} such that
      ++ \axiom{p(x)} is true, and \axiom{minIndex(a) - 1} if there is no such x.
   if S has SetCategory then
      position: (S, %)	-> Integer
	++ position(x,a) returns the index i of the first occurrence of x in a,
	++ and \axiom{minIndex(a) - 1} if there is no such x.
      position: (S,%,Integer) -> Integer
	++ position(x,a,n) returns the index i of the first occurrence of x in
	++ \axiom{a} where \axiom{i >= n}, and \axiom{minIndex(a) - 1} if no such x is found.
   if S has OrderedSet then
      OrderedSet
      merge: (%,%) -> %
	++ merge(u,v) merges u and v in ascending order.
	++ Note: \axiom{merge(u,v) = merge(<=,u,v)}.
      sort: % -> %
	++ sort(u) returns an u with elements in ascending order.
	++ Note: \axiom{sort(u) = sort(<=,u)}.
      sorted?: % -> Boolean
	++ sorted?(u) tests if the elements of u are in ascending order.
   if % has shallowlyMutable then
      copyInto!: (%,%,Integer) -> %
	++ copyInto!(u,v,i) returns aggregate u containing a copy of
	++ v inserted at element i.
      reverse!: % -> %
	++ reverse!(u) returns u with its elements in reverse order.
      sort!: ((S,S)->Boolean,%) -> %
	++ sort!(p,u) returns u with its elements ordered by p.
      if S has OrderedSet then sort!: % -> %
	++ sort!(u) returns u with its elements in ascending order.
 add
    if S has SetCategory then
      position(x:S, t:%) == position(x, t, minIndex t)

    if S has OrderedSet then
--    sorted? l	  == sorted?(_<$S, l)
      sorted? l	  == sorted?(#1 < #2 or #1 = #2, l)
      merge(x, y) == merge(_<$S, x, y)
      sort l	  == sort(_<$S, l)

    if % has shallowlyMutable then
      reverse x	 == reverse! copy x
      sort(f, l) == sort!(f, copy l)

      if S has OrderedSet then
	sort! l == sort!(_<$S, l)

@

\section{category A1AGG OneDimensionalArrayAggregate}

<<category A1AGG OneDimensionalArrayAggregate>>=
import Type
import FiniteLinearAggregate
)abbrev category A1AGG OneDimensionalArrayAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ One-dimensional-array aggregates serves as models for one-dimensional arrays.
++ Categorically, these aggregates are finite linear aggregates
++ with the \spadatt{shallowlyMutable} property, that is, any component of
++ the array may be changed without affecting the
++ identity of the overall array.
++ Array data structures are typically represented by a fixed area in storage and
++ therefore cannot efficiently grow or shrink on demand as can list structures
++ (see however \spadtype{FlexibleArray} for a data structure which
++ is a cross between a list and an array).
++ Iteration over, and access to, elements of arrays is extremely fast
++ (and often can be optimized to open-code).
++ Insertion and deletion however is generally slow since an entirely new
++ data structure must be created for the result.
OneDimensionalArrayAggregate(S:Type): Category ==
    FiniteLinearAggregate S with shallowlyMutable
  add
    parts x	    == [qelt(x, i) for i in minIndex x .. maxIndex x]
    sort!(f, a) == quickSort(f, a)$FiniteLinearAggregateSort(S, %)

    any?(f, a) ==
      for i in minIndex a .. maxIndex a repeat
	f qelt(a, i) => return true
      false

    every?(f, a) ==
      for i in minIndex a .. maxIndex a repeat
	not(f qelt(a, i)) => return false
      true

    position(f:S -> Boolean, a:%) ==
      for i in minIndex a .. maxIndex a repeat
	f qelt(a, i) => return i
      minIndex(a) - 1

    find(f, a) ==
      for i in minIndex a .. maxIndex a repeat
	f qelt(a, i) => return qelt(a, i)
      "failed"

    count(f:S->Boolean, a:%) ==
      n:NonNegativeInteger := 0
      for i in minIndex a .. maxIndex a repeat
	if f(qelt(a, i)) then n := n+1
      n

    map!(f, a) ==
      for i in minIndex a .. maxIndex a repeat
	qsetelt!(a, i, f qelt(a, i))
      a

    setelt(a:%, s:UniversalSegment(Integer), x:S) ==
      l := lo s; h := if hasHi s then hi s else maxIndex a
      l < minIndex a or h > maxIndex a => error "index out of range"
      for k in l..h repeat qsetelt!(a, k, x)
      x

    reduce(f, a) ==
      empty? a => error "cannot reduce an empty aggregate"
      r := qelt(a, m := minIndex a)
      for k in m+1 .. maxIndex a repeat r := f(r, qelt(a, k))
      r

    reduce(f, a, identity) ==
      for k in minIndex a .. maxIndex a repeat
	identity := f(identity, qelt(a, k))
      identity

    if S has SetCategory then
       reduce(f, a, identity,absorber) ==
	 for k in minIndex a .. maxIndex a while identity ~= absorber
		repeat identity := f(identity, qelt(a, k))
	 identity

-- this is necessary since new has disappeared.
    stupidnew: (NonNegativeInteger, %, %) -> %
    stupidget: List % -> S
-- a and b are not both empty if n > 0
    stupidnew(n, a, b) ==
      zero? n => empty()
      new(n, (empty? a => qelt(b, minIndex b); qelt(a, minIndex a)))
-- at least one element of l must be non-empty
    stupidget l ==
      for a in l repeat
	not empty? a => return first a
      error "Should not happen"

    map(f, a, b) ==
      m := max(minIndex a, minIndex b)
      n := min(maxIndex a, maxIndex b)
      l := max(0, n - m + 1)::NonNegativeInteger
      c := stupidnew(l, a, b)
      for i in minIndex(c).. for j in m..n repeat
	qsetelt!(c, i, f(qelt(a, j), qelt(b, j)))
      c

--  map(f, a, b, x) ==
--    m := min(minIndex a, minIndex b)
--    n := max(maxIndex a, maxIndex b)
--    l := (n - m + 1)::NonNegativeInteger
--    c := new l
--    for i in minIndex(c).. for j in m..n repeat
--	qsetelt!(c, i, f(a(j, x), b(j, x)))
--    c

    merge(f, a, b) ==
      r := stupidnew(#a + #b, a, b)
      i := minIndex a
      m := maxIndex a
      j := minIndex b
      n := maxIndex b
      k := minIndex(r)
      while i <= m and j <= n repeat
	if f(qelt(a, i), qelt(b, j)) then
	  qsetelt!(r, k, qelt(a, i))
	  i := i+1
	else
	  qsetelt!(r, k, qelt(b, j))
	  j := j+1
        k := k + 1
      while i <= m repeat
        qsetelt!(r, k, elt(a, i))
        k := k + 1
        i := i + 1
      while j <= n repeat
        qsetelt!(r, k, elt(b, j))
        k := k + 1
        j := j + 1
      r

    elt(a:%, s:UniversalSegment(Integer)) ==
      l := lo s
      h := if hasHi s then hi s else maxIndex a
      l < minIndex a or h > maxIndex a => error "index out of range"
      r := stupidnew(max(0, h - l + 1)::NonNegativeInteger, a, a)
      for k in minIndex r.. for i in l..h repeat
	qsetelt!(r, k, qelt(a, i))
      r

    insert(a:%, b:%, i:Integer) ==
      m := minIndex b
      n := maxIndex b
      i < m or i > n => error "index out of range"
      y := stupidnew(#a + #b, a, b)
      k := minIndex y
      for j in m..i-1 repeat
	qsetelt!(y, k, qelt(b, j))
        k := k + 1
      for j in minIndex a .. maxIndex a repeat
	qsetelt!(y, k, qelt(a, j))
        k := k + 1
      for j in i..n repeat
        qsetelt!(y, k, qelt(b, j))
        k := k + 1
      y

    copy x ==
      y := stupidnew(#x, x, x)
      for i in minIndex x .. maxIndex x for j in minIndex y .. repeat
	qsetelt!(y, j, qelt(x, i))
      y

    copyInto!(y, x, s) ==
      s < minIndex y or s + #x > maxIndex y + 1 =>
					      error "index out of range"
      for i in minIndex x .. maxIndex x for j in s.. repeat
	qsetelt!(y, j, qelt(x, i))
      y

    construct l ==
--    a := new(#l)
      empty? l => empty()
      a := new(#l, first l)
      for i in minIndex(a).. for x in l repeat qsetelt!(a, i, x)
      a

    delete(a:%, s:UniversalSegment(Integer)) ==
      l := lo s; h := if hasHi s then hi s else maxIndex a
      l < minIndex a or h > maxIndex a => error "index out of range"
      h < l => copy a
      r := stupidnew((#a - h + l - 1)::NonNegativeInteger, a, a)
      k := minIndex(r)
      for i in minIndex a..l-1 repeat
	qsetelt!(r, k, qelt(a, i))
        k := k + 1
      for i in h+1 .. maxIndex a repeat
	qsetelt!(r, k, qelt(a, i))
        k := k + 1
      r

    delete(x:%, i:Integer) ==
      i < minIndex x or i > maxIndex x => error "index out of range"
      y := stupidnew((#x - 1)::NonNegativeInteger, x, x)
      k := minIndex y
      for j in minIndex x..i-1 repeat
	qsetelt!(y, k, qelt(x, j))
        k := k + 1
      for j in i+1 .. maxIndex x repeat
	qsetelt!(y, k, qelt(x, j))
        k := k + 1
      y

    reverse! x ==
      m := minIndex x
      n := maxIndex x
      for i in 0..((n-m) quo 2) repeat swap!(x, m+i, n-i)
      x

    concat l ==
      empty? l => empty()
      n := +/[#a for a in l]
      i := minIndex(r := new(n, stupidget l))
      for a in l repeat
	copyInto!(r, a, i)
	i := i + #a
      r

    sorted?(f, a) ==
      for i in minIndex(a)..maxIndex(a)-1 repeat
	not f(qelt(a, i), qelt(a, i + 1)) => return false
      true

    concat(x:%, y:%) ==
      z := stupidnew(#x + #y, x, y)
      copyInto!(z, x, i := minIndex z)
      copyInto!(z, y, i + #x)
      z

    if S has CoercibleTo(OutputForm) then
      coerce(r:%):OutputForm ==
	bracket commaSeparate
	      [qelt(r, k)::OutputForm for k in minIndex r .. maxIndex r]

    if S has SetCategory then
      x = y ==
	#x ~= #y => false
	for i in minIndex x .. maxIndex x repeat
	  not(qelt(x, i) = qelt(y, i)) => return false
	true

      position(x:S, t:%, s:Integer) ==
	n := maxIndex t
	s < minIndex t or s > n => error "index out of range"
	for k in s..n repeat
	  qelt(t, k) = x => return k
	minIndex(t) - 1

    if S has OrderedSet then
      a < b ==
	for i in minIndex a .. maxIndex a
	  for j in minIndex b .. maxIndex b repeat
	    qelt(a, i) ~= qelt(b, j) => return a.i < b.j
	#a < #b


@

\section{category ELAGG ExtensibleLinearAggregate}

<<category ELAGG ExtensibleLinearAggregate>>=
import Type
import LinearAggregate
import OrderedSet
import Integer
)abbrev category ELAGG ExtensibleLinearAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ An extensible aggregate is one which allows insertion and deletion of entries.
++ These aggregates are models of lists and streams which are represented
++ by linked structures so as to make insertion, deletion, and
++ concatenation efficient. However, access to elements of these
++ extensible aggregates is generally slow since access is made from the end.
++ See \spadtype{FlexibleArray} for an exception.
ExtensibleLinearAggregate(S:Type):Category == LinearAggregate S with
   shallowlyMutable
   concat!: (%,S) -> %
     ++ concat!(u,x) destructively adds element x to the end of u.
   concat!: (%,%) -> %
     ++ concat!(u,v) destructively appends v to the end of u.
     ++ v is unchanged
   delete!: (%,Integer) -> %
     ++ delete!(u,i) destructively deletes the \axiom{i}th element of u.
   delete!: (%,UniversalSegment(Integer)) -> %
     ++ delete!(u,i..j) destructively deletes elements u.i through u.j.
   remove!: (S->Boolean,%) -> %
     ++ remove!(p,u) destructively removes all elements x of
     ++ u such that \axiom{p(x)} is true.
   insert!: (S,%,Integer) -> %
     ++ insert!(x,u,i) destructively inserts x into u at position i.
   insert!: (%,%,Integer) -> %
     ++ insert!(v,u,i) destructively inserts aggregate v into u at position i.
   merge!: ((S,S)->Boolean,%,%) -> %
     ++ merge!(p,u,v) destructively merges u and v using predicate p.
   select!: (S->Boolean,%) -> %
     ++ select!(p,u) destructively changes u by keeping only values x such that
     ++ \axiom{p(x)}.
   if S has SetCategory then
     remove!: (S,%) -> %
       ++ remove!(x,u) destructively removes all values x from u.
     removeDuplicates!: % -> %
       ++ removeDuplicates!(u) destructively removes duplicates from u.
   if S has OrderedSet then merge!: (%,%) -> %
       ++ merge!(u,v) destructively merges u and v in ascending order.
 add
   delete(x:%, i:Integer)	   == delete!(copy x, i)
   delete(x:%, i:UniversalSegment(Integer))	   == delete!(copy x, i)
   remove(f:S -> Boolean, x:%)   == remove!(f, copy x)
   insert(s:S, x:%, i:Integer)   == insert!(s, copy x, i)
   insert(w:%, x:%, i:Integer)   == insert!(copy w, copy x, i)
   select(f, x)		   == select!(f, copy x)
   concat(x:%, y:%)	   == concat!(copy x, y)
   concat(x:%, y:S)	   == concat!(copy x, new(1, y))
   concat!(x:%, y:S)	   == concat!(x, new(1, y))
   if S has SetCategory then
     remove(s:S, x:%)	     == remove!(s, copy x)
     remove!(s:S, x:%)	     == remove!(#1 = s, x)
     removeDuplicates(x:%)   == removeDuplicates!(copy x)

   if S has OrderedSet then
     merge!(x, y) == merge!(_<$S, x, y)

@

\section{category LSAGG ListAggregate}

<<category LSAGG ListAggregate>>=
import Type
import StreamAggregate
import FiniteLinearAggregate
import ExtensibleLinearAggregate
)abbrev category LSAGG ListAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A list aggregate is a model for a linked list data structure.
++ A linked list is a versatile
++ data structure. Insertion and deletion are efficient and
++ searching is a linear operation.
ListAggregate(S:Type): Category == Join(StreamAggregate S,
   FiniteLinearAggregate S, ExtensibleLinearAggregate S) with
      list: S -> %
	++ list(x) returns the list of one element x.
 add
   cycleMax ==> 1000

   mergeSort: ((S, S) -> Boolean, %, Integer) -> %

   sort!(f, l)	      == mergeSort(f, l, #l)
   list x		   == concat(x, empty())
   reduce(f, x)		   ==
     empty? x => error "reducing over an empty list needs the 3 argument form"
     reduce(f, rest x, first x)
   merge(f, p, q)	   == merge!(f, copy p, copy q)

   select!(f, x) ==
     while not empty? x and not f first x repeat x := rest x
     empty? x => x
     y := x
     z := rest y
     while not empty? z repeat
       if f first z then (y := z; z := rest z)
		    else (z := rest z; setrest!(y, z))
     x

   merge!(f, p, q) ==
     empty? p => q
     empty? q => p
     eq?(p, q) => error "cannot merge a list into itself"
     if f(first p, first q)
       then (r := t := p; p := rest p)
       else (r := t := q; q := rest q)
     while not empty? p and not empty? q repeat
       if f(first p, first q)
	 then (setrest!(t, p); t := p; p := rest p)
	 else (setrest!(t, q); t := q; q := rest q)
     setrest!(t, if empty? p then q else p)
     r

   insert!(s:S, x:%, i:Integer) ==
     i < (m := minIndex x) => error "index out of range"
     i = m => concat(s, x)
     y := rest(x, (i - 1 - m)::NonNegativeInteger)
     z := rest y
     setrest!(y, concat(s, z))
     x

   insert!(w:%, x:%, i:Integer) ==
     i < (m := minIndex x) => error "index out of range"
     i = m => concat!(w, x)
     y := rest(x, (i - 1 - m)::NonNegativeInteger)
     z := rest y
     setrest!(y, w)
     concat!(y, z)
     x

   remove!(f:S -> Boolean, x:%) ==
     while not empty? x and f first x repeat x := rest x
     empty? x => x
     p := x
     q := rest x
     while not empty? q repeat
       if f first q then q := setrest!(p, rest q)
		    else (p := q; q := rest q)
     x

   delete!(x:%, i:Integer) ==
     i < (m := minIndex x) => error "index out of range"
     i = m => rest x
     y := rest(x, (i - 1 - m)::NonNegativeInteger)
     setrest!(y, rest(y, 2))
     x

   delete!(x:%, i:UniversalSegment(Integer)) ==
     (l := lo i) < (m := minIndex x) => error "index out of range"
     h := if hasHi i then hi i else maxIndex x
     h < l => x
     l = m => rest(x, (h + 1 - m)::NonNegativeInteger)
     t := rest(x, (l - 1 - m)::NonNegativeInteger)
     setrest!(t, rest(t, (h - l + 2)::NonNegativeInteger))
     x

   find(f, x) ==
     while not empty? x and not f first x repeat x := rest x
     empty? x => "failed"
     first x

   position(f:S -> Boolean, x:%) ==
     k := minIndex(x)
     while not empty? x and not f first x repeat
       x := rest x
       k := k + 1
     empty? x => minIndex(x) - 1
     k

   mergeSort(f, p, n) ==
     if n = 2 and f(first rest p, first p) then p := reverse! p
     n < 3 => p
     l := (n quo 2)::NonNegativeInteger
     q := split!(p, l)
     p := mergeSort(f, p, l)
     q := mergeSort(f, q, n - l)
     merge!(f, p, q)

   sorted?(f, l) ==
     empty? l => true
     p := rest l
     while not empty? p repeat
       not f(first l, first p) => return false
       p := rest(l := p)
     true

   reduce(f, x, i) ==
     r := i
     while not empty? x repeat (r := f(r, first x); x := rest x)
     r

   if S has SetCategory then
      reduce(f, x, i,a) ==
	r := i
	while not empty? x and r ~= a repeat
	  r := f(r, first x)
	  x := rest x
	r

   new(n, s) ==
     l := empty()
     for k in 1..n repeat l := concat(s, l)
     l

   map(f, x, y) ==
     z := empty()
     while not empty? x and not empty? y repeat
       z := concat(f(first x, first y), z)
       x := rest x
       y := rest y
     reverse! z

-- map(f, x, y, d) ==
--   z := empty()
--   while not empty? x and not empty? y repeat
--     z := concat(f(first x, first y), z)
--     x := rest x
--     y := rest y
--   z := reverseInPlace z
--   if not empty? x then
--	z := concat!(z, map(f(#1, d), x))
--   if not empty? y then
--	z := concat!(z, map(f(d, #1), y))
--   z

   reverse! x ==
     empty? x => x
     empty?(y := rest x) => x
     setrest!(x, empty())
     while not empty? y repeat
       z := rest y
       setrest!(y, x)
       x := y
       y := z
     x

   copy x ==
     y := empty()
     for k in 0.. while not empty? x repeat
       k = cycleMax and cyclic? x => error "cyclic list"
       y := concat(first x, y)
       x := rest x
     reverse! y

   copyInto!(y, x, s) ==
     s < (m := minIndex y) => error "index out of range"
     z := rest(y, (s - m)::NonNegativeInteger)
     while not empty? z and not empty? x repeat
       setfirst!(z, first x)
       x := rest x
       z := rest z
     y

   if S has SetCategory then
     position(w, x, s) ==
       s < (m := minIndex x) => error "index out of range"
       x := rest(x, (s - m)::NonNegativeInteger)
       k := s
       while not empty? x and w ~= first x repeat
	 x := rest x
         k := k + 1
       empty? x => minIndex x - 1
       k

     removeDuplicates! l ==
       p := l
       while not empty? p repeat
	 p := setrest!(p, remove!(#1 = first p, rest p))
       l

   if S has OrderedSet then
     x < y ==
	while not empty? x and not empty? y repeat
	  first x ~= first y => return(first x < first y)
	  x := rest x
	  y := rest y
	empty? x => not empty? y
	false

@

\section{category ALAGG AssociationListAggregate}

<<category ALAGG AssociationListAggregate>>=
import SetCategory
import TableAggregate
import ListAggregate
)abbrev category ALAGG AssociationListAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ An association list is a list of key entry pairs which may be viewed
++ as a table.	It is a poor mans version of a table:
++ searching for a key is a linear operation.
AssociationListAggregate(Key:SetCategory,Entry:SetCategory): Category ==
   Join(TableAggregate(Key, Entry), ListAggregate Record(key:Key,entry:Entry)) with
      assoc: (Key, %) -> Union(Record(key:Key,entry:Entry), "failed")
	++ assoc(k,u) returns the element x in association list u stored
	++ with key k, or "failed" if u has no key k.

@

\section{category SRAGG StringAggregate}

<<category SRAGG StringAggregate>>=
import OneDimensionalArrayAggregate Character
import UniversalSegment
import Boolean
import Character
import CharacterClass
import Integer
)abbrev category SRAGG StringAggregate
++ Author: Stephen Watt and Michael Monagan. revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ A string aggregate is a category for strings, that is,
++ one dimensional arrays of characters.
StringAggregate: Category == OneDimensionalArrayAggregate Character with
    lowerCase	    : % -> %
      ++ lowerCase(s) returns the string with all characters in lower case.
    lowerCase!: % -> %
      ++ lowerCase!(s) destructively replaces the alphabetic characters
      ++ in s by lower case.
    upperCase	    : % -> %
      ++ upperCase(s) returns the string with all characters in upper case.
    upperCase!: % -> %
      ++ upperCase!(s) destructively replaces the alphabetic characters
      ++ in s by upper case characters.
    prefix?	    : (%, %) -> Boolean
      ++ prefix?(s,t) tests if the string s is the initial substring of t.
      ++ Note: \axiom{prefix?(s,t) == reduce(and,[s.i = t.i for i in 0..maxIndex s])}.
    suffix?	    : (%, %) -> Boolean
      ++ suffix?(s,t) tests if the string s is the final substring of t.
      ++ Note: \axiom{suffix?(s,t) == reduce(and,[s.i = t.(n - m + i) for i in 0..maxIndex s])}
      ++ where m and n denote the maxIndex of s and t respectively.
    substring?: (%, %, Integer) -> Boolean
      ++ substring?(s,t,i) tests if s is a substring of t beginning at
      ++ index i.
      ++ Note: \axiom{substring?(s,t,0) = prefix?(s,t)}.
    match: (%, %, Character) -> NonNegativeInteger
      ++ match(p,s,wc) tests if pattern \axiom{p} matches subject \axiom{s}
      ++ where \axiom{wc} is a wild card character. If no match occurs,
      ++ the index \axiom{0} is returned; otheriwse, the value returned
      ++ is the first index of the first character in the subject matching
      ++ the subject (excluding that matched by an initial wild-card).
      ++ For example, \axiom{match("*to*","yorktown","*")} returns \axiom{5}
      ++ indicating a successful match starting at index \axiom{5} of
      ++ \axiom{"yorktown"}.
    match?: (%, %, Character) -> Boolean
      ++ match?(s,t,c) tests if s matches t except perhaps for
      ++ multiple and consecutive occurrences of character c.
      ++ Typically c is the blank character.
    replace	    : (%, UniversalSegment(Integer), %) -> %
      ++ replace(s,i..j,t) replaces the substring \axiom{s(i..j)} of s by string t.
    position	    : (%, %, Integer) -> Integer
      ++ position(s,t,i) returns the position j of the substring s in string t,
      ++ where \axiom{j >= i} is required.
    position	    : (CharacterClass, %, Integer) -> Integer
      ++ position(cc,t,i) returns the position \axiom{j >= i} in t of
      ++ the first character belonging to cc.
    coerce	    : Character -> %
      ++ coerce(c) returns c as a string s with the character c.

    split: (%, Character) -> List %
      ++ split(s,c) returns a list of substrings delimited by character c.
    split: (%, CharacterClass) -> List %
      ++ split(s,cc) returns a list of substrings delimited by characters in cc.

    trim: (%, Character) -> %
      ++ trim(s,c) returns s with all characters c deleted from right
      ++ and left ends.
      ++ For example, \axiom{trim(" abc ", char " ")} returns \axiom{"abc"}.
    trim: (%, CharacterClass) -> %
      ++ trim(s,cc) returns s with all characters in cc deleted from right
      ++ and left ends.
      ++ For example, \axiom{trim("(abc)", charClass "()")} returns \axiom{"abc"}.
    leftTrim: (%, Character) -> %
      ++ leftTrim(s,c) returns s with all leading characters c deleted.
      ++ For example, \axiom{leftTrim("  abc  ", char " ")} returns \axiom{"abc  "}.
    leftTrim: (%, CharacterClass) -> %
      ++ leftTrim(s,cc) returns s with all leading characters in cc deleted.
      ++ For example, \axiom{leftTrim("(abc)", charClass "()")} returns \axiom{"abc)"}.
    rightTrim: (%, Character) -> %
      ++ rightTrim(s,c) returns s with all trailing occurrences of c deleted.
      ++ For example, \axiom{rightTrim("  abc  ", char " ")} returns \axiom{"  abc"}.
    rightTrim: (%, CharacterClass) -> %
      ++ rightTrim(s,cc) returns s with all trailing occurences of
      ++ characters in cc deleted.
      ++ For example, \axiom{rightTrim("(abc)", charClass "()")} returns \axiom{"(abc"}.
    elt: (%, %) -> %
      ++ elt(s,t) returns the concatenation of s and t. It is provided to
      ++ allow juxtaposition of strings to work as concatenation.
      ++ For example, \axiom{"smoo" "shed"} returns \axiom{"smooshed"}.
 add
   trim(s: %, c:  Character)	  == leftTrim(rightTrim(s, c),	c)
   trim(s: %, cc: CharacterClass) == leftTrim(rightTrim(s, cc), cc)

   lowerCase s		 == lowerCase! copy s
   upperCase s		 == upperCase! copy s
   prefix?(s, t)	 == substring?(s, t, minIndex t)
   coerce(c:Character):% == new(1, c)
   elt(s:%, t:%): %	 == concat(s,t)$%

@

\section{category BTAGG BitAggregate}

<<category BTAGG BitAggregate>>=
import OrderedSet
import Logic
import OneDimensionalArrayAggregate Boolean
)abbrev category BTAGG BitAggregate
++ Author: Michael Monagan; revised by Manuel Bronstein and Richard Jenks
++ Date Created: August 87 through August 88
++ Date Last Updated: April 1991
++ Basic Operations:
++ Related Constructors:
++ Also See:
++ AMS Classifications:
++ Keywords:
++ References:
++ Description:
++ The bit aggregate category models aggregates representing large
++ quantities of Boolean data.
BitAggregate(): Category ==
  Join(OrderedSet, BooleanLogic, Logic, OneDimensionalArrayAggregate Boolean) with
    nand : (%, %) -> %
      ++ nand(a,b) returns the logical {\em nand} of bit aggregates \axiom{a}
      ++ and \axiom{b}.
    nor	 : (%, %) -> %
      ++ nor(a,b) returns the logical {\em nor} of bit aggregates \axiom{a} and 
      ++ \axiom{b}.
    xor	 : (%, %) -> %
      ++ xor(a,b) returns the logical {\em exclusive-or} of bit aggregates
      ++ \axiom{a} and \axiom{b}.

 add
   not v      == map(_not, v)
   ~ v        == map(_~, v)
   v /\ u     == map(_/_\, v, u)
   v \/ u     == map(_\_/, v, u)
   nand(v, u) == map(nand, v, u)
   nor(v, u)  == map(nor, v, u)

@

\section{License}

<<license>>=
--Copyright (c) 1991-2002, The Numerical ALgorithms Group Ltd.
--All rights reserved.
--Copyright (C) 2007-2010, Gabriel Dos Reis.
--All rights reserved.
--
--Redistribution and use in source and binary forms, with or without
--modification, are permitted provided that the following conditions are
--met:
--
--    - Redistributions of source code must retain the above copyright
--      notice, this list of conditions and the following disclaimer.
--
--    - Redistributions in binary form must reproduce the above copyright
--      notice, this list of conditions and the following disclaimer in
--      the documentation and/or other materials provided with the
--      distribution.
--
--    - Neither the name of The Numerical ALgorithms Group Ltd. nor the
--      names of its contributors may be used to endorse or promote products
--      derived from this software without specific prior written permission.
--
--THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS
--IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
--TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
--PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER
--OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
--EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
--PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
--PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
--LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
--NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
--SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
@
<<*>>=
<<license>>

<<category AGG Aggregate>>
<<category HOAGG HomogeneousAggregate>>
<<category CLAGG Collection>>
<<category BGAGG BagAggregate>>
<<category SKAGG StackAggregate>>
<<category QUAGG QueueAggregate>>
<<category DQAGG DequeueAggregate>>
<<category PRQAGG PriorityQueueAggregate>>
<<category DIOPS DictionaryOperations>>
<<category DIAGG Dictionary>>
<<category MDAGG MultiDictionary>>
<<category SETAGG SetAggregate>>
<<category FSAGG FiniteSetAggregate>>
<<category MSETAGG MultisetAggregate>>
<<category OMSAGG OrderedMultisetAggregate>>
<<category KDAGG KeyedDictionary>>
<<category ELTAB Eltable>>
<<category ELTAGG EltableAggregate>>
<<category IXAGG IndexedAggregate>>
<<category TBAGG TableAggregate>>
<<category RCAGG RecursiveAggregate>>
<<category BRAGG BinaryRecursiveAggregate>>
<<category DLAGG DoublyLinkedAggregate>>
<<category URAGG UnaryRecursiveAggregate>>
<<category STAGG StreamAggregate>>
<<category LNAGG LinearAggregate>>
<<category FLAGG FiniteLinearAggregate>>
<<category A1AGG OneDimensionalArrayAggregate>>
<<category ELAGG ExtensibleLinearAggregate>>
<<category LSAGG ListAggregate>>
<<category ALAGG AssociationListAggregate>>
<<category SRAGG StringAggregate>>
<<category BTAGG BitAggregate>>

@
\eject
\begin{thebibliography}{99}
\bibitem{1} nothing
\end{thebibliography}
\end{document}