Copyright | (c) Andrey Mokhov 2016-2022 |
---|---|
License | MIT (see the file LICENSE) |
Maintainer | andrey.mokhov@gmail.com |
Stability | experimental |
Safe Haskell | None |
Language | Haskell2010 |
Alga is a library for algebraic construction and manipulation of graphs in Haskell. See this paper for the motivation behind the library, the underlying theory, and implementation details.
This module defines the data type Graph
for algebraic graphs that are known
to be non-empty at compile time. To avoid name clashes with Algebra.Graph,
this module can be imported qualified:
import qualified Algebra.Graph.NonEmpty as NonEmpty
The naming convention generally follows that of Data.List.NonEmpty: we use
suffix 1
to indicate the functions whose interface must be changed compared
to Algebra.Graph, e.g. vertices1
.
Synopsis
- data Graph a
- toNonEmpty :: Graph a -> Maybe (Graph a)
- vertex :: a -> Graph a
- edge :: a -> a -> Graph a
- overlay :: Graph a -> Graph a -> Graph a
- overlay1 :: Graph a -> Graph a -> Graph a
- connect :: Graph a -> Graph a -> Graph a
- vertices1 :: NonEmpty a -> Graph a
- edges1 :: NonEmpty (a, a) -> Graph a
- overlays1 :: NonEmpty (Graph a) -> Graph a
- connects1 :: NonEmpty (Graph a) -> Graph a
- foldg1 :: (a -> b) -> (b -> b -> b) -> (b -> b -> b) -> Graph a -> b
- isSubgraphOf :: Ord a => Graph a -> Graph a -> Bool
- (===) :: Eq a => Graph a -> Graph a -> Bool
- size :: Graph a -> Int
- hasVertex :: Eq a => a -> Graph a -> Bool
- hasEdge :: Eq a => a -> a -> Graph a -> Bool
- vertexCount :: Ord a => Graph a -> Int
- edgeCount :: Ord a => Graph a -> Int
- vertexList1 :: Ord a => Graph a -> NonEmpty a
- edgeList :: Ord a => Graph a -> [(a, a)]
- vertexSet :: Ord a => Graph a -> Set a
- edgeSet :: Ord a => Graph a -> Set (a, a)
- path1 :: NonEmpty a -> Graph a
- circuit1 :: NonEmpty a -> Graph a
- clique1 :: NonEmpty a -> Graph a
- biclique1 :: NonEmpty a -> NonEmpty a -> Graph a
- star :: a -> [a] -> Graph a
- stars1 :: NonEmpty (a, [a]) -> Graph a
- tree :: Tree a -> Graph a
- mesh1 :: NonEmpty a -> NonEmpty b -> Graph (a, b)
- torus1 :: NonEmpty a -> NonEmpty b -> Graph (a, b)
- removeVertex1 :: Eq a => a -> Graph a -> Maybe (Graph a)
- removeEdge :: Eq a => a -> a -> Graph a -> Graph a
- replaceVertex :: Eq a => a -> a -> Graph a -> Graph a
- mergeVertices :: (a -> Bool) -> a -> Graph a -> Graph a
- splitVertex1 :: Eq a => a -> NonEmpty a -> Graph a -> Graph a
- transpose :: Graph a -> Graph a
- induce1 :: (a -> Bool) -> Graph a -> Maybe (Graph a)
- induceJust1 :: Graph (Maybe a) -> Maybe (Graph a)
- simplify :: Ord a => Graph a -> Graph a
- sparsify :: Graph a -> Graph (Either Int a)
- sparsifyKL :: Int -> Graph Int -> Graph
- box :: Graph a -> Graph b -> Graph (a, b)
Non-empty algebraic graphs
Non-empty algebraic graphs, which are constructed using three primitives:
vertex
, overlay
and connect
. See module Algebra.Graph for algebraic
graphs that can be empty.
We define a Num
instance as a convenient notation for working with graphs:
0 ==vertex
0 1 + 2 ==overlay
(vertex
1) (vertex
2) 1 * 2 ==connect
(vertex
1) (vertex
2) 1 + 2 * 3 ==overlay
(vertex
1) (connect
(vertex
2) (vertex
3)) 1 * (2 + 3) ==connect
(vertex
1) (overlay
(vertex
2) (vertex
3))
Note: the signum
method of the type class Num
cannot be implemented and
will throw an error. Furthermore, the Num
instance does not satisfy several
"customary laws" of Num
, which dictate that fromInteger
0
and
fromInteger
1
should act as additive and multiplicative identities, and
negate
as additive inverse. Nevertheless, overloading fromInteger
, +
and
*
is very convenient when working with algebraic graphs; we hope that in
future Haskell's Prelude will provide a more fine-grained class hierarchy for
algebraic structures, which we would be able to utilise without violating any
laws.
The Eq
instance satisfies the following laws of non-empty algebraic graphs.
overlay
is commutative, associative and idempotent:x + y == y + x x + (y + z) == (x + y) + z x + x == x
connect
is associative:x * (y * z) == (x * y) * z
connect
distributes overoverlay
:x * (y + z) == x * y + x * z (x + y) * z == x * z + y * z
connect
can be decomposed:x * y * z == x * y + x * z + y * z
connect
satisfies absorption and saturation:x * y + x + y == x * y x * x * x == x * x
When specifying the time and memory complexity of graph algorithms, n will
denote the number of vertices in the graph, m will denote the number of
edges in the graph, and s will denote the size of the corresponding Graph
expression, defined as the number of vertex leaves (note that n <= s). If
g
is a Graph
, the corresponding n, m and s can be computed as follows:
n ==vertexCount
g m ==edgeCount
g s ==size
g
Converting a Graph
to the corresponding
AdjacencyMap
takes O(s + m * log(m)) time and O(s + m) memory. This is also the
complexity of the graph equality test, because it is currently implemented by
converting graph expressions to canonical representations based on adjacency
maps.
The total order Ord
on graphs is defined using size-lexicographic comparison:
- Compare the number of vertices. In case of a tie, continue.
- Compare the sets of vertices. In case of a tie, continue.
- Compare the number of edges. In case of a tie, continue.
- Compare the sets of edges.
Here are a few examples:
vertex
1 <vertex
2vertex
3 <edge
1 2vertex
1 <edge
1 1edge
1 1 <edge
1 2edge
1 2 <edge
1 1 +edge
2 2edge
1 2 <edge
1 3
Note that the resulting order refines the isSubgraphOf
relation and is
compatible with overlay
and connect
operations:
isSubgraphOf
x y ==> x <= y
x <= x + y x + y <= x * y
Instances
Basic graph construction primitives
vertex :: a -> Graph a Source #
Construct the graph comprising a single isolated vertex. An alias for the
constructor Vertex
.
hasVertex
x (vertex y) == (x == y)vertexCount
(vertex x) == 1edgeCount
(vertex x) == 0size
(vertex x) == 1
edge :: a -> a -> Graph a Source #
Construct the graph comprising a single edge.
edge x y ==connect
(vertex
x) (vertex
y)hasEdge
x y (edge x y) == TrueedgeCount
(edge x y) == 1vertexCount
(edge 1 1) == 1vertexCount
(edge 1 2) == 2
overlay :: Graph a -> Graph a -> Graph a Source #
Overlay two graphs. An alias for the constructor Overlay
. This is a
commutative, associative and idempotent operation.
Complexity: O(1) time and memory, O(s1 + s2) size.
hasVertex
z (overlay x y) ==hasVertex
z x ||hasVertex
z yvertexCount
(overlay x y) >=vertexCount
xvertexCount
(overlay x y) <=vertexCount
x +vertexCount
yedgeCount
(overlay x y) >=edgeCount
xedgeCount
(overlay x y) <=edgeCount
x +edgeCount
ysize
(overlay x y) ==size
x +size
yvertexCount
(overlay 1 2) == 2edgeCount
(overlay 1 2) == 0
overlay1 :: Graph a -> Graph a -> Graph a Source #
Overlay a possibly empty graph (from Algebra.Graph) with a non-empty
graph. If the first argument is empty
, the function returns the second
argument; otherwise it is semantically the same as overlay
.
Complexity: O(s1) time and memory, and O(s1 + s2) size.
overlay1empty
x == x x /=empty
==> overlay1 x y == overlay (fromJust $ toNonEmpty x) y
connect :: Graph a -> Graph a -> Graph a Source #
Connect two graphs. An alias for the constructor Connect
. This is an
associative operation, which distributes over overlay
and obeys the
decomposition axiom.
Complexity: O(1) time and memory, O(s1 + s2) size. Note that the number
of edges in the resulting graph is quadratic with respect to the number of
vertices of the arguments: m = O(m1 + m2 + n1 * n2).
hasVertex
z (connect x y) ==hasVertex
z x ||hasVertex
z yvertexCount
(connect x y) >=vertexCount
xvertexCount
(connect x y) <=vertexCount
x +vertexCount
yedgeCount
(connect x y) >=edgeCount
xedgeCount
(connect x y) >=edgeCount
yedgeCount
(connect x y) >=vertexCount
x *vertexCount
yedgeCount
(connect x y) <=vertexCount
x *vertexCount
y +edgeCount
x +edgeCount
ysize
(connect x y) ==size
x +size
yvertexCount
(connect 1 2) == 2edgeCount
(connect 1 2) == 1
overlays1 :: NonEmpty (Graph a) -> Graph a Source #
Overlay a given list of graphs. Complexity: O(L) time and memory, and O(S) size, where L is the length of the given list, and S is the sum of sizes of the graphs in the list.
overlays1 [x] == x
overlays1 [x,y] == overlay
x y
connects1 :: NonEmpty (Graph a) -> Graph a Source #
Connect a given list of graphs. Complexity: O(L) time and memory, and O(S) size, where L is the length of the given list, and S is the sum of sizes of the graphs in the list.
connects1 [x] == x
connects1 [x,y] == connect
x y
Graph folding
foldg1 :: (a -> b) -> (b -> b -> b) -> (b -> b -> b) -> Graph a -> b Source #
Generalised graph folding: recursively collapse a Graph
by
applying the provided functions to the leaves and internal nodes of the
expression. The order of arguments is: vertex, overlay and connect.
Complexity: O(s) applications of the given functions. As an example, the
complexity of size
is O(s), since const
and +
have constant costs.
foldg1vertex
overlay
connect
== id foldg1vertex
overlay
(flip
connect
) ==transpose
foldg1 (const
1) (+) (+) ==size
foldg1 (== x) (||) (||) ==hasVertex
x
Relations on graphs
isSubgraphOf :: Ord a => Graph a -> Graph a -> Bool Source #
The isSubgraphOf
function takes two graphs and returns True
if the
first graph is a subgraph of the second.
Complexity: O(s + m * log(m)) time. Note that the number of edges m of a
graph can be quadratic with respect to the expression size s.
isSubgraphOf x (overlay
x y) == True isSubgraphOf (overlay
x y) (connect
x y) == True isSubgraphOf (path1
xs) (circuit1
xs) == True isSubgraphOf x y ==> x <= y
(===) :: Eq a => Graph a -> Graph a -> Bool infix 4 Source #
Structural equality on graph expressions. Complexity: O(s) time.
x === x == True x + y === x + y == True 1 + 2 === 2 + 1 == False x + y === x * y == False
Graph properties
size :: Graph a -> Int Source #
The size of a graph, i.e. the number of leaves of the expression. Complexity: O(s) time.
size (vertex
x) == 1 size (overlay
x y) == size x + size y size (connect
x y) == size x + size y size x >= 1 size x >=vertexCount
x
hasVertex :: Eq a => a -> Graph a -> Bool Source #
Check if a graph contains a given vertex. Complexity: O(s) time.
hasVertex x (vertex
y) == (x == y)
edgeList :: Ord a => Graph a -> [(a, a)] Source #
The sorted list of edges of a graph. Complexity: O(s + m * log(m)) time and O(m) memory. Note that the number of edges m of a graph can be quadratic with respect to the expression size s.
edgeList (vertex
x) == [] edgeList (edge
x y) == [(x,y)] edgeList (star
2 [3,1]) == [(2,1), (2,3)] edgeList .edges1
==nub
.sort
.toList
edgeList .transpose
==sort
.map
swap
. edgeList
Standard families of graphs
clique1 :: NonEmpty a -> Graph a Source #
The clique on a list of vertices. Complexity: O(L) time, memory and size, where L is the length of the given list.
clique1 [x] ==vertex
x clique1 [x,y] ==edge
x y clique1 [x,y,z] ==edges1
[(x,y), (x,z), (y,z)] clique1 (xs<>
ys) ==connect
(clique1 xs) (clique1 ys) clique1 .reverse
==transpose
. clique1
stars1 :: NonEmpty (a, [a]) -> Graph a Source #
The stars formed by overlaying a non-empty list of star
s.
Complexity: O(L) time, memory and size, where L is the total size of the
input.
stars1 [(x, [] )] ==vertex
x stars1 [(x, [y])] ==edge
x y stars1 [(x, ys )] ==star
x ys stars1 ==overlays1
.fmap
(uncurry
star
)overlay
(stars1 xs) (stars1 ys) == stars1 (xs<>
ys)
tree :: Tree a -> Graph a Source #
The tree graph constructed from a given Tree
data structure.
Complexity: O(T) time, memory and size, where T is the size of the
given tree (i.e. the number of vertices in the tree).
tree (Node x []) ==vertex
x tree (Node x [Node y [Node z []]]) ==path1
[x,y,z] tree (Node x [Node y [], Node z []]) ==star
x [y,z] tree (Node 1 [Node 2 [], Node 3 [Node 4 [], Node 5 []]]) ==edges1
[(1,2), (1,3), (3,4), (3,5)]
mesh1 :: NonEmpty a -> NonEmpty b -> Graph (a, b) Source #
Construct a mesh graph from two lists of vertices. Complexity: O(L1 * L2) time, memory and size, where L1 and L2 are the lengths of the given lists.
mesh1 [x] [y] ==vertex
(x, y) mesh1 xs ys ==box
(path1
xs) (path1
ys) mesh1 [1,2,3] ['a', 'b'] ==edges1
[ ((1,'a'),(1,'b')), ((1,'a'),(2,'a')) , ((1,'b'),(2,'b')), ((2,'a'),(2,'b')) , ((2,'a'),(3,'a')), ((2,'b'),(3,'b')) , ((3,'a'),(3,'b')) ]
torus1 :: NonEmpty a -> NonEmpty b -> Graph (a, b) Source #
Construct a torus graph from two lists of vertices. Complexity: O(L1 * L2) time, memory and size, where L1 and L2 are the lengths of the given lists.
torus1 [x] [y] ==edge
(x,y) (x,y) torus1 xs ys ==box
(circuit1
xs) (circuit1
ys) torus1 [1,2] ['a', 'b'] ==edges1
[ ((1,'a'),(1,'b')), ((1,'a'),(2,'a')) , ((1,'b'),(1,'a')), ((1,'b'),(2,'b')) , ((2,'a'),(1,'a')), ((2,'a'),(2,'b')) , ((2,'b'),(1,'b')), ((2,'b'),(2,'a')) ]
Graph transformation
removeVertex1 :: Eq a => a -> Graph a -> Maybe (Graph a) Source #
Remove a vertex from a given graph. Returns Nothing
if the resulting
graph is empty.
Complexity: O(s) time, memory and size.
removeVertex1 x (vertex
x) == Nothing removeVertex1 1 (vertex
2) == Just (vertex
2) removeVertex1 x (edge
x x) == Nothing removeVertex1 1 (edge
1 2) == Just (vertex
2) removeVertex1 x>=>
removeVertex1 x == removeVertex1 x
replaceVertex :: Eq a => a -> a -> Graph a -> Graph a Source #
The function replaceVertex
x y
replaces vertex x
with vertex y
in a
given Graph
. If y
already exists, x
and y
will be merged.
Complexity: O(s) time, memory and size.
replaceVertex x x == id replaceVertex x y (vertex
x) ==vertex
y replaceVertex x y ==mergeVertices
(== x) y
mergeVertices :: (a -> Bool) -> a -> Graph a -> Graph a Source #
Merge vertices satisfying a given predicate into a given vertex. Complexity: O(s) time, memory and size, assuming that the predicate takes constant time.
mergeVertices (const
False) x == id mergeVertices (== x) y ==replaceVertex
x y mergeVerticeseven
1 (0 * 2) == 1 * 1 mergeVerticesodd
1 (3 + 4 * 5) == 4 * 1
splitVertex1 :: Eq a => a -> NonEmpty a -> Graph a -> Graph a Source #
Split a vertex into a list of vertices with the same connectivity. Complexity: O(s + k * L) time, memory and size, where k is the number of occurrences of the vertex in the expression and L is the length of the given list.
splitVertex1 x [x] == id
splitVertex1 x [y] == replaceVertex
x y
splitVertex1 1 [0,1] $ 1 * (2 + 3) == (0 + 1) * (2 + 3)
induce1 :: (a -> Bool) -> Graph a -> Maybe (Graph a) Source #
Construct the induced subgraph of a given graph by removing the
vertices that do not satisfy a given predicate. Returns Nothing
if the
resulting graph is empty.
Complexity: O(s) time, memory and size, assuming that the predicate takes
constant time.
induce1 (const
True ) x == Just x induce1 (const
False) x == Nothing induce1 (/= x) ==removeVertex1
x induce1 p>=>
induce1 q == induce1 (\x -> p x && q x)
induceJust1 :: Graph (Maybe a) -> Maybe (Graph a) Source #
Construct the induced subgraph of a given graph by removing the vertices
that are Nothing
. Returns Nothing
if the resulting graph is empty.
Complexity: O(s) time, memory and size.
induceJust1 (vertex
Nothing
) ==Nothing
induceJust1 (edge
(Just
x)Nothing
) ==Just
(vertex
x) induceJust1 .fmap
Just
==Just
induceJust1 .fmap
(\x -> if p x thenJust
x elseNothing
) ==induce1
p
simplify :: Ord a => Graph a -> Graph a Source #
Simplify a graph expression. Semantically, this is the identity function, but it simplifies a given expression according to the laws of the algebra. The function does not compute the simplest possible expression, but uses heuristics to obtain useful simplifications in reasonable time. Complexity: the function performs O(s) graph comparisons. It is guaranteed that the size of the result does not exceed the size of the given expression.
simplify == idsize
(simplify x) <=size
x simplify 1===
1 simplify (1 + 1)===
1 simplify (1 + 2 + 1)===
1 + 2 simplify (1 * 1 * 1)===
1 * 1
sparsify :: Graph a -> Graph (Either Int a) Source #
Sparsify a graph by adding intermediate Left
Int
vertices between the
original vertices (wrapping the latter in Right
) such that the resulting
graph is sparse, i.e. contains only O(s) edges, but preserves the
reachability relation between the original vertices. Sparsification is useful
when working with dense graphs, as it can reduce the number of edges from
O(n^2) down to O(n) by replacing cliques, bicliques and similar densely
connected structures by sparse subgraphs built out of intermediate vertices.
Complexity: O(s) time, memory and size.
sort
.reachable
x ==sort
.rights
.reachable
(Right
x) . sparsifyvertexCount
(sparsify x) <=vertexCount
x +size
x + 1edgeCount
(sparsify x) <= 3 *size
xsize
(sparsify x) <= 3 *size
x
sparsifyKL :: Int -> Graph Int -> Graph Source #
Sparsify a graph whose vertices are integers in the range [1..n]
, where
n
is the first argument of the function, producing an array-based graph
representation from Data.Graph (introduced by King and Launchbury, hence
the name of the function). In the resulting graph, vertices [1..n]
correspond to the original vertices, and all vertices greater than n
are
introduced by the sparsification procedure.
Complexity: O(s) time and memory. Note that thanks to sparsification, the resulting graph has a linear number of edges with respect to the size of the original algebraic representation even though the latter can potentially contain a quadratic O(s^2) number of edges.
sort
.reachable
k ==sort
.filter
(<= n) .flip
reachable
k . sparsifyKL nlength
(vertices
$ sparsifyKL n x) <=vertexCount
x +size
x + 1length
(edges
$ sparsifyKL n x) <= 3 *size
x
Graph composition
box :: Graph a -> Graph b -> Graph (a, b) Source #
Compute the Cartesian product of graphs. Complexity: O(s1 * s2) time, memory and size, where s1 and s2 are the sizes of the given graphs.
box (path1
[0,1]) (path1
['a','b']) ==edges1
[ ((0,'a'), (0,'b')) , ((0,'a'), (1,'a')) , ((0,'b'), (1,'b')) , ((1,'a'), (1,'b')) ]
Up to isomorphism between the resulting vertex types, this operation is
commutative, associative, distributes over overlay
, and has
singleton graphs as identities. Below ~~
stands for equality up to an
isomorphism, e.g. (x,
()) ~~ x
.
box x y ~~ box y x box x (box y z) ~~ box (box x y) z box x (overlay
y z) ==overlay
(box x y) (box x z) box x (vertex
()) ~~ xtranspose
(box x y) == box (transpose
x) (transpose
y)vertexCount
(box x y) ==vertexCount
x *vertexCount
yedgeCount
(box x y) <=vertexCount
x *edgeCount
y +edgeCount
x *vertexCount
y