Spanning trees¶
This module is a collection of algorithms on spanning trees. Also included in the collection are algorithms for minimum spanning trees. See the book [JoynerNguyenCohen2010] for descriptions of spanning tree algorithms, including minimum spanning trees.
See also
Todo
 Rewrite
kruskal()
to use priority queues.  Parallel version of Boruvka’s algorithm.
 Randomized spanning tree construction.
REFERENCES:
[Aldous90]  D. Aldous, The random walk construction of uniform spanning trees, SIAM J Discrete Math 3 (1990), 450465. 
[Broder89]  A. Broder, Generating random spanning trees, Proceedings of the 30th IEEE Symposium on Foundations of Computer Science, 1989, pp. 442447. doi:10.1109/SFCS.1989.63516, <http://www.cs.cmu.edu/~15859n/RelatedWork/BroderGenRanSpanningTrees.pdf>_ 
[CormenEtAl2001]  Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, and Clifford Stein. Introduction to Algorithms. 2nd edition, The MIT Press, 2001. 
[GoodrichTamassia2001]  Michael T. Goodrich and Roberto Tamassia. Data Structures and Algorithms in Java. 2nd edition, John Wiley & Sons, 2001. 
[JoynerNguyenCohen2010]  David Joyner, Minh Van Nguyen, and Nathann Cohen. Algorithmic Graph Theory. 2010, http://code.google.com/p/graphtheoryalgorithmsbook/ 
[Sahni2000]  (1, 2) Sartaj Sahni. Data Structures, Algorithms, and Applications in Java. McGrawHill, 2000. 
Methods¶

sage.graphs.spanning_tree.
boruvka
(G, wfunction=None, check=False, by_weight=True)¶ Minimum spanning tree using Boruvka’s algorithm.
This function assumes that we can only compute minimum spanning trees for undirected graphs. Such graphs can be weighted or unweighted, and they can have multiple edges (since we are computing the minimum spanning tree, only the minimum weight among all \((u,v)\)edges is considered, for each pair of vertices \(u\), \(v\)).
INPUT:
G
– an undirected graph.wfunction
– weight function (default:None
); a function that inputs an edgee
and outputs its weight. An edge has the form(u,v,l)
, whereu
andv
are vertices,l
is a label (that can be of any kind). Thewfunction
can be used to transform the label into a weight. In particular: if
wfunction
is notNone
, the weight of an edgee
iswfunction(e)
;  if
wfunction
isNone
(default) andg
is weighted (that is,g.weighted()==True
), the weight of an edgee=(u,v,l)
isl
, independently on which kind of objectl
is: the ordering of labels relies on Python’s operator<
;  if
wfunction
isNone
andg
is not weighted, we set all weights to 1 (hence, the output can be any spanning tree).
 if
check
– boolean (default:False
); whether to first perform sanity checks on the input graphG
. Default:check=False
. If we togglecheck=True
, the following sanity checks are first performed onG
prior to running Boruvka’s algorithm on that input graph: Is
G
the null graph or graph on one vertex?  Is
G
disconnected?  Is
G
a tree?
By default, we turn off the sanity checks for performance reasons. This means that by default the function assumes that its input graph is connected, and has at least one vertex. Otherwise, you should set
check=True
to perform some sanity checks and preprocessing on the input graph. Is
by_weight
– boolean (default:False
); whether to find MST by using weights of edges provided. Default:by_weight=True
. Ifwfunction
is given, MST is calculated using the weights of edges as per the function. Ifwfunction
isNone
, the weight of an edgee=(u,v,l)
isl
if graph is weighted, or all edge weights are considered1
if graph is unweighted. If we toggleby_weight=False
, all weights are considered as1
and MST is calculated.
OUTPUT:
The edges of a minimum spanning tree of
G
, if one exists, otherwise returns the empty list.See also
EXAMPLES:
An example from pages 727–728 in [Sahni2000]:
sage: from sage.graphs.spanning_tree import boruvka sage: G = Graph({1:{2:28, 6:10}, 2:{3:16, 7:14}, 3:{4:12}, 4:{5:22, 7:18}, 5:{6:25, 7:24}}) sage: G.weighted(True) sage: E = boruvka(G, check=True); E [(1, 6, 10), (2, 7, 14), (3, 4, 12), (4, 5, 22), (5, 6, 25), (2, 3, 16)] sage: boruvka(G, by_weight=True) [(1, 6, 10), (2, 7, 14), (3, 4, 12), (4, 5, 22), (5, 6, 25), (2, 3, 16)] sage: sorted(boruvka(G, by_weight=False)) [(1, 2, 28), (1, 6, 10), (2, 3, 16), (2, 7, 14), (3, 4, 12), (4, 5, 22)]
An example with custom edge labels:
sage: G = Graph([[0,1,1],[1,2,1],[2,0,10]], weighted=True) sage: weight = lambda e:3e[0]e[1] sage: boruvka(G, wfunction=lambda e:3e[0]e[1], by_weight=True) [(0, 2, 10), (1, 2, 1)] sage: boruvka(G, wfunction=lambda e:float(1/e[2]), by_weight=True) [(0, 2, 10), (0, 1, 1)]
An example of disconnected graph with
check
disabled:sage: from sage.graphs.spanning_tree import boruvka sage: G = Graph({1:{2:28}, 3:{4:16}}, weighted=True) sage: boruvka(G, check=False) []

sage.graphs.spanning_tree.
kruskal
(G, wfunction=None, check=False)¶ Minimum spanning tree using Kruskal’s algorithm.
This function assumes that we can only compute minimum spanning trees for undirected graphs. Such graphs can be weighted or unweighted, and they can have multiple edges (since we are computing the minimum spanning tree, only the minimum weight among all \((u,v)\)edges is considered, for each pair of vertices \(u\), \(v\)).
INPUT:
G
– an undirected graph.weight_function
(function)  a function that inputs an edgee
and outputs its weight. An edge has the form(u,v,l)
, whereu
andv
are vertices,l
is a label (that can be of any kind). Theweight_function
can be used to transform the label into a weight. In particular: if
weight_function
is notNone
, the weight of an edgee
isweight_function(e)
;  if
weight_function
isNone
(default) andg
is weighted (that is,g.weighted()==True
), the weight of an edgee=(u,v,l)
isl
, independently on which kind of objectl
is: the ordering of labels relies on Python’s operator<
;  if
weight_function
isNone
andg
is not weighted, we set all weights to 1 (hence, the output can be any spanning tree).
 if
check
– Whether to first perform sanity checks on the input graphG
. Default:check=False
. If we togglecheck=True
, the following sanity checks are first performed onG
prior to running Kruskal’s algorithm on that input graph: Is
G
the null graph?  Is
G
disconnected?  Is
G
a tree?  Does
G
have selfloops?  Does
G
have multiple edges?
By default, we turn off the sanity checks for performance reasons. This means that by default the function assumes that its input graph is connected, and has at least one vertex. Otherwise, you should set
check=True
to perform some sanity checks and preprocessing on the input graph. IfG
has multiple edges or selfloops, the algorithm still works, but the runningtime can be improved if these edges are removed. To further improve the runtime of this function, you should call it directly instead of using it indirectly viasage.graphs.generic_graph.GenericGraph.min_spanning_tree()
. Is
OUTPUT:
The edges of a minimum spanning tree of
G
, if one exists, otherwise returns the empty list.EXAMPLES:
An example from pages 727–728 in [Sahni2000].
sage: from sage.graphs.spanning_tree import kruskal sage: G = Graph({1:{2:28, 6:10}, 2:{3:16, 7:14}, 3:{4:12}, 4:{5:22, 7:18}, 5:{6:25, 7:24}}) sage: G.weighted(True) sage: E = kruskal(G, check=True); E [(1, 6, 10), (3, 4, 12), (2, 7, 14), (2, 3, 16), (4, 5, 22), (5, 6, 25)]
Variants of the previous example.
sage: H = Graph(G.edges(labels=False)) sage: kruskal(H, check=True) [(1, 2, None), (1, 6, None), (2, 3, None), (2, 7, None), (3, 4, None), (4, 5, None)] sage: G.allow_loops(True) sage: G.allow_multiple_edges(True) sage: G Looped multigraph on 7 vertices sage: for i in range(20): ....: u = randint(1, 7) ....: v = randint(1, 7) ....: w = randint(0, 20) ....: G.add_edge(u, v, w) sage: H = copy(G) sage: H Looped multigraph on 7 vertices sage: def sanitize(G): ....: G.allow_loops(False) ....: G.allow_multiple_edges(False, keep_label='min') sage: sanitize(H) sage: H Graph on 7 vertices sage: sum(e[2] for e in kruskal(G, check=True)) == sum(e[2] for e in kruskal(H, check=True)) True
An example from pages 599–601 in [GoodrichTamassia2001].
sage: G = Graph({"SFO":{"BOS":2704, "ORD":1846, "DFW":1464, "LAX":337}, ....: "BOS":{"ORD":867, "JFK":187, "MIA":1258}, ....: "ORD":{"PVD":849, "JFK":740, "BWI":621, "DFW":802}, ....: "DFW":{"JFK":1391, "MIA":1121, "LAX":1235}, ....: "LAX":{"MIA":2342}, ....: "PVD":{"JFK":144}, ....: "JFK":{"MIA":1090, "BWI":184}, ....: "BWI":{"MIA":946}}) sage: G.weighted(True) sage: kruskal(G, check=True) [('JFK', 'PVD', 144), ('BWI', 'JFK', 184), ('BOS', 'JFK', 187), ('LAX', 'SFO', 337), ('BWI', 'ORD', 621), ('DFW', 'ORD', 802), ('BWI', 'MIA', 946), ('DFW', 'LAX', 1235)]
An example from pages 568–569 in [CormenEtAl2001].
sage: G = Graph({"a":{"b":4, "h":8}, "b":{"c":8, "h":11}, ....: "c":{"d":7, "f":4, "i":2}, "d":{"e":9, "f":14}, ....: "e":{"f":10}, "f":{"g":2}, "g":{"h":1, "i":6}, "h":{"i":7}}) sage: G.weighted(True) sage: T = Graph(kruskal(G, check=True), format='list_of_edges') sage: sum(T.edge_labels()) 37 sage: T.is_tree() True
An example with custom edge labels:
sage: G = Graph([[0,1,1],[1,2,1],[2,0,10]], weighted=True) sage: weight = lambda e:3e[0]e[1] sage: sorted(kruskal(G, check=True)) [(0, 1, 1), (1, 2, 1)] sage: sorted(kruskal(G, wfunction=weight, check=True)) [(0, 2, 10), (1, 2, 1)] sage: sorted(kruskal(G, wfunction=weight, check=False)) [(0, 2, 10), (1, 2, 1)]

sage.graphs.spanning_tree.
kruskal_iterator
(G, wfunction=None, check=False)¶ Return an iterator implementation of Kruskal algorithm.
OUTPUT:
The edges of a minimum spanning tree of
G
, one by one.See also
EXAMPLES:
sage: from sage.graphs.spanning_tree import kruskal_iterator sage: G = Graph({1:{2:28, 6:10}, 2:{3:16, 7:14}, 3:{4:12}, 4:{5:22, 7:18}, 5:{6:25, 7:24}}) sage: G.weighted(True) sage: next(kruskal_iterator(G, check=True)) (1, 6, 10)

sage.graphs.spanning_tree.
random_spanning_tree
(self, output_as_graph=False)¶ Return a random spanning tree of the graph.
This uses the AldousBroder algorithm ([Broder89], [Aldous90]) to generate a random spanning tree with the uniform distribution, as follows.
Start from any vertex. Perform a random walk by choosing at every step one neighbor uniformly at random. Every time a new vertex \(j\) is met, add the edge \((i, j)\) to the spanning tree, where \(i\) is the previous vertex in the random walk.
INPUT:
output_as_graph
– boolean (default:False
); whether to return a list of edges or a graph
See also
EXAMPLES:
sage: G = graphs.TietzeGraph() sage: G.random_spanning_tree(output_as_graph=True) Graph on 12 vertices sage: rg = G.random_spanning_tree(); rg # random [(0, 9), (9, 11), (0, 8), (8, 7), (7, 6), (7, 2), (2, 1), (1, 5), (9, 10), (5, 4), (2, 3)] sage: Graph(rg).is_tree() True
A visual example for the grid graph:
sage: G = graphs.Grid2dGraph(6, 6) sage: pos = G.get_pos() sage: T = G.random_spanning_tree(True) sage: T.set_pos(pos) sage: T.show(vertex_labels=False)