pgr_dijkstraCost

pgr_dijkstraCost

Using Dijkstra algorithm implemented by Boost.Graph, and extract only the aggregate cost of the shortest path(s) found, for the combination of vertices given.

_images/boost-inside.jpeg

Boost Graph Inside

Availability

  • Version 3.1.0
    • New Proposed functions:
      • pgr_dijkstraCost(combinations)
  • Version 2.2.0
    • New Official function

Description

The pgr_dijkstraCost algorithm, is a good choice to calculate the sum of the costs of the shortest path for a subset of pairs of nodes of the graph. We make use of the Boost’s implementation of dijkstra which runs in \(O(V \log V + E)\) time.

The main characteristics are:
  • It does not return a path.
  • Returns the sum of the costs of the shortest path for pair combination of nodes in the graph.
  • Process is done only on edges with positive costs.
  • Values are returned when there is a path.
    • The returned values are in the form of a set of (start_vid, end_vid, agg_cost).
    • When the starting vertex and ending vertex are the same, there is no path.
      • The agg_cost int the non included values (v, v) is 0
    • When the starting vertex and ending vertex are the different and there is no path.
      • The agg_cost in the non included values (u, v) is \(\infty\)
  • Let be the case the values returned are stored in a table, so the unique index would be the pair: (start_vid, end_vid).
  • For undirected graphs, the results are symmetric.
    • The agg_cost of (u, v) is the same as for (v, u).
  • Any duplicated value in the start_vids or end_vids is ignored.
  • The returned values are ordered:
    • start_vid ascending
    • end_vid ascending
  • Running time: \(O(| start\_vids | * (V \log V + E))\)

Signatures

Summary

pgr_dijkstraCost(edges_sql, from_vid,  to_vid  [, directed])
pgr_dijkstraCost(edges_sql, from_vid,  to_vids [, directed])
pgr_dijkstraCost(edges_sql, from_vids, to_vid  [, directed])
pgr_dijkstraCost(edges_sql, from_vids, to_vids [, directed])
pgr_dijkstraCost(edges_sql, combinations_sql   [, directed]) -- Proposed on v3.1
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET

Using defaults

pgr_dijkstraCost(edges_sql, from_vid,  to_vid)
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET
Example:From vertex \(2\) to vertex \(3\) on a directed graph
SELECT * FROM pgr_dijkstraCost(
    'select id, source, target, cost, reverse_cost from edge_table',
    2, 3);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         2 |       3 |        5
(1 row)

One to One

pgr_dijkstraCost(edges_sql, from_vid,  to_vid  [, directed])
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET
Example:From vertex \(2\) to vertex \(3\) on an undirected graph
SELECT * FROM pgr_dijkstraCost(
    'select id, source, target, cost, reverse_cost from edge_table',
    2, 3, false);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         2 |       3 |        1
(1 row)

One to Many

pgr_dijkstraCost(edges_sql, from_vid,  to_vids [, directed])
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET
Example:From vertex \(2\) to vertices \(\{3, 11\}\) on a directed graph
SELECT * FROM pgr_dijkstraCost(
    'select id, source, target, cost, reverse_cost from edge_table',
    2, ARRAY[3, 11]);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         2 |       3 |        5
         2 |      11 |        3
(2 rows)

Many to One

pgr_dijkstraCost(edges_sql, from_vids, to_vid  [, directed])
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET
Example:From vertices \(\{2, 7\}\) to vertex \(3\) on a directed graph
SELECT * FROM pgr_dijkstraCost(
    'select id, source, target, cost, reverse_cost from edge_table',
    ARRAY[2, 7], 3);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         2 |       3 |        5
         7 |       3 |        6
(2 rows)

Many to Many

pgr_dijkstraCost(edges_sql, from_vids, to_vids [, directed])
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET
Example:From vertices \(\{2, 7\}\) to vertices \(\{3, 11\}\) on a directed graph
SELECT * FROM pgr_dijkstraCost(
    'select id, source, target, cost, reverse_cost from edge_table',
    ARRAY[2, 7], ARRAY[3, 11]);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         2 |       3 |        5
         2 |      11 |        3
         7 |       3 |        6
         7 |      11 |        4
(4 rows)

Combinations

pgr_dijkstraCost(TEXT edges_sql, TEXT combination_sql, BOOLEAN directed:=true);
RETURNS SET OF (start_vid, end_vid, agg_cost)
OR EMPTY SET
Example:Using a combinations table on an undirected graph
SELECT * FROM pgr_dijkstraCost(
    'SELECT id, source, target, cost, reverse_cost FROM edge_table',
    'SELECT source, target FROM combinations_table',
    FALSE
);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         1 |       2 |        1
         1 |       4 |        3
         2 |       1 |        1
         2 |       4 |        2
(4 rows)

Parameters

Parameter Type Default Description
Edges SQL TEXT   Edges query as described below
Combinations SQL TEXT   Combinations query as described below
start_vid BIGINT   Identifier of the starting vertex of the path.
start_vids ARRAY[BIGINT]   Array of identifiers of starting vertices.
end_vid BIGINT   Identifier of the ending vertex of the path.
end_vids ARRAY[BIGINT]   Array of identifiers of ending vertices.
directed BOOLEAN true
  • When true Graph is considered Directed
  • When false the graph is considered as Undirected.

Inner query

Edges query

Column Type Default Description
id ANY-INTEGER   Identifier of the edge.
source ANY-INTEGER   Identifier of the first end point vertex of the edge.
target ANY-INTEGER   Identifier of the second end point vertex of the edge.
cost ANY-NUMERICAL  

Weight of the edge (source, target)

  • When negative: edge (source, target) does not exist, therefore it’s not part of the graph.
reverse_cost ANY-NUMERICAL -1

Weight of the edge (target, source),

  • When negative: edge (target, source) does not exist, therefore it’s not part of the graph.

Where:

ANY-INTEGER:SMALLINT, INTEGER, BIGINT
ANY-NUMERICAL:SMALLINT, INTEGER, BIGINT, REAL, FLOAT

Combinations query

Column Type Default Description
source ANY-INTEGER   Identifier of the first end point vertex of the edge.
target ANY-INTEGER   Identifier of the second end point vertex of the edge.

Where:

ANY-INTEGER:SMALLINT, INTEGER, BIGINT

Return Columns

Returns SET OF (start_vid, end_vid, agg_cost)

Column Type Description
start_vid BIGINT Identifier of the starting vertex.
end_vid BIGINT Identifier of the ending vertex.
agg_cost FLOAT Aggregate cost from start_vid to end_vid.

Additional Examples

Example 1:Demonstration of repeated values are ignored, and result is sorted.
SELECT * FROM pgr_dijkstraCost(
        'select id, source, target, cost, reverse_cost from edge_table',
            ARRAY[5, 3, 4, 3, 3, 4], ARRAY[3, 5, 3, 4]);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         3 |       4 |        3
         3 |       5 |        2
         4 |       3 |        1
         4 |       5 |        3
         5 |       3 |        4
         5 |       4 |        3
(6 rows)

Example 2:Making start_vids the same as end_vids
SELECT * FROM pgr_dijkstraCost(
        'select id, source, target, cost, reverse_cost from edge_table',
            ARRAY[5, 3, 4], ARRAY[5, 3, 4]);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         3 |       4 |        3
         3 |       5 |        2
         4 |       3 |        1
         4 |       5 |        3
         5 |       3 |        4
         5 |       4 |        3
(6 rows)

Example 3:Four manually assigned (source, target) vertex combinations
SELECT * FROM pgr_dijkstraCost(
    'SELECT id, source, target, cost FROM edge_table',
    'SELECT * FROM (VALUES (2, 3), (2, 5), (11, 3), (11, 5)) AS combinations (source, target)',
    FALSE
);
 start_vid | end_vid | agg_cost
-----------+---------+----------
         2 |       3 |        3
         2 |       5 |        1
        11 |       3 |        2
        11 |       5 |        2
(4 rows)