DiagrammeR (version 1.0.0)

do_bfs: Use the breadth-first search (bfs) algorithm

Description

With a chosen or random node serving as the starting point, perform a breadth-first search of the whole graph and return the node ID values visited. The bfs algorithm differs from depth-first search (dfs) in that bfs will follow tree branches branches one level at a time until terminating at leaf node (dfs traverses branches as far as possible).

Usage

do_bfs(graph, node = NULL, direction = "all")

Arguments

graph

a graph object of class dgr_graph that is created using create_graph.

node

an optional node ID value to specify a single starting point for the bfs. If not provided, a random node from the graph will be chosen.

direction

using all (the default), the bfs will ignore edge direction while traversing through the graph. With out and in, traversals between adjacent nodes will respect the edge direction.

Value

a vector containing node ID values for nodes visited during the breadth-first search. The order of the node IDs corresponds to the order visited.

Examples

Run this code
# NOT RUN {
# Create a graph containing
# two balanced trees
graph <-
  create_graph() %>%
  add_balanced_tree(
    k = 2, h = 2) %>%
  add_balanced_tree(
    k = 3, h = 2)

# Perform a breadth-first
# search of the graph,
# beginning at the root node
# `1` (the default
# `direction = "all"` doesn't
# take edge direction into
# account)
graph %>%
  do_bfs(node = 1)

# If not specifying a
# starting node, the function
# will begin the search from
# a random node
graph %>%
  do_bfs()

# It's also possible to
# perform bfs while taking
# into account edge direction;
# using `direction = "in"`
# causes the bfs routine to
# visit nodes along inward edges
graph %>%
  do_bfs(
    node = 1,
    direction = "in")

# Using `direction = "out"`
# results in the bfs moving
# along solely outward edges
graph %>%
  do_bfs(
    node = 1,
    direction = "out")
# }

Run the code above in your browser using DataCamp Workspace