Bug in graph/dijkstra_shortest_paths_no_color_map.hpp

[hokix@…]

Hi,

I'm a new user to Boost Graph Library and I'm just start to figure out how to use it. So I apologize if I'm wrong.

While testing dijkstra algorithm, I found a possible bug in graph/dijkstra_shortest_paths_no_color_map.hpp.

In function dijkstra_shortest_paths_no_color_map_no_init, the code does not initialize distance_map[start_vertex] to 0 before push it into vertex_queue, as the document says here http://www.boost.org/doc/libs/1_57_0/libs/graph/doc/dijkstra_shortest_paths_no_color_map.html

So this will always return at the start vertex if the distance_map is initialized to infinity:

if (!distance_compare(min_vertex_distance, distance_infinity)) {
// This is the minimum vertex, so all other vertices are unreachable
return;
}

Hui Xiao

[anonymous]

Replying to hokix@…:

Hi,

I'm a new user to Boost Graph Library and I'm just start to figure out how to use it. So I apologize if I'm wrong.

While testing dijkstra algorithm, I found a possible bug in graph/dijkstra_shortest_paths_no_color_map.hpp.

In function dijkstra_shortest_paths_no_color_map_no_init, the code does not initialize distance_map[start_vertex] to 0 before push it into vertex_queue, as the document says here http://www.boost.org/doc/libs/1_57_0/libs/graph/doc/dijkstra_shortest_paths_no_color_map.html

So this will always return at the start vertex if the distance_map is initialized to infinity:

if (!distance_compare(min_vertex_distance, distance_infinity)) {
// This is the minimum vertex, so all other vertices are unreachable
return;
}

Hui Xiao

I'm sorry, I didn't notice that the initialization is in function dijkstra_shortest_paths_no_color_map and I have to init distance_map[start_vertex] to 0 myself.