I am using a p-graph structure in the C programming language. There
are two types of vertices: family and individual. The edges for a
family are "husband", "wife", with multiple "child" edges. The edges
for the individual are "parents" and "marriage". Individuals can only
be connected to families and vice versa. My question is, how can I
properly represent this structure so that I can perform Dijkstra's
shortest path algorithm? Also, what is the specific method to perform
Dijkstra's shortest path algorithm on the p-graph structure, or in a
more general case, how to perform Dijkstra's shortest path algorithm
with multiple types of vertices/edges of equal weight?

---

Request for Question Clarification by maniac-ga on 07 Apr 2004 05:41 PDT

Hello Diddydustin,

Could you please describe your use of a "p-graph structure"? I am
assuming you do not refer to process graphs such as found with a
search like
  "p graph" structure
or at
  http://www.p-graph.com/intro.html

Once you clarify that, we should be able to find several good
references (some w/ source code) on Dijkstra's shortest path
algorithm.
  --Maniac

---

Clarification of Question by diddydustin-ga on 07 Apr 2004 07:18 PDT

Sure,

You are correct, I am not referring to a process graph. I am using
this for a genealogical program. I am reading a file with commands
like, I 13 P 2 M 4, which means: "Individual 13 has parents 2 and is
married in family 4." Or F 4 H 9 W 6 C 4 C 5 means "Family 4 has
husband 9, wife 6, and children 4 5". Once I read in all of this data,
I need to store it in a structure, then perform the shortest path
algorithm to determine how two individuals are connected. For example,
"individual 5 family 2 individual 8 family 6 individual 8" would show
how 5 and 8 are connceted.

Does this clarify?