graph library questions

From: Krishna Padmasola (krishna_at_[hidden])
Date: 2000-09-29 00:15:07

Jeremy,

Since you mentioned that the ggcl mailing will be discontinued, I am 
posting my questions here. I apologize if this is not the appropriate 
list for such questions.

I have some questions regarding the implementation and use of
property 
accessors and vertex/edge descriptors:

- How is a property stored and accessed (I mean the implementation)? 
Does this depend on the type of selector we use for vertex/edge? How 
much space does it take and what is the time complexity for accessing 
the property?

- Are internal properties stored with a vertex/edge? Does the 
vertex/edge descriptor have a way to access a property?

- Is there a way to find a vertex/edge with a certain value for a 
given property? e.g., give me the 'red' vertex? What is the time 
complexity for this operation? (I was not able to find this).

- If my application stores the vertex descriptors (created during 
add_vertex) for future use, under what conditions do they remain
valid 
in future? Does adding new vertices using add_vertex invalidate any 
references to the earlier vertex descriptors ? Is this dependent on 
the kind of VertexList selector in use (listS/vecS)?

- Since any non trivial application will have properties associated 
with the vertices/edges, in your opinion, what is the most efficient 
way to go about doing this? In my application don't know the number
of 
vertices in advance, so currently I am using add_vertex, and adding 
the property to the resulting vertex descriptor.

I tried to find some of the answers by going through the source code, 
but as I am just getting familiar with STL, it is a daunting task at 
the moment :-)

Thanks in advance!

Krishna.