Subject: Re: [boost] [gsoc] Interest in BGL v2?
From: Alex Hagen-Zanker (ahh34_at_[hidden])
Date: 2011-03-29 04:37:32
On 29/03/2011 02:20, Jeremiah Willcock wrote:
> On Mon, 28 Mar 2011, Andrew Sutton wrote:
>
> (snip)
>
>>>> Another feature that has been on BGL's want list is the
>>>> implementation of algorithm objects.
>>>
>>> For example?
>>
>> Just a class that wraps the algorithm. There are several in the BGL.
>> The technique offers a little more flexibility when the number of
>> parameters is large.
>
> I think the original intent of algorithm objects for BGL is inversion
> of control flow: instead of something like BFS calling visitor methods
> at event points, you have a BFS object that suspends itself where it
> would have called the visitor, and then is continued by surrounding
> code. This makes it easier to do things like interleave two BFS runs
> at the same time, which you can't do with the existing model (without
> threads or coroutines).
To me that would be the most valuable aspect of algorithm classes. For
the purpose of suspending and restarting the Dijkstra shortest path
algorithm I built a class around existing BGL code that works perfectly
fine for me. I am happy to share it, even though it is surely not up to
Boost standards.
Alex