Leif Mortenson wrote:
Berin Loritsch wrote:

Leif Mortenson wrote:


The DirectedAcyclicGraphVerifierTestCase is currently failing, I am working
on fixing it. Just so nobody else attacks it at the same time. Hopefully
will have a commit in a couple hours.

Fixed.


Thanks, but that was a hint to hold off as I was just testing a larger fix before
committing it. :-)


Many of the problems I have been having the last few days turned out to be
caused by problems with the way the DirectedAcyclicGraphVerifier was
testing for loops in the dependency graph.

Ok.



I fixed some problems with the construction of the vertices list in
AbstractContainer and modified the graph code so that it is now possible
to display an error message like the following when a cyclic graph is detected:


Component1 -> Component2 -> Component3 -> Component2

I like the ability to see the cycle, but can't you work that out with the algorithm that was there before?

The container also displays the order in which the components will be
initialized and then shutdown when debug output is enabled. This should
make it easier for users to understand the dependencies of their application.

That is a nice feature, so I'd like to keep that.



So far, everything seems to now be working and I am now able to get my application up and running.

I fear that the changes you did would make things a bit less efficient. I actually used to have two different methods for testing if a "graph" had a cycle, and since reduced it into one. I did catch a couple of minor issues in the process, but I would like to keep the topological sort using the same algorithm if at all possible--and fix my implementation.

--
"You know the world is going crazy when the best
rapper is a white guy, the best golfer is a black guy,
The Swiss hold the America's Cup, France is
accusing the US of arrogance, and Germany doesn't want
to go to war. And the 3 most powerful men in America
are named 'Bush', 'Dick', and 'Colon' (sic)".

-----Chris Rock


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to