Re: [jgrapht-users] JGraphT vs. JUNG

[John S. <js...@gm...>]

Hi Rob,

A major point of distinction used to be that JGraphT's API was based
on generics, whereas JUNG's was not.  However, with the release of
Jung 2.0, this is no longer the case as they have made the API quite
similar to JGraphT.

In general, JGraphT focuses more on the data structure toolkit aspect
so that you can come up with an optimized representations (including
automatic indexing via custom listeners), as well as create
sophisticated views into existing graphs (e.g. induced subgraph).
JUNG's algorithm library is probably quite a bit bigger than JGraphT's
at this point, but I think there are some algorithms that only appear
in one library or the other.  JUNG has a lot more visualization
support.

If someone were interested in unifying the two libraries to get the
best of both, I would be very supportive of that since I am a bit
overloaded in terms of the number of open source projects I'm involved
with these days (Apache Hive, LucidDB, Eigenbase, ...).

If you get a better answer, you might ask the JUNG people to update
this FAQ entry (and I could link it from the JGraphT website):

http://jung.sourceforge.net/faq.html#different

JVS

On 9/29/10, Rob Patro <rob...@gm...> wrote:
> Hi All,
>
> I was wondering if anyone might be able to provide me with a short
> comparison between JGraphT and JUNG.
> What are the primary differences between these two libraries?  How do they
> compare in terms of features and performance?
> How do they compare in terms of API and ease-of-use?
>
> Thanks,
> Rob
>