jgrapht-users — JGraphT user list.

Re: [jgrapht-users] Maximally connected component for dependencies?

[H.N. de R. <hnr...@gr...>]

Hi,

no, the ConnectivityInspector ignores the direction of the edges. So if your graph has edges s->t->u and you call connectedSetOf(t), you will get {s, t, u} as the result. If you want to get precisely those nodes that are reachable from t (traversing edges in the "proper" direction only), you can do a breadth first search from t with BreadthFirstInspector.

If you want to do this for many nodes, it may be faster to first calculate the transitive closure of the graph and then check the out-neighbours of your node. The transitive closure G' of G has an edge u->v precisely when there is a path u==>v in G.

Do you have a unique top level term from which all useful terms are reachable, or is your construction more complicated?

Regards,
   Ernst

----- Reply message -----
From: org...@io...
To: <jgr...@li...>
Subject: [jgrapht-users] Maximally connected component for dependencies?
Date: Mon, Dec 30, 2013 21:32


'Lo.

As mentioned previously, I'm using jgrapht to handle dependencies in
a small programming language I'm developing. I was originally only using
the library to handle module imports, but I've since moved to using the
library to track dependencies between all types and terms in the
language.

For example: In the term graph, if a term t contains a reference to a
term u, the graph contains an edge from t -> u.

I'd like to use the connectivity information during code generation to
exclude all unreferenced terms from the final program. Is the
ConnectivityInspector the right way to handle this? Specifically, given 
a "top level" term t, I want all of those terms to which t refers, and 
the terms to which those terms refer, and so on, and so on, excluding
any otherwise unreferenced terms.

I think that basically means instantiating a new ConnectivityInspector
for the term graph and then calling connectedSetOf(t)? 

Am I correct in thinking that, given that the graph is directed and
acyclic, that I won't get terms that depend on t (vertices with
outgoing edges to t)? If that's correct, then presumably I *would*
get those terms if the graph was undirected?

Apologies if these are blindingly obvious questions. I'm not all that
familiar with graph algorithms.

M




------------------------------------------------------------------------------
Rapidly troubleshoot problems before they affect your business. Most IT 
organizations don't have a clear picture of how application performance 
affects their revenue. With AppDynamics, you get 100% visibility into your 
Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
_______________________________________________
jgrapht-users mailing list
jgr...@li...
https://lists.sourceforge.net/lists/listinfo/jgrapht-users

[jgrapht-users] Maximally connected component for dependencies?

[<org...@io...>]

'Lo.

As mentioned previously, I'm using jgrapht to handle dependencies in
a small programming language I'm developing. I was originally only using
the library to handle module imports, but I've since moved to using the
library to track dependencies between all types and terms in the
language.

For example: In the term graph, if a term t contains a reference to a
term u, the graph contains an edge from t -> u.

I'd like to use the connectivity information during code generation to
exclude all unreferenced terms from the final program. Is the
ConnectivityInspector the right way to handle this? Specifically, given 
a "top level" term t, I want all of those terms to which t refers, and 
the terms to which those terms refer, and so on, and so on, excluding
any otherwise unreferenced terms.

I think that basically means instantiating a new ConnectivityInspector
for the term graph and then calling connectedSetOf(t)? 

Am I correct in thinking that, given that the graph is directed and
acyclic, that I won't get terms that depend on t (vertices with
outgoing edges to t)? If that's correct, then presumably I *would*
get those terms if the graph was undirected?

Apologies if these are blindingly obvious questions. I'm not all that
familiar with graph algorithms.

M

[jgrapht-users] Continuous Integration via Travis

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

Hey,

We've set up Travis to monitor JGraphT build/test health.  You can
check the latest results here:

https://travis-ci.org/jgrapht/jgrapht/

I try to always run tests locally before merging any pull request, but
this will give us another level of protection, and should also be
useful in the future if I am ever joined by other committers.

Thanks to Peter Goldstein for reminding me to get this activated!

JVS

Re: [jgrapht-users] Getting an ordered set of edges for graph cycles?

[<org...@io...>]

On Thu, 19 Dec 2013 20:26:38 +0100
"H.N. de Ridder" <hnr...@gr...> wrote:
> 
> If adding an edge u->v to G would create a cycle that wasn't there before, then
> G contains a path v==>u. You can find this path using e.g.
> DijkstraShortestPath, which can tell you both the vertices and edges on the
> path in the proper order (getPath()).
> 
> Hope this helps,
>    Ernst

Ah, thank you. Works perfectly!

M

Re: [jgrapht-users] Getting an ordered set of edges for graph cycles?

[H.N. de R. <hnr...@gr...>]

On Thu, Dec 19, 2013 at 05:26:30PM +0000, org...@io... wrote:
> Hello.
> 
> I'm attempting to use jgrapht to handle dependencies in a small
> programming language.
> 
> Essentially, I want to ensure that a "module" can only import known
> modules and that imports cannot introduce cyclic dependencies. I want
> to give good error messages when the user manages either of the
> previous errors: A list of the import statements that caused the cycle.
> 
> Something like:
> 
> error: cyclic import detected
>      import of N at m.z:23
>   -> import of M at n.z:12
>   -> import of N at m.z:23
> 
> I've tried the following:
> 
> 1. Use a DirectedGraph where modules are the vertices and import
> declarations are the edges. This allows me to catch imports of unknown
> modules by catching IllegalArgumentExceptions raised by addEdge. I
> then run the CycleDetector after each addition of an edge.
> Unfortunately, the CycleDetector can only return a set of vertices
> rather than the edges (so I can't directly tell the user exactly which
> series of import declarations caused the cycle). It also comes with a
> rather scary warning that it might not actually catch all cycles (and
> recommends the use of the StrongConnectivityInspector, but doesn't say
> exactly how to use it and the documentation is non-obvious).

I think the warning for findCyclesContainingVertex(v) means that the Set it
returns might not contain every cycle containing v, but it will contain at
least one cycle containing v. In any case your second approach seems to me
the simpler one:

> 2. Use a DirectedAcyclicGraph. This allows me to catch imports of
> unknown modules by checking to see if an edge could actually be
> inserted (the boolean return value of addDagEdge). It also raises
> exceptions when cycles are introduced, but this gives me even less
> information than the CycleDetector (it just tells me that they've
> happened, not why).

If adding an edge u->v to G would create a cycle that wasn't there before, then
G contains a path v==>u. You can find this path using e.g.
DijkstraShortestPath, which can tell you both the vertices and edges on the
path in the proper order (getPath()).

Hope this helps,
   Ernst

P.S. DirectedAcyclicGraph is in the experimental package. Does anybody know
how reliable it is? The unit test seems quite thorough.

-- 
Information System on Graph Classes and their Inclusions (ISGCI)
http://www.graphclasses.org

[jgrapht-users] Getting an ordered set of edges for graph cycles?

[<org...@io...>]

Hello.

I'm attempting to use jgrapht to handle dependencies in a small
programming language.

Essentially, I want to ensure that a "module" can only import known
modules and that imports cannot introduce cyclic dependencies. I want
to give good error messages when the user manages either of the
previous errors: A list of the import statements that caused the cycle.

Something like:

error: cyclic import detected
     import of N at m.z:23
  -> import of M at n.z:12
  -> import of N at m.z:23

I've tried the following:

1. Use a DirectedGraph where modules are the vertices and import
declarations are the edges. This allows me to catch imports of unknown
modules by catching IllegalArgumentExceptions raised by addEdge. I
then run the CycleDetector after each addition of an edge.
Unfortunately, the CycleDetector can only return a set of vertices
rather than the edges (so I can't directly tell the user exactly which
series of import declarations caused the cycle). It also comes with a
rather scary warning that it might not actually catch all cycles (and
recommends the use of the StrongConnectivityInspector, but doesn't say
exactly how to use it and the documentation is non-obvious).

2. Use a DirectedAcyclicGraph. This allows me to catch imports of
unknown modules by checking to see if an edge could actually be
inserted (the boolean return value of addDagEdge). It also raises
exceptions when cycles are introduced, but this gives me even less
information than the CycleDetector (it just tells me that they've
happened, not why).

Is there some better way to achieve what I want? My ideal scenario is
that I add edges to graphs and upon adding an edge, I get a list of all
the *edges and vertices* that led to a cycle in some well-defined order.

M

[jgrapht-users] JGraphT Version 0.9.0 Released

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

After almost two years since the 0.8.3 release, this is a major
release involving some big renovations to project structure:

* move from Sourceforge to github
* dual licensing (LGPL + EPL)
* use mvn instead of ant for build
* publish official release artifacts to Maven Central

There are also a number of new algorithms and bugfixes; you can see
the full change list here:

https://github.com/jgrapht/jgrapht/blob/jgrapht-0.9.0/HISTORY.md

A huge shout-out goes to Andreas Schnaiter, Isaac Kleinman, Owen
Jacobson, Oliver Schrenk, and Oliver Kopp for their help with the
project modernization efforts.  It has been really gratifying to see
everyone pitch in to keep JGraphT in step with the times.

And as always, a hearty thanks to all who made suggestions and code
contributions for this release, big or small.

JVS

[jgrapht-users] JGraphT 0.9.0 in Maven Central

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

Hey all,

I'm very happy to announce that Isaac Kleinman has successfully
published the official 0.9.0 release artifacts in Maven Central:

http://search.maven.org/#search%7Cga%7C1%7Cjgrapht

I'll be following up with the more traditional portion of the release
(.zip and .tar.gz archive uploads to Sourceforge); once I'm done with
that, I'll send the full release announcement email.

JVS

Re: [jgrapht-users] JGraphT Relicensing Process

[Idrissa D. <idy...@gm...>]

OK I will test using those located in the snapshots repo.

Thanks!
Le 11/12/2013 21:26, John Sichi a écrit :
> Isaac and I are still working on getting the official artifacts
> published into Maven Central, but they should be functionally
> equivalent to those snapshots in Sonatype.
>
> On Wed, Dec 11, 2013 at 11:35 AM, Idrissa DIENG <idy...@gm...> wrote:
>> Hi Szabolcs,
>>
>> thanks for your quite response!
>>
>> Le 11/12/2013 20:33, Szabolcs Besenyei a écrit :
>>
>> Hi!
>>
>> http://oss.sonatype.org/content/repositories/snapshots/org/jgrapht/
>>
>> Szabolcs Besenyei
>>
>>
>> 2013/12/11 Idrissa DIENG <idy...@gm...>
>>> Hi John,
>>>
>>> I saw that the JGraphT 0.9.0 has already released in the GIT development
>>> stream.
>>>
>>> Where I can found please the official download artifacts?
>>>
>>> Le 13/11/2013 10:12, Idrissa DIENG a écrit :
>>>
>>> Hi John,
>>>
>>> Thanks for your quite response. This is very good new.
>>>
>>> Best regards,
>>> Idrissa DIENG
>>>
>>> Le 13/11/2013 00:22, John Sichi a écrit :
>>>
>>> On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG <idy...@gm...> wrote:
>>>> Can you please give us the latest status of relicensing process?
>>>> Exist there an estimation for the next release that should provide
>>>> JGraphT under EPL license? What is the estimation effort?
>>>
>>> Hi Idrissa,
>>>
>>> We got sufficient contributor approval, so the only thing remaining is to
>>> change all the file headers.  There are a few other items which need work
>>> before we can make the 0.9 release including the license change; we're
>>> hoping to have that out before year's end.
>>>
>>> JVS
>>>
>>>
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Rapidly troubleshoot problems before they affect your business. Most IT
>>> organizations don't have a clear picture of how application performance
>>> affects their revenue. With AppDynamics, you get 100% visibility into your
>>> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics
>>> Pro!
>>>
>>> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
>>> _______________________________________________
>>> jgrapht-users mailing list
>>> jgr...@li...
>>> https://lists.sourceforge.net/lists/listinfo/jgrapht-users
>>>
>>

Re: [jgrapht-users] JGraphT Relicensing Process

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

Isaac and I are still working on getting the official artifacts
published into Maven Central, but they should be functionally
equivalent to those snapshots in Sonatype.

On Wed, Dec 11, 2013 at 11:35 AM, Idrissa DIENG <idy...@gm...> wrote:
> Hi Szabolcs,
>
> thanks for your quite response!
>
> Le 11/12/2013 20:33, Szabolcs Besenyei a écrit :
>
> Hi!
>
> http://oss.sonatype.org/content/repositories/snapshots/org/jgrapht/
>
> Szabolcs Besenyei
>
>
> 2013/12/11 Idrissa DIENG <idy...@gm...>
>>
>> Hi John,
>>
>> I saw that the JGraphT 0.9.0 has already released in the GIT development
>> stream.
>>
>> Where I can found please the official download artifacts?
>>
>> Le 13/11/2013 10:12, Idrissa DIENG a écrit :
>>
>> Hi John,
>>
>> Thanks for your quite response. This is very good new.
>>
>> Best regards,
>> Idrissa DIENG
>>
>> Le 13/11/2013 00:22, John Sichi a écrit :
>>
>> On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG <idy...@gm...> wrote:
>>>
>>> Can you please give us the latest status of relicensing process?
>>> Exist there an estimation for the next release that should provide
>>> JGraphT under EPL license? What is the estimation effort?
>>
>>
>> Hi Idrissa,
>>
>> We got sufficient contributor approval, so the only thing remaining is to
>> change all the file headers.  There are a few other items which need work
>> before we can make the 0.9 release including the license change; we're
>> hoping to have that out before year's end.
>>
>> JVS
>>
>>
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Rapidly troubleshoot problems before they affect your business. Most IT
>> organizations don't have a clear picture of how application performance
>> affects their revenue. With AppDynamics, you get 100% visibility into your
>> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics
>> Pro!
>>
>> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
>> _______________________________________________
>> jgrapht-users mailing list
>> jgr...@li...
>> https://lists.sourceforge.net/lists/listinfo/jgrapht-users
>>
>
>

Re: [jgrapht-users] JGraphT Relicensing Process

[Idrissa D. <idy...@gm...>]

Hi Szabolcs,

thanks for your quite response!

Le 11/12/2013 20:33, Szabolcs Besenyei a écrit :
> Hi!
>
> http://oss.sonatype.org/content/repositories/snapshots/org/jgrapht/
>
> Szabolcs Besenyei
>
>
> 2013/12/11 Idrissa DIENG <idy...@gm... <mailto:idy...@gm...>>
>
>     Hi John,
>
>     I saw that the JGraphT 0.9.0 has already released in the GIT
>     development stream.
>
>     Where I can found please the official download artifacts?
>
>     Le 13/11/2013 10:12, Idrissa DIENG a écrit :
>>     Hi John,
>>
>>     Thanks for your quite response. This is very good new.
>>
>>     Best regards,
>>     Idrissa DIENG
>>
>>     Le 13/11/2013 00:22, John Sichi a écrit :
>>>     On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG
>>>     <idy...@gm... <mailto:idy...@gm...>> wrote:
>>>
>>>         Can you please give us the latest status of relicensing process?
>>>         Exist there an estimation for the next release that should
>>>         provide
>>>         JGraphT under EPL license? What is the estimation effort?
>>>
>>>
>>>     Hi Idrissa,
>>>
>>>     We got sufficient contributor approval, so the only thing
>>>     remaining is to change all the file headers.  There are a few
>>>     other items which need work before we can make the 0.9 release
>>>     including the license change; we're hoping to have that out
>>>     before year's end.
>>>
>>>     JVS
>>>      
>>
>
>
>     ------------------------------------------------------------------------------
>     Rapidly troubleshoot problems before they affect your business.
>     Most IT
>     organizations don't have a clear picture of how application
>     performance
>     affects their revenue. With AppDynamics, you get 100% visibility
>     into your
>     Java,.NET, & PHP application. Start your 15-day FREE TRIAL of
>     AppDynamics Pro!
>     http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
>     _______________________________________________
>     jgrapht-users mailing list
>     jgr...@li...
>     <mailto:jgr...@li...>
>     https://lists.sourceforge.net/lists/listinfo/jgrapht-users
>
>

Re: [jgrapht-users] JGraphT Relicensing Process

[Szabolcs B. <bes...@gm...>]

Hi!

http://oss.sonatype.org/content/repositories/snapshots/org/jgrapht/

Szabolcs Besenyei


2013/12/11 Idrissa DIENG <idy...@gm...>

>  Hi John,
>
> I saw that the JGraphT 0.9.0 has already released in the GIT development
> stream.
>
> Where I can found please the official download artifacts?
>
> Le 13/11/2013 10:12, Idrissa DIENG a écrit :
>
> Hi John,
>
> Thanks for your quite response. This is very good new.
>
> Best regards,
> Idrissa DIENG
>
> Le 13/11/2013 00:22, John Sichi a écrit :
>
> On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG <idy...@gm...> wrote:
>
>> Can you please give us the latest status of relicensing process?
>> Exist there an estimation for the next release that should provide
>> JGraphT under EPL license? What is the estimation effort?
>>
>
>  Hi Idrissa,
>
>  We got sufficient contributor approval, so the only thing remaining is
> to change all the file headers.  There are a few other items which need
> work before we can make the 0.9 release including the license change; we're
> hoping to have that out before year's end.
>
>  JVS
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Rapidly troubleshoot problems before they affect your business. Most IT
> organizations don't have a clear picture of how application performance
> affects their revenue. With AppDynamics, you get 100% visibility into your
> Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics
> Pro!
> http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
> _______________________________________________
> jgrapht-users mailing list
> jgr...@li...
> https://lists.sourceforge.net/lists/listinfo/jgrapht-users
>
>

Re: [jgrapht-users] JGraphT Relicensing Process

[Idrissa D. <idy...@gm...>]

Hi John,

I saw that the JGraphT 0.9.0 has already released in the GIT development
stream.

Where I can found please the official download artifacts?

Le 13/11/2013 10:12, Idrissa DIENG a écrit :
> Hi John,
>
> Thanks for your quite response. This is very good new.
>
> Best regards,
> Idrissa DIENG
>
> Le 13/11/2013 00:22, John Sichi a écrit :
>> On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG <idy...@gm...
>> <mailto:idy...@gm...>> wrote:
>>
>>     Can you please give us the latest status of relicensing process?
>>     Exist there an estimation for the next release that should provide
>>     JGraphT under EPL license? What is the estimation effort?
>>
>>
>> Hi Idrissa,
>>
>> We got sufficient contributor approval, so the only thing remaining
>> is to change all the file headers.  There are a few other items which
>> need work before we can make the 0.9 release including the license
>> change; we're hoping to have that out before year's end.
>>
>> JVS
>>  
>

[jgrapht-users] JGraphT in a public Maven repository

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

Hey all,

I'm excited to report that at long last we're close to having the
latest JGraphT artifacts published in Maven Central.  You can help by
testing out what we've got already to make sure that our
build/publication process is correct.  (Note that the old 0.7.3 and
earlier artifacts in Maven Central were uploaded by random users, not
as part of any release process.)

As the first step, we've published a recent snapshot build to the
sonatype repository.  In your build, use this repository URL:

http://oss.sonatype.org/content/repositories/snapshots

And then add this dependency:

<dependency>
  <groupId>org.jgrapht</groupId>
  <artifactId>jgrapht-core</artifactId>
  <version>0.8.3-SNAPSHOT</version>
</dependency>

Note that the version is still incorrect (0.8.3 was the last released
version, and the next released version will be 0.9.0).  It should be
0.9.0-SNAPSHOT.  We'll have this fixed soon.

Please report any problems you encounter as issues in github so that
we can make sure that they're addressed prior to release.  Once we
make the release, we'll publish the final artifacts to Maven Central.

Thanks,
JVS

Re: [jgrapht-users] BlockCutpointGraph - getBlock

[H.N. de R. <hnr...@gr...>]

On Fri, Nov 29, 2013 at 04:06:48PM +0100, Carl Witt wrote:

Hello Carl,

> I have a question concerning the public UndirectedGraph 
> <http://jgrapht.org/javadoc/org/jgrapht/UndirectedGraph.html><V 
> <http://jgrapht.org/javadoc/org/jgrapht/alg/BlockCutpointGraph.html>,E 
> <http://jgrapht.org/javadoc/org/jgrapht/alg/BlockCutpointGraph.html>> 
> *getBlock*(V 
> <http://jgrapht.org/javadoc/org/jgrapht/alg/BlockCutpointGraph.html> vertex) 
> method.
> The documentation says
> 
>      Returns the vertex if vertex is a cutpoint, and otherwise returns 
> the block (biconnected component) containing the vertex.
> 
> But the example given in the introductory section of the 
> BlockCutpointGraph class clarifies that the a vertex can be part of more 
> than one block
>
> Is the documentation of the getBlock method being imprecise (should it 
> read "returns a block containing the vertex") or is the underlying 
> definition of a block not the one from Mathworld?

I don't know how it could happen, but cutpoint is one of the very few
concepts in graph theory that have only one meaning. A cutpoint is indeed in
more than one block (x is a cutpoint iff x is in more than one block), but
non-cutpoints are in precisely one block.

So getBlock(x) returns x if x is a cutpoint and the unique block that
contains x, otherwise.

Hope this clears things up,
   Ernst

-- 
Information System on Graph Classes and their Inclusions (ISGCI)
http://www.graphclasses.org

[jgrapht-users] BlockCutpointGraph - getBlock

[Carl W. <wit...@st...>]

Hello JGraphT Users,

I have a question concerning the public UndirectedGraph 
<http://jgrapht.org/javadoc/org/jgrapht/UndirectedGraph.html><V 
<http://jgrapht.org/javadoc/org/jgrapht/alg/BlockCutpointGraph.html>,E 
<http://jgrapht.org/javadoc/org/jgrapht/alg/BlockCutpointGraph.html>> 
*getBlock*(V 
<http://jgrapht.org/javadoc/org/jgrapht/alg/BlockCutpointGraph.html> vertex) 
method.
The documentation says

     Returns the vertex if vertex is a cutpoint, and otherwise returns 
the block (biconnected component) containing the vertex.

But the example given in the introductory section of the 
BlockCutpointGraph class clarifies that the a vertex can be part of more 
than one block

     http://mathworld.wolfram.com/Block.html

Is the documentation of the getBlock method being imprecise (should it 
read "returns a block containing the vertex") or is the underlying 
definition of a block not the one from Mathworld?

Best regards,
Carl Witt

Re: [jgrapht-users] Class Cast Exception's within JGrapht

[H.N. de R. <hnr...@gr...>]

On Sat, Nov 23, 2013 at 09:29:23PM -0700, Deepa Mallela wrote:

Hello Deepa,

can cou give a (minimal) code example in which you get the exception? Without
code, it's difficult to say what's going wrong.

Regards,
   Ernst


> I see a lot of java.lang.classcastException 's in the JGrapht Libray.
> 
> For example, when I tried to test  the classes in org.jgrapht.alg package
> such as  DijkstraShortestPath I am getting the class cast Exception's
> from org.jgrapht.graph, org.jgrapht.traverse, and org.jgrapht.util.
> 
>   And BellmanFordShortestPath algorithm, I see the same kind of exception's
> in AbstractPathElement, BellmanFordIterator classes from the
> org.jgrapht.alg package.


-- 
Information System on Graph Classes and their Inclusions (ISGCI)
http://www.graphclasses.org

[jgrapht-users] Class Cast Exception's within JGrapht

[Deepa M. <deepamallela@u.boisestate.edu>]

Dear Member's community,

I see a lot of java.lang.classcastException 's in the JGrapht Libray.

For example, when I tried to test  the classes in org.jgrapht.alg package
such as  DijkstraShortestPath I am getting the class cast Exception's
from org.jgrapht.graph, org.jgrapht.traverse, and org.jgrapht.util.

  And BellmanFordShortestPath algorithm, I see the same kind of exception's
in AbstractPathElement, BellmanFordIterator classes from the
org.jgrapht.alg package.

Is there any solid reason behind this.Can you please enlighten me upon this.

Your help is appreciated

-- 
Thanks &Regards,
Deepa Mallela.

Re: [jgrapht-users] JGraphT Relicensing Process

[Idrissa D. <idy...@gm...>]

Hi John,

Thanks for your quite response. This is very good new.

Best regards,
Idrissa DIENG

Le 13/11/2013 00:22, John Sichi a écrit :
> On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG <idy...@gm...
> <mailto:idy...@gm...>> wrote:
>
>     Can you please give us the latest status of relicensing process?
>     Exist there an estimation for the next release that should provide
>     JGraphT under EPL license? What is the estimation effort?
>
>
> Hi Idrissa,
>
> We got sufficient contributor approval, so the only thing remaining is
> to change all the file headers.  There are a few other items which
> need work before we can make the 0.9 release including the license
> change; we're hoping to have that out before year's end.
>
> JVS
>  

Re: [jgrapht-users] JGraphT Relicensing Process

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

On Tue, Nov 12, 2013 at 8:23 AM, Idrissa DIENG <idy...@gm...> wrote:

> Can you please give us the latest status of relicensing process?
> Exist there an estimation for the next release that should provide
> JGraphT under EPL license? What is the estimation effort?
>

Hi Idrissa,

We got sufficient contributor approval, so the only thing remaining is to
change all the file headers.  There are a few other items which need work
before we can make the 0.9 release including the license change; we're
hoping to have that out before year's end.

JVS

[jgrapht-users] JGraphT Relicensing Process

[Idrissa D. <idy...@gm...>]

Dear all,

We currently use the JGrapth library internally and we plan to use it in
an EclipseLabs open source project.

I found in the Wiki page
(https://github.com/jgrapht/jgrapht/wiki/Relicensing) that a relicensing
initiative is started. Therefore, we are very interested in the JGraphT
relicensing process.

Can you please give us the latest status of relicensing process?
Exist there an estimation for the next release that should provide
JGraphT under EPL license? What is the estimation effort?

Thanks in advance for your response.

[jgrapht-users] JGraphT Relicensing Process

[Idrissa D. <idy...@gm...>]

Dear all,

We currently use the JGrapth library internally and we plan to use it in
an EclipseLabs open source project.

I found in the Wiki page
(https://github.com/jgrapht/jgrapht/wiki/Relicensing) that a relicensing
initiative is started. Therefore, we are very interested in the JGraphT
relicensing process.

Can you please give us the latest status of relicensing process?
Exist there an estimation for the next release that should provide
JGraphT under EPL license? What is the estimation effort?

Thanks in advance for your response.

Re: [jgrapht-users] loops in simpleweighted graph

[H.N. de R. <hnr...@gr...>]

On Sun, Nov 10, 2013 at 10:43:10AM -0800, Hamideh Bakhtiari wrote:

> I want to have loops in my simpleweightedgraph structure. any idea how to do that?

You can't. A simple graph by definition has no loops. You'll need to use
WeightedPseudoGraph which allows both loops and multi-edges.

Ernst

-- 
Information System on Graph Classes and their Inclusions (ISGCI)
http://www.graphclasses.org

[jgrapht-users] loops in simpleweighted graph

[Hamideh B. <ham...@ya...>]

Hi All,

I want to have loops in my simpleweightedgraph structure. any idea how to do that?

Thanks.

[jgrapht-users] performance techniques

[Luc H. <Luc...@in...>]

Good morning,

I'm interested in performance techniques for graph algorithms. In 
particular I would like to know what have been made into JGraphT to make 
it efficient.

Regards,
Luc.



-- 
Luc Hogie
COMRED Research Unit (I3S(CNRS-UNS) INRIA)
I3S Laboratory, University of Nice Sophia Antipolis

http://www-sop.inria.fr/members/Luc.Hogie/
luc...@cn...
+33 4 89 73 24 25 (office)
+33 6 80 91 40 71 (mobile)
Skype ID: luchogie