can you post this link again? i cannot search it right now.

On Thursday, May 10, 2012 8:10:21 AM UTC-7, Jacopo Farina wrote:
>
> Have you already seen this old mail ? 
> http://lists.neo4j.org/pipermail/user/2011-March/007735.html
> It's about strongly connected components in graphs, and a SCC imply the 
> existence of at least one cycle. I used it with a 7M nodes graph and worked 
> pretty well, but that code is based on the embedded version of Neo4j.
> Cheers,
> Jacopo Farina
>
> 2012/5/10 Mattias Persson <[email protected] <javascript:>>
>
>>
>>
>> 2012/5/10 M.Alipour <[email protected] <javascript:>>
>>
>> Thanks,
>>> I want to get "all cycles" in my graph, then, by remove these cycles,
>>> convert my graph to a DAG.
>>> Graph size:290000 nodes.
>>>
>>> "all cycles" is not really something that computes to me. You have to 
>> have some kind of relationship types/directions to traverse your paths 
>> with. Graphs are connected and if you just look at nodes and relationships 
>> there are cycles absolutely everywhere. 
>>
>>>
>>> On May 10, 12:05 pm, Mattias Persson <[email protected]>
>>> wrote:
>>> > You can create a traversal with certain criterias and look at each 
>>> returned
>>> > path if it contains any node more than once. What are querying, paths
>>> > between start/end nodes or something else?
>>> >
>>> > 2012/5/10 M.Alipour <[email protected]>
>>> >
>>> > > Hi,
>>> > > Is there a way to "print" "paths that have cycles" in a Directed
>>> > > Cycling Graph that
>>> > > created by Neo4j graph database service?
>>> > > Thanks.
>>> >
>>> > --
>>> > Mattias Persson, [[email protected]]
>>> > Hacker, Neo Technologywww.neotechnology.com
>>>
>>
>>
>>
>> -- 
>> Mattias Persson, [[email protected] <javascript:>]
>> Hacker, Neo Technology
>> www.neotechnology.com
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Neo4j" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to