Re: [jgrapht-users] Documentation, implementation or me?

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

Mathematically, a graph is a pair of sets (vertices and edges), so we
model it on java.util.Set, which has the same behavior (regardless of
hash vs tree, you can't change keys after adding them).  If you want
mutability, you should define a containing class which uses the
default object identity hash code and supply that as your vertex
class; hang the set as a data member.

We have an existing wiki page on this topic; I'll add a link to it
from the Javadoc:

http://pub.eigenbase.org/wiki/JGraphT:EqualsAndHashCode

JVS

On Mon, Nov 15, 2010 at 6:37 PM, Jonathan Cederberg
<jon...@it...> wrote:
> Exactly what I thought. But... is this really what a graph should be? I mean, there is no hint in the docs as far as I can see, and although I know it is not good practice to change things in a collection from the outside, it still seems like something you might want to do. In particular if the graph represents some meta level info...
>
> /J
>
> On 16 nov 2010, at 03.18, John Sichi wrote:
>
>> Hi Jonathan,
>>
>> s2 is a mutable object, and you changed it after adding it to the
>> graph.  That's why you got unexpected results.
>>
>> See the section "Good keys" in this article:
>>
>> http://www.ibm.com/developerworks/java/library/j-jtp02183.html
>>
>> JVS
>>
>> On Mon, Nov 15, 2010 at 5:23 PM, Jonathan Cederberg
>> <jon...@it...> wrote:
>>> I am a huge fan of jgrapht. However, I have come across a problem with it.
>>> My understanding is that that the code below should work, but it doesn't. I
>>> have not dug into the source, but I can imagine that there is a hashcode
>>> issue, as it no longer seem to find the vertex that is there.
>>> I am not sure if I have read the docs wrong, programmed wrong, or if this is
>>> simply a bug. The latter seems unlikely though, since it is such a core
>>> feature that this type of problem should have been caught long ago.
>>> Thanks in advance for any help!
>>> /Jonathan
>>>
>>> import java.util.LinkedHashSet;
>>> import java.util.Set;
>>> import org.jgrapht.graph.DefaultEdge;
>>> import org.jgrapht.graph.SimpleGraph;
>>> public class Main {
>>> /**
>>> * @param args
>>> */
>>> public static void main(String[] args) {
>>> SimpleGraph<Set<String>, DefaultEdge> sg =
>>> new SimpleGraph<Set<String>, DefaultEdge>(DefaultEdge.class);
>>>
>>> Set<String> s1, s2;
>>> s2 = new LinkedHashSet<String>();
>>> s2.add("y");
>>> s2.add("z");
>>>
>>> sg.addVertex(s2);
>>>
>>> s2.remove("z");
>>>
>>> // this is true, as expected
>>> assert(sg.vertexSet().size()==1) : "There is no vertex any more.";
>>> assert(sg.vertexSet().iterator().next()==s2) : "Not the same vertex!";
>>>
>>> // but this is false!
>>> assert(sg.containsVertex(s2)) : "The vertex is gone!";
>>> }
>>> }
>>>
>>> ------------------------------------------------------------------------------
>>> Beautiful is writing same markup. Internet Explorer 9 supports
>>> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
>>> Spend less time writing and  rewriting code and more time creating great
>>> experiences on the web. Be a part of the beta today
>>> http://p.sf.net/sfu/msIE9-sfdev2dev
>>> _______________________________________________
>>> jgrapht-users mailing list
>>> jgr...@li...
>>> https://lists.sourceforge.net/lists/listinfo/jgrapht-users
>>>
>>>
>
>
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1,  ECMAScript5, and DOM L2 & L3.
> Spend less time writing and  rewriting code and more time creating great
> experiences on the web. Be a part of the beta today
> http://p.sf.net/sfu/msIE9-sfdev2dev
> _______________________________________________
> jgrapht-users mailing list
> jgr...@li...
> https://lists.sourceforge.net/lists/listinfo/jgrapht-users
>