igraph-help
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [igraph] RuntimeWarning when reading GraphML


From: Chris Wj
Subject: Re: [igraph] RuntimeWarning when reading GraphML
Date: Fri, 6 Feb 2009 15:14:29 -0500


> OK, but what is the problem with the output? There are no attributes, so no attributes are written.

Nothing is wrong with it. I was taken back by the fact that an ID attribute was written as a data element when I didn't specify/change them myself. It just seemed inconsistent and I was confused by that.

On Fri, Feb 6, 2009 at 2:53 PM, Gábor Csárdi <address@hidden> wrote:
On Fri, Feb 6, 2009 at 8:37 PM, Chris Wj <address@hidden> wrote:
>> I don't get that. Igraph outputs only "n0", "n1", etc. as <node
>> id="n0"> and outputs ALL vertex attributes (including "id") as <data
>> key= > etc.
>
> This is what I did in 0.6 to produce that output.
> G = igraph.Graph(5, edges=[(1,2),(2,3),(3,4)])
> G.write_graphml("no-attrs.graphml")

OK, but what is the problem with the output? There are no attributes,
so no attributes are written.

>> Someone else (or it was you? :) said last week that we should ignore
>> these completely. :)
> Oh really?! No that wasn't me. What was the reason?

It was not on the list, but in private, but there was no particular
reason, I believe. :)

G.

> On Fri, Feb 6, 2009 at 2:22 PM, Gábor Csárdi <address@hidden> wrote:
>>
>> On Fri, Feb 6, 2009 at 8:09 PM, Chris Wj <address@hidden> wrote:
>> > I've looked at a bunch of examples and the GraphML docs/tutorials. I
>> > have
>> > only seed the form <node id="n0"> to specify the id.
>> >
>> > Also, igraph outputs only:
>> > ...
>> > <node id="n0">
>> >     </node>
>> >     <node id="n1">
>> >     </node>
>> > ...
>> > when there are no other attributes specified. It comes out in the
>> > attributes
>> > list as soon as you add any attribute value.
>>
>> I don't get that. Igraph outputs only "n0", "n1", etc. as <node
>> id="n0"> and outputs ALL vertex attributes (including "id") as <data
>> key= > etc.
>>
>> > If it is used like this everywhere, I'd say the <node id="n0"> format
>> > should
>> > be the primary method of specifying node number.
>>
>> Someone else (or it was you? :) said last week that we should ignore
>> these completely. :)
>>
>> > My only other question is
>> > will a node ever have a value other than 'n#' as the id, and if so, will
>> > it
>> > not be read correctly by igraph?
>>
>> Igraph writes only "n0", "n1", etc., but can read whatever is in the file.
>>
>> At least this is the case for the current version, the next will
>> behave (probably) the way I described in my previous post.
>>
>> G.
>>
>> > -Chris
>> >
>> > On Fri, Feb 6, 2009 at 12:52 PM, Gábor Csárdi <address@hidden>
>> > wrote:
>> >>
>> >> So, as I see it, this would be almost exactly the current behavior,
>> >> except for renaming 'id' to "GraphML:id" when reading the graph. Oh,
>> >> and perhaps using "GraphML:id" as <node id="n3"> ids when writing out
>> >> the graph.
>> >>
>> >> G.
>> >>
>> >> On Fri, Feb 6, 2009 at 6:36 PM, Tamas Nepusz <address@hidden> wrote:
>> >> >> 1) Whether to read the <node id="n3"> type of ids from the graph, or
>> >> >> not. It they are not read, then we might lose some information. If
>> >> >> they are read, that could lead to confusion.
>> >> >
>> >> > I think we should read them anyway, but renaming the auto-generated
>> >> > "id"
>> >> > attribute to "GraphML:id" seems a good idea; chances are that some of
>> >> > our
>> >> > users will use the "id" attribute for other purposes, but it's
>> >> > unlikely
>> >> > with
>> >> > "GraphML:id".
>> >> >
>> >> >> 2) Whether to use the "id" (igraph) vertex attribute as <node
>> >> >> id="n3">
>> >> >> type of ids, when we write out a GraphML file. Perhaps not.
>> >> >
>> >> > I agree; in general, nothing guarantees that the "id" attributes are
>> >> > unique
>> >> > and although igraph will happily save the graph as GraphML in this
>> >> > case,
>> >> > it
>> >> > won't be able to read it back.
>> >> >
>> >> > --
>> >> > Tamas
>> >> >
>> >> >
>> >> >
>> >> > _______________________________________________
>> >> > igraph-help mailing list
>> >> > address@hidden
>> >> > http://lists.nongnu.org/mailman/listinfo/igraph-help
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >> Gabor Csardi <address@hidden>     UNIL DGM
>> >>
>> >>
>> >> _______________________________________________
>> >> igraph-help mailing list
>> >> address@hidden
>> >> http://lists.nongnu.org/mailman/listinfo/igraph-help
>> >
>> >
>> > _______________________________________________
>> > igraph-help mailing list
>> > address@hidden
>> > http://lists.nongnu.org/mailman/listinfo/igraph-help
>> >
>> >
>>
>>
>>
>> --
>> Gabor Csardi <address@hidden>     UNIL DGM
>>
>>
>> _______________________________________________
>> igraph-help mailing list
>> address@hidden
>> http://lists.nongnu.org/mailman/listinfo/igraph-help
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help
>
>



--
Gabor Csardi <address@hidden>     UNIL DGM


_______________________________________________
igraph-help mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/igraph-help


reply via email to

[Prev in Thread] Current Thread [Next in Thread]