Re: [igraph] R igraph: get.shortest.paths returns error

[Tobias Heckmann]

hm, I've tried your code snippet with both machines.

First my "old" one:

>sessionInfo()

R version 2.15.1 (2012-06-22)
Platform: i386-pc-mingw32/i386 (32-bit)

locale:
[1] LC_COLLATE=English_United Kingdom.1252 LC_CTYPE=English_United 
Kingdom.1252    LC_MONETARY=English_United Kingdom.1252 LC_NUMERIC=C 
LC_TIME=English_United Kingdom.1252

attached base packages:
[1] tcltk     stats     graphics  grDevices utils     datasets methods   
base

other attached packages:
 [1] RSAGA_0.93-1          plyr_1.7.1 shapefiles_0.6        
foreign_0.8-50        gstat_1.0-12 spacetime_0.7-0       
xts_0.8-6             zoo_1.7-7 sp_0.9-99             sqldf_0.4-6.4
[11] RSQLite.extfuns_0.0.1 RSQLite_0.11.1 chron_2.3-42          
gsubfn_0.6-3          proto_0.3-9.2 DBI_0.2-5             
Matrix_1.0-6          lattice_0.20-6 igraph_0.6-2

loaded via a namespace (and not attached):
[1] grid_2.15.1

The snippet returns >50 warnings with respect to "couldn't reach some 
vertices", but no error

---------------------------------------------------------------------------------------------------

Now my "new" computer:

>sessionInfo()

R version 2.15.1 (2012-06-22)
Platform: x86_64-pc-mingw32/x64 (64-bit)

locale:
[1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252    
LC_MONETARY=German_Germany.1252 LC_NUMERIC=C                    
LC_TIME=German_Germany.1252

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods base

other attached packages:
[1] igraph_0.6-2

loaded via a namespace (and not attached):
[1] tools_2.15.1

here, the snippet returns again the error
"At structural_properties.c:659 : Size of the `vertices' and the `to' 
should match, Invalid value"

-------------
I'd like to point out that the only differences I can see are the 
Patform and the Locale. Both machines use the same R (2.15.1) and igraph 
(0.6-2) versions.
So I can't confirm that it is a matter of the R version...

Thank you anyway for your help !
Cheers
Tobias


Am 08/08/2012 16:34, schrieb Gábor Csárdi:
> Thanks for the report! I cannot reproduce this on OSX and R 2.15.1, using
>
> for (i in 1:10000) {
>    g <- grg.game(100, 0.2)
>    get.shortest.paths(g, from=1, to=c(2,3,4,5), output="epath")
> }
>
> So I am not sure what is happening here. I am running OS X 10.6.8,
> btw, and igraph 0.6-2, of course.
>
> G.
>
> On Wed, Aug 8, 2012 at 5:19 AM, Tamas Nepusz <address@hidden> wrote:
> > Dear Tobias,
> >
> > Thanks for the feedback. It seems that your problem is version-specific,
> > i.e. it occurs with igraph in R 2.15.1 (confirmed on Mac OS X) but not in R
> > 2.14.1 (confirmed on Linux). I have created a bug report for this and
> > assigned it to Gabor, the developer of the R interface. You can subscribe to
> > the bug report on the following page if you would like to be notified when
> > this is fixed:
> >
> > https://bugs.launchpad.net/igraph/+bug/1034312
> >
> > Best,
> > Tamas
> >
> >
> > On 08/08/2012 10:56 AM, Tobias Heckmann wrote:
> > > Dear Tamas,
> > >
> > > my RData file is unfortunately too large to send (43 MB).
> > > I have now tried a random graph, and get.shortest.paths returns the same
> > > error... Hence, it doesn't seem to depend on my data. I've attached the
> > > RData file for my graph and my sessionInfo.
> > >
> > > Thank you very much for dealing with my problem !
> > > Best regards
> > > Tobias
> > >
> > >
> > > Am 08/08/2012 09:48, schrieb Tamás Nepusz:
> > > > Dear Tobias,
> > > >
> > > > This is strange; I still cannot reproduce your issue on my machine, neither with "ordinary" R 
> > > > vectors nor with vertex sequences. Could you send me an Rdata file with an example graph and a 
> > > > "startnode" and "tonodes" vector that causes problems on your machine?
> > > >
> > > > Best,--
> > > > T.
> > > >
> > > >
> > > > On Wednesday, 8 August 2012 at 09:25, Tobias Heckmann wrote:
> > > >
> > > > > Dear Tamas,
> > > > >
> > > > > in the original version of my script (which worked on my old machine),
> > > > > startnode is a simple R vector:
> > > > > > str(startnode)
> > > > > num 177822
> > > > >
> > > > > while tonodes seems to be a vector sequence:
> > > > > > str(tonodes)
> > > > > Class 'igraph.vs' atomic [1:286] 157812 158206 158605 158989 159367 ...
> > > > > ..- attr(*, "env")=<environment: 0x1013523c>
> > > > >
> > > > > in ?get.shortest.paths, it reads that the "to" argument should be a
> > > > > "numeric vector, the vertices to which the shortest paths will be
> > > > > calculated..."
> > > > > Hence, I also tried as.numeric(tonodes) in the "to" argument, but the
> > > > > error remains the same.
> > > > >
> > > > > Cheers,
> > > > > Tobias
> > > > >
> > > > > Am 08/08/2012 08:43, schrieb Tamas Nepusz:
> > > > > > Dear Tobias,
> > > > > >
> > > > > > > Until today, the following command worked well:
> > > > > > >
> > > > > > > paths<-get.shortest.paths(graph,from=startnode,to=tonodes,mode="out",weights=NA,output="epath")
> > > > > > >
> > > > > > > Today, I installed R on a 64bit machine and tried running my script again.
> > > > > > > This time, the command returned the following error:
> > > > > > >
> > > > > > > "At structural_properties.c:659 : Size of the `vertices' and the `to'
> > > > > > > should match, Invalid value"
> > > > > >
> > > > > > I have tried this now on a random graph with random start and end nodes and
> > > > > > it seemed to work for me, so can you please let us know what is there in the
> > > > > > "startnode" and "tonodes" variables? Are they simple R vectors?
> > > > > >
> > > > > > Cheers,
> > > > > > Tamas
> > > > > >
> > > > > > _______________________________________________
> > > > > > igraph-help mailing list
> > > > > > address@hidden (mailto:address@hidden)
> > > > > > https://lists.nongnu.org/mailman/listinfo/igraph-help
> > > > >
> > > > >
> > > > >
> > > > > --
> > > > > ---
> > > > > Dr. Tobias Heckmann
> > > > > Wiss. Angestellter
> > > > > Lehrstuhl für Physische Geographie
> > > > > Katholische Universität Eichstätt-Ingolstadt
> > > > > Ostenstraße 18
> > > > > 85072 Eichstätt
> > > > >
> > > > > Tel.: 08421-93-1309
> > > > > Fax: 08421-93-1787
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > igraph-help mailing list
> > > > > address@hidden (mailto:address@hidden)
> > > > > https://lists.nongnu.org/mailman/listinfo/igraph-help
> > > >
> > > >
> > > > _______________________________________________
> > > > igraph-help mailing list
> > > > address@hidden
> > > > https://lists.nongnu.org/mailman/listinfo/igraph-help
> > >
> > >
> > >
> > > _______________________________________________
> > > igraph-help mailing list
> > > address@hidden
> > > https://lists.nongnu.org/mailman/listinfo/igraph-help
> > _______________________________________________
> > igraph-help mailing list
> > address@hidden
> > https://lists.nongnu.org/mailman/listinfo/igraph-help


--
---
Dr. Tobias Heckmann
Wiss. Angestellter
Lehrstuhl für Physische Geographie
Katholische Universität Eichstätt-Ingolstadt
Ostenstraße 18
85072 Eichstätt

Tel.: 08421-93-1309
Fax: 08421-93-1787