Re: [igraph] Odd error message

[Peter Flom]

OK a bit more information and additional problems

 

The links in the file were 13 digit numbers (they are ISBNs).  Excel (which is how the data were stored) converts these to scientific notation. Even if I convert it to a number, when I store it as a .csv file, the scientific notation returns.

 

For ONE file, I got the weird error message below.  BUT EVEN WORSE, for other files, I got no errors, just wrong results (with wrong links and stuff).  The links seem to be coming from the oddly notated ISBNs.

 

I realize this is more of an R problem combined with Excel, and I will look on R resources, but it’s a real danger here, and I wanted to let people know. 

 

Peter Flom

Peter Flom Consulting

http://www.statisticalanalysisconsulting.com/

http://www.IAmLearningDisabled.com

 

From: Peter Flom [mailto:address@hidden
Sent: Saturday, April 21, 2012 12:11 PM
To: 'address@hidden'
Subject: Odd error message

 

Hello again

 

When I run the following code

<<<< 

DVg <- graph.empty()

DVg <- add.vertices(DVg, nrow(DVtraits),

                     ISBN=as.character(DVtraits[,1]),

                     title=as.character(DVtraits[,2]), author = as.character(DVtraits[,3]),

                                                             publisher = as.character(DVtraits[,4]), salesrank = as.character(DVtraits[,6]),

                                                            category = as.character(DVtraits[,5]))

                                                           

names <- V(DVg)$ISBN

ids <- 1:length(names)-1

names(ids) <- names

 

from <- as.character(DVlinks[,1])

to <- as.character(DVlinks[,2])

 

edges <- matrix(c(ids[from], ids[to]), nc=2)

DVg <- add.edges(DVg, t(edges))                  

>>>> 

 

I get (after running the last line) an error:

 

 

Traceback:

1: .Call("R_igraph_add_edges", graph, as.numeric(edges), PACKAGE = "igraph")

2: add.edges(DVg, t(edges))

Error in add.edges(DVg, t(edges)) :

  caught access violation - continue with care

 

Last time this happened, it turned out to be that there were some NA in the from and to, but there are not this time.

 

I am not sure how to create a toy example…. Or what else would be helpful here.

 

Thanks (and if someone wants me to send the data, I can send it)

 

Peter

 

Peter Flom

Peter Flom Consulting

http://www.statisticalanalysisconsulting.com/

http://www.IAmLearningDisabled.com