Re: [igraph] Intersect graphs with different vertices

[Peter Pirolli]

Thanks for the response. I did make some minor modifications and the code 
worked just fine for directed graphs. I'll take a look on Monday and see what 
is worth sending to you (I'm out of the country until then).

And, thanks again for providing the code. If anything comes out of our work I 
will acknowledge your help. We're trying to visualize the evolution of 
Wikipedia, including the social network evolution.

--Pete



On Sep 3, 2010, at 11:51 AM, Magnus Thor Torfason wrote:

> Hi Pete,
> 
> I'm glad that the code worked out for you, and sorry that I didn't respond 
> earlier.
> 
> Anyway, to answer your question about adapting the code for directed graphs: 
> No, there is no reason why this should not work with directed graphs. I 
> simply had not tested them with directed graphs, and the check was only a 
> matter of defensive programming. When the choice is between an error and a 
> potentially unexpected result I tend to opt for an error (especially when the 
> code can then be updated and tested with a given special case in mind).
> 
> I have actually now made these and a few other igraph utility functions 
> available on my web page:
> 
> http://www.torfason.net/source-code/
> 
> If you are using the functions more or less as-is, it would probably be most 
> useful to download them from there, since any updates or bug fixes would be 
> reflected there.
> 
> And on that note, you mentioned that you added some attribute copying code. 
> I'd very much like to incorporate any improvements into my code, if you're 
> willing to share.
> 
> Best,
> Magnus Thor
> 
> On 8/26/2010 1:39 PM, Peter Pirolli wrote:
>> Hi:
>> 
>> I came across your code in
>> 
>> http://lists.nongnu.org/archive/html/igraph-help/2009-10/msg00087.html
>> 
>> which looks to be exactly what I need (thank you)--except that I am working 
>> with directed graphs. When I modified the relevant lines to permit directed 
>> graphs and do the appropriate attribute copying, everything seems to work 
>> fine for the simple examples that I tried. Can you tell me why you 
>> originally restricted the functions to undirected graphs?
>> 
>> --Pete
> 
>