On Thursday, June 14, 2012 3:27:41 PM UTC+1, jason wrote:
>
> So, to be clear, the new networkx function does not give you as much
> information as the old one?
Yes, that is the case. The (weighted) clustering coefficient algorithm
(with weighted *edges*)
computes some auxiliary *vertex* weights. The old method used to return all
of it, the new one
just returns the clustering coefficients
> Generally, in the Sage wrapper for the
> networkx function, you'd "correct" the return output to take care of the
> backwards incompatibility. Is there a way to get the missing
> information easily?
>
Well, kind of easily. I can just copy over the lines from the old networkx
and since it is pure python it will probably work. Should I do that with an
additional "backwards_compatible" boolean variable set on True by default?
def sage_function(..., backwards_compatible = True)
if not backward_compatible:
results = networkx.function(...)
else:
deprecation_warning(...)
# Copy about 20 lines of old networkx code here
results = ... (obtained from the old code)
return results
Adding the missing information alone takes about the same work as computing
the whole thing...
J.
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
