Mark Engelberg
Mark Engelberg
I think you've correctly identified that Ubergraph would benefit from graph attributes, and I'm glad you were able to find a reasonable workaround for your specific use case. I would...
This all looks great. Thanks for the contribution. I don't want to push a release while I'm traveling and won't be readily available to support any issues caused by a...
I think you're right that the call to edge-description->edge is unnecessary. I was probably thinking that since the call to undirected-edge? is a protocol that can theoretically be extended to...
Thanks for this. Although resolve-node-or-edge is still "constant time", as I began to merge it, I found myself uncomfortable with adding the extra overhead of verifying the edge's existence every...
Thanks. Since this would change the function's return type from a set to a seq, I need to do some double-checking to be sure that's not a problem.
For background, I think the reason I didn't bother with attaching the metadata here is that originally, that was sort of a hack so that I when passed graphs to...
Looking at it, I guess it doesn't currently promise a set return value in all possible code branches.
After doing some soul-searching on this issue over the past couple of weeks, I've come to the conclusion that it's a bad idea for end users to expect the edges...
I have completed the changes to code and README to prevent edge objects from being used in graph constructors, and added node-with-attrs and edge-with-attrs. Take a look and let me...
I just ran into this issue too.