Paul Schoenfelder
Paul Schoenfelder
Shouldn't they handle this with a default clause? It gives them the same choice and semantics right? You can't discard messages from the log, so either you define that clause...
Well once the bad message is received it has effectively poisoned that node right? Regardless of how the error is handled, since it has been persisted to the log. So...
I've merged this manually, thanks for the contribution!
#43 has been merged, and will be released in 0.15.0, which at least provides a solution for situations in which you can provide a better vertex ID than phash2. I'm...
@anttipoi The reason to have a notion of a "vertex id" at all is to avoid having many copies of large vertex values in the graph datastructure. By using integer...
I considered it when building the library, but haven't had time to revisit it. I'm open to working with someone on getting a PR merged for adding such support though!
If you can open a PR with a test that reproduces the issue, I can take it from there - if it will be too tricky to get it repro'd...
Thanks for the PR, not sure why I failed to get this merged previously, seems to have fallen off my radar.
You're correct in that currently the order we visit edges/vertices is based on the order of the underlying map/mapset. If we follow any order at all, it should be insertion...
The pathfinding has not been implemented for undirected graphs yet, but if anyone is interested in lending a hand there, I'd love to get support fleshed out so that we...