feat: make Csr::from_sorted_edges generic over edge type#783
feat: make Csr::from_sorted_edges generic over edge type#783uulm-janbaudisch wants to merge 1 commit intopetgraph:masterfrom
Csr::from_sorted_edges generic over edge type#783Conversation
uulm-janbaudisch
changed the title
Csr::from_sorted_edges generic over edge typeCsr::from_sorted_edges generic over edge type
|
I think this was done some times ago due to the requirements of unambiguity of the edge order. |
|
What I noticed when using CSRs for undirected graphs is that edges have to be present in both directions for |
starovoid
force-pushed
the
csr-from-edges-undirected
branch
from
4b41b80 to
a2b5892
Compare
starovoid
added
the
C-feature-accepted
starovoid
force-pushed
the
csr-from-edges-undirected
branch
from
a2b5892 to
5baf298
Compare
Let's do that. |
starovoid
force-pushed
the
csr-from-edges-undirected
branch
from
5baf298 to
7d21f27
Compare
starovoid
force-pushed
the
csr-from-edges-undirected
branch
from
7d21f27 to
2726f3c
Compare
|
Since this PR seems stale, I was wondering if it would be alright if I would pick it up. That is, fork your branch and finish it such that your work gets merged : ) |
uulm-janbaudisch
force-pushed
the
csr-from-edges-undirected
branch
from
bb207b2 to
26f106a
Compare
|
Sorry for the delay. I added a note to the documentation. |
uulm-janbaudisch
force-pushed
the
csr-from-edges-undirected
branch
from
26f106a to
b6f7190
Compare
No worries at all :) Alright, I see 🤔 I'll look into it then and prob take over the PR 👍 I appreciate the quick response |
|
Okay, I will close this in favor of #861 then : ) |
… increase edge_count in Csr::from_sorted_edges (#861) This PR continues the work done in #783. Thus, this resolves #588. The trait bound on the `Csr` to be `Directed` for `from_sorted_edges` was lifted. Thus, it was noticed that at no point in `from_sorted_edges`, the `edge_count` is increased. This is only relevant for `Undirected` graphs, as `Directed` graphs use their `self.column.len()` as the `edge_count()` under the hood. --------- Co-authored-by: Jan Baudisch <[email protected]>
… increase edge_count in Csr::from_sorted_edges (petgraph#861) This PR continues the work done in petgraph#783. Thus, this resolves petgraph#588. The trait bound on the `Csr` to be `Directed` for `from_sorted_edges` was lifted. Thus, it was noticed that at no point in `from_sorted_edges`, the `edge_count` is increased. This is only relevant for `Undirected` graphs, as `Directed` graphs use their `self.column.len()` as the `edge_count()` under the hood. --------- Co-authored-by: Jan Baudisch <[email protected]>
… increase edge_count in Csr::from_sorted_edges (petgraph#861) This PR continues the work done in petgraph#783. Thus, this resolves petgraph#588. The trait bound on the `Csr` to be `Directed` for `from_sorted_edges` was lifted. Thus, it was noticed that at no point in `from_sorted_edges`, the `edge_count` is increased. This is only relevant for `Undirected` graphs, as `Directed` graphs use their `self.column.len()` as the `edge_count()` under the hood. --------- Co-authored-by: Jan Baudisch <[email protected]>
3 participants
I am not sure if there was a reason for the constraint, but it seems to work fine for undirected graphs.
Fixes #588.