PyDiGraph: get ingoing edge indices of a node

[hakkape]

What is the expected enhancement?

Similar to incident_edges() which can return the IDs of edges going out OR going in and out of a node in a directed graph, i would like to have a function to obtain the IDs of just the ingoing edges of a node. There is in_edges(), but this only returns tuples of (head_node, tail_node, data), so i would need to call another function to get the ID of each edge.

[IvanIsCoding]

I assume we could use a in_edge_indices function. That seems specific but reasonable

[ThisuraGallage]

Hi... looks like this issue is free for grabs. May I take this up?

Edit: Just a bit confused about the proposed incident_edge_indices function. As far as I understand, the incident_edges() function already gives us indices for edges going out or both in and out (all incident edges). What seems to be missing is a function that would return only the indices of ingoing edges. So, I was wondering, what would the incident_edge_indices function return? Could you help me clarify this? Apologies if I’ve misunderstood anything!

[IvanIsCoding]

I will post a comment with the details later but yes this issue is available as a good first issue

[ThisuraGallage]

Great, I'll take on the issue!

[ThisuraGallage]

Hi,
I went through the code and read the contribution guidelines in CONTRIBUTING.md. I believe these are the places I need to make the changes, right? Apologies if I’m misunderstanding anything—kinda new to this!

Add the new function in, https://github.com/Qiskit/rustworkx/blob/main/src/digraph.rs
Add tests in, https://github.com/Qiskit/rustworkx/blob/main/tests/digraph/test_edges.py
Function signature in, https://github.com/Qiskit/rustworkx/blob/main/rustworkx/rustworkx.pyi
and a release note

[IvanIsCoding]

Yes, the places you outlined are correct! Sorry I haven’t had time to update the details yet but what inferred from Contributing.md should be enough to get you started.

I will try to add the details when I can and/or review your PR if you finish before I update

[ThisuraGallage]

No worries! I've created a draft PR, Please review it at your convenience and let me know your thoughts. I'm not entirely sure about the function signature as well. I'll add documentation once everything is finalized, thanks !

[IvanIsCoding]

No worries! I've created a draft PR, Please review it at your convenience and let me know your thoughts. I'm not entirely sure about the function signature as well. I'll add documentation once everything is finalized, thanks !

After revising all of our APIs, the method should be in_edge_indices and it should just be the logic of in_edges but returning only the EdgeIndices.

An optional improvement is adding out_edge_indices which is the same as out_edges but again for the other direction

[ThisuraGallage]

Hi, I added the requested changes in #1369 . I kept the tests for the undirected graph consistent with the other tests in the file. Let me know if those need the same improvement u suggested for tests in directed graph. ( I assumed the variables for edge indices were used to make the edge and directions clear in the test functions ). Thanks !