perf(analyze): reuse degrees for surprise scoring (#914)

Co-authored-by: hanmo1 <hanmo1@lenovo.com>

graphify/analyze.py

@@ -182,6 +182,7 @@ def _surprise_score(
     node_community: dict[str, int],
     u_source: str,
     v_source: str,
+    degrees: dict[str, int] | None = None,
 ) -> tuple[int, list[str]]:
     """Score how surprising a cross-file edge is. Returns (score, reasons)."""
     score = 0
@@ -236,8 +237,8 @@ def _surprise_score(
         reasons.append("semantically similar concepts with no structural link")
 
     # 5. Peripheral→hub: a low-degree node connecting to a high-degree one
-    deg_u = G.degree(u)
-    deg_v = G.degree(v)
+    deg_u = degrees[u] if degrees is not None else G.degree(u)
+    deg_v = degrees[v] if degrees is not None else G.degree(v)
     if min(deg_u, deg_v) <= 2 and max(deg_u, deg_v) >= 5:
         score += 1
         peripheral = G.nodes[u].get("label", u) if deg_u <= 2 else G.nodes[v].get("label", v)
@@ -262,6 +263,7 @@ def _cross_file_surprises(G: nx.Graph, communities: dict[int, list[str]], top_n:
     Each result includes a 'why' field explaining what makes it non-obvious.
     """
     node_community = _node_community_map(communities)
+    degrees = dict(G.degree())
     candidates = []
 
     for u, v, data in G.edges(data=True):
@@ -279,7 +281,7 @@ def _cross_file_surprises(G: nx.Graph, communities: dict[int, list[str]], top_n:
         if not u_source or not v_source or u_source == v_source:
             continue
 
-        score, reasons = _surprise_score(G, u, v, data, node_community, u_source, v_source)
+        score, reasons = _surprise_score(G, u, v, data, node_community, u_source, v_source, degrees)
         src_id = data.get("_src", u)
         if src_id not in G.nodes:
             src_id = u

tests/test_analyze.py

@@ -105,6 +105,27 @@ def test_surprising_connections_ambiguous_scores_higher_than_extracted():
     assert score_amb > score_ext
 
 
+def test_surprise_score_accepts_precomputed_degrees():
+    G = nx.Graph()
+    for nid, label, src in [
+        ("hub", "Hub", "repo1/hub.py"),
+        ("leaf", "Leaf", "repo2/leaf.py"),
+        ("n1", "N1", "repo1/n1.py"),
+        ("n2", "N2", "repo1/n2.py"),
+        ("n3", "N3", "repo1/n3.py"),
+        ("n4", "N4", "repo1/n4.py"),
+    ]:
+        G.add_node(nid, label=label, source_file=src, file_type="code")
+    for node in ("leaf", "n1", "n2", "n3", "n4"):
+        G.add_edge("hub", node, relation="calls", confidence="EXTRACTED", weight=1.0)
+
+    nc = {"hub": 0, "leaf": 1}
+    edge = G.edges["hub", "leaf"]
+    args = (G, "hub", "leaf", edge, nc, "repo1/hub.py", "repo2/leaf.py")
+
+    assert _surprise_score(*args) == _surprise_score(*args, dict(G.degree()))
+
+
 def test_surprising_connections_cross_type_scores_higher():
     """Code↔paper edge should score higher than code↔code edge."""
     G = nx.Graph()