Index .psm1, anchor Swift import targets, dedupe no-cluster edges

- #1315: add .psm1 to CODE_EXTENSIONS + _DISPATCH so PowerShell modules are indexed
- #1327: synthesize a module node for Swift import targets (new LanguageConfig
  flag synthesize_import_module_nodes) so imports edges survive build.py pruning;
  strengthen the Swift dangling-edge test to also assert edge targets
- #1317: dedupe parallel edges by (source,target,relation) in the --no-cluster
  and incremental update write paths so edge counts are deterministic and
  `update` is idempotent

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tests/test_build.py

@@ -2,10 +2,36 @@ import json
 from pathlib import Path
 import networkx as nx
 from networkx.readwrite import json_graph
-from graphify.build import build_from_json, build, build_merge, edge_data, edge_datas
+from graphify.build import build_from_json, build, build_merge, edge_data, edge_datas, dedupe_edges
 
 FIXTURES = Path(__file__).parent / "fixtures"
 
+
+def test_dedupe_edges_collapses_exact_parallels():
+    # #1317: --no-cluster / incremental update concatenate edge lists raw.
+    edges = [
+        {"source": "a", "target": "b", "relation": "calls", "source_location": "L1"},
+        {"source": "a", "target": "b", "relation": "calls", "source_location": "L9"},  # dup
+        {"source": "a", "target": "b", "relation": "imports"},  # different relation: kept
+        {"source": "b", "target": "c", "relation": "calls"},
+    ]
+    out = dedupe_edges(edges)
+    keys = [(e["source"], e["target"], e["relation"]) for e in out]
+    assert keys == [("a", "b", "calls"), ("a", "b", "imports"), ("b", "c", "calls")]
+    # first occurrence wins (keeps L1, not L9)
+    assert out[0]["source_location"] == "L1"
+
+
+def test_dedupe_edges_is_idempotent():
+    edges = [
+        {"source": "a", "target": "b", "relation": "calls"},
+        {"source": "a", "target": "b", "relation": "calls"},
+    ]
+    once = dedupe_edges(edges)
+    twice = dedupe_edges(once + edges)  # simulate a second `update` re-concatenating
+    assert len(once) == 1
+    assert len(twice) == 1
+
 def load_extraction():
     return json.loads((FIXTURES / "extraction.json").read_text())
 

tests/test_detect.py

@@ -11,6 +11,10 @@ def test_classify_python():
 def test_classify_typescript():
     assert classify_file(Path("bar.ts")) == FileType.CODE
 
+def test_classify_powershell_module():
+    # #1315: .psm1 modules were never indexed (CODE_EXTENSIONS gap).
+    assert classify_file(Path("Utils.psm1")) == FileType.CODE
+
 def test_classify_markdown():
     assert classify_file(Path("README.md")) == FileType.DOCUMENT
 

tests/test_languages.py

@@ -596,6 +596,28 @@ def test_swift_no_dangling_edges():
     node_ids = {n["id"] for n in r["nodes"]}
     for e in r["edges"]:
         assert e["source"] in node_ids
+        # #1327: targets must resolve to a node too, else build.py prunes the edge.
+        assert e["target"] in node_ids, f"dangling target {e['target']} ({e['relation']})"
+
+
+def test_swift_imports_survive_build():
+    # #1327: `import Foundation` / `import UIKit` previously emitted edges to bare
+    # module ids with no backing node, so build.py dropped 100% of Swift imports.
+    from graphify.build import build_from_json
+    r = extract_swift(FIXTURES / "sample.swift")
+    import_edges = [e for e in r["edges"] if e["relation"] == "imports"]
+    assert import_edges, "extractor should emit Swift import edges"
+    node_ids = {n["id"] for n in r["nodes"]}
+    for e in import_edges:
+        assert e["target"] in node_ids  # synthesized module node exists
+    # No private bookkeeping key should leak into output edges.
+    assert all("_import_label" not in e for e in r["edges"])
+    # Edges must survive the build (which prunes edges with unknown endpoints).
+    G = build_from_json(r)
+    surviving = [
+        (u, v) for u, v, d in G.edges(data=True) if d.get("relation") == "imports"
+    ]
+    assert surviving, "Swift import edges must survive build_from_json (#1327)"
 
 def test_swift_finds_actor():
     r = extract_swift(FIXTURES / "sample.swift")
@@ -1005,6 +1027,20 @@ def test_powershell_no_error():
     assert "error" not in r
 
 
+def test_powershell_psm1_dispatched_and_extracted(tmp_path):
+    # #1315: .psm1 modules were never indexed — no dispatch entry, no CODE_EXTENSIONS.
+    from graphify.extract import _get_extractor
+    mod = tmp_path / "Utils.psm1"
+    mod.write_text(
+        "function Get-Greeting { param([string]$Name) return \"Hi $Name\" }\n",
+        encoding="utf-8",
+    )
+    assert _get_extractor(mod) is extract_powershell
+    r = extract_powershell(mod)
+    assert "error" not in r
+    assert any("Get-Greeting" in n["label"] for n in r["nodes"])
+
+
 def test_powershell_finds_class_and_method():
     r = extract_powershell(FIXTURES / "sample.ps1")
     labels = [n["label"] for n in r["nodes"]]