agent-coderag 1.5.0__tar.gz → 1.5.1__tar.gz

{agent_coderag-1.5.0/agent_coderag.egg-info → agent_coderag-1.5.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-coderag
-Version: 1.5.0
+Version: 1.5.1
 Summary: Lightweight semantic code search and distillation utility for AI coding agents. It solves the API knowledge gap via real-time local signature extraction and intent analysis without PyTorch. Optimized for token efficiency, it compresses codebase context into compact semantic summaries stored in a local DuckDB vector similarity index.
 Author-email: Igor Boloban <naranor@gmail.com>
 License: MIT

{agent_coderag-1.5.0 → agent_coderag-1.5.1/agent_coderag.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-coderag
-Version: 1.5.0
+Version: 1.5.1
 Summary: Lightweight semantic code search and distillation utility for AI coding agents. It solves the API knowledge gap via real-time local signature extraction and intent analysis without PyTorch. Optimized for token efficiency, it compresses codebase context into compact semantic summaries stored in a local DuckDB vector similarity index.
 Author-email: Igor Boloban <naranor@gmail.com>
 License: MIT

{agent_coderag-1.5.0 → agent_coderag-1.5.1}/code_rag/entry/args.py

@@ -118,3 +118,7 @@ def main() -> None:
     except Exception as exc:
         logging.getLogger(__name__).error("Unexpected error: %s", exc)
         sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

{agent_coderag-1.5.0 → agent_coderag-1.5.1}/code_rag/parsers/tree_sitter.py

@@ -74,6 +74,7 @@ class TreeSitterParser(IParser):
                 "file_path": identity,
                 "units": units,
                 "scope": None,
+                "name_counts": {},
             }
             self._recursive_distill(tree.root_node, ctx)
             return units
@@ -87,21 +88,35 @@ class TreeSitterParser(IParser):
 
     def _recursive_distill(self, node: Node, ctx: Dict[str, Any]) -> None:
         config = ctx["config"]
-        current_scope = ctx["scope"]
+        child_scope = ctx["scope"]
+        child_counts = ctx["name_counts"]
 
         if node.type in config.canonical_map:
-            self._process_node(node, ctx)
-            if config.canonical_map.get(node.type) == "CLASS":
-                node_name = self._resolve_name(node)
-                scope = ctx["scope"]
-                current_scope = f"{scope}.{node_name}" if scope else node_name
-
-        # Recurse into children with updated scope
-        child_ctx = {**ctx, "scope": current_scope}
+            qname = self._qualified_name(node, ctx)
+            self._process_node(node, ctx, qname)
+            if config.canonical_map.get(node.type) in ("CLASS", "FUNCTION", "METHOD"):
+                child_scope = qname
+                child_counts = {}
+
+        child_ctx = {**ctx, "scope": child_scope, "name_counts": child_counts}
         for child in node.children:
             self._recursive_distill(child, child_ctx)
 
-    def _process_node(self, node: Node, ctx: Dict[str, Any]) -> None:
+    def _qualified_name(self, node: Node, ctx: Dict[str, Any]) -> str:
+        """Qualified id segment. Enclosing functions are part of the scope.
+
+        A repeated name in the same scope gets ``#2``, ``#3``, … so two
+        different bodies cannot share one primary key.
+        """
+        node_name = self._resolve_name(node)
+        counts: Dict[str, int] = ctx["name_counts"]
+        counts[node_name] = counts.get(node_name, 0) + 1
+        seen = counts[node_name]
+        label = node_name if seen == 1 else f"{node_name}#{seen}"
+        scope = ctx["scope"]
+        return f"{scope}.{label}" if scope else label
+
+    def _process_node(self, node: Node, ctx: Dict[str, Any], qname: str) -> None:
         """Processes a single node and adds it to the units list."""
         node_name = self._resolve_name(node)
         docstring = self._extract_docstring(node)
@@ -113,9 +128,6 @@ class TreeSitterParser(IParser):
 
         kind = self._determine_kind(node, ctx["config"])
 
-        # Stable ID based on qualified name
-        scope = ctx["scope"]
-        qname = f"{scope}.{node_name}" if scope else node_name
         unit_id = f"{ctx['file_path']}:{qname}"
 
         unit = KnowledgeUnit(

{agent_coderag-1.5.0 → agent_coderag-1.5.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "agent-coderag"
-version = "1.5.0"
+version = "1.5.1"
 description = "Lightweight semantic code search and distillation utility for AI coding agents. It solves the API knowledge gap via real-time local signature extraction and intent analysis without PyTorch. Optimized for token efficiency, it compresses codebase context into compact semantic summaries stored in a local DuckDB vector similarity index."
 readme = "README.md"
 requires-python = ">=3.10"

{agent_coderag-1.5.0 → agent_coderag-1.5.1}/tests/test_cli_startup.py

@@ -1,3 +1,4 @@
+import runpy
 import subprocess
 import sys
 from types import SimpleNamespace
@@ -33,6 +34,12 @@ assert "onnxruntime" not in sys.modules
     assert result.returncode == 0, result.stderr
 
 
+def test_dunder_main_invokes_main(capsys):
+    with patch("sys.argv", ["agent-coderag"]):
+        runpy.run_path(cli_args.__file__, run_name="__main__")
+    assert "CodeRAG" in capsys.readouterr().out
+
+
 def test_main_help_and_verbose_do_not_dispatch(capsys):
     with patch("sys.argv", ["agent-coderag"]):
         cli_args.main()

{agent_coderag-1.5.0 → agent_coderag-1.5.1}/tests/test_tree_sitter_parser.py

@@ -114,6 +114,58 @@ const arrow = (x) => {
         os.unlink(temp_path)
 
 
+@pytest.mark.asyncio
+async def test_nested_functions_keep_distinct_ids():
+    parser = TreeSitterParser()
+    content = """
+def test_one():
+    def helper():
+        return 1
+
+def test_two():
+    def helper():
+        return 2
+
+def same_scope():
+    def helper():
+        return 3
+
+    def helper():
+        return 4
+"""
+    with tempfile.NamedTemporaryFile(suffix=".py", delete=False) as f:
+        f.write(content.encode("utf-8"))
+        temp_path = f.name
+
+    try:
+        units = await parser.distill_file(temp_path)
+        ids = [unit.id for unit in units]
+        assert f"{temp_path}:test_one.helper" in ids
+        assert f"{temp_path}:test_two.helper" in ids
+        assert f"{temp_path}:same_scope.helper" in ids
+        assert f"{temp_path}:same_scope.helper#2" in ids
+        assert len(ids) == len(set(ids))
+    finally:
+        os.unlink(temp_path)
+
+
+@pytest.mark.asyncio
+async def test_export_does_not_wrap_inner_function_scope():
+    parser = TreeSitterParser()
+    content = "export function exportedFn() { return 1; }\n"
+    with tempfile.NamedTemporaryFile(suffix=".js", delete=False) as f:
+        f.write(content.encode("utf-8"))
+        temp_path = f.name
+
+    try:
+        units = await parser.distill_file(temp_path)
+        ids = {unit.id for unit in units}
+        assert f"{temp_path}:anonymous" in ids
+        assert f"{temp_path}:exportedFn" in ids
+    finally:
+        os.unlink(temp_path)
+
+
 @pytest.mark.asyncio
 async def test_missing_grammar():
     parser = TreeSitterParser()