codexa 0.4.0__py3-none-any.whl

semantic_code_intelligence/tests/test_phase25.py

@@ -0,0 +1,496 @@
+"""Phase 25 — Incremental Indexing & Quality Refactors.
+
+Tests verify:
+  1.  VectorStore.remove_by_file — correct removal and index rebuild
+  2.  Incremental indexing — stale vector removal, deleted file cleanup
+  3.  HF_TOKEN configuration — env var propagation
+  4.  CallGraph regex matching — word boundary accuracy
+  5.  Web viz data format — edges key in callgraph response
+  6.  Silent exception logging — debug messages instead of bare pass
+  7.  Refactored CLI output helpers — quality_cmd, metrics_cmd
+  8.  Module imports and version
+"""
+
+from __future__ import annotations
+
+import os
+import re
+from pathlib import Path
+from unittest.mock import MagicMock, patch
+
+import numpy as np
+import pytest
+
+from semantic_code_intelligence.storage.vector_store import ChunkMetadata, VectorStore
+
+_PROJECT_ROOT = Path(__file__).resolve().parents[2]
+
+
+# ===================================================================
+# 1. VectorStore.remove_by_file
+# ===================================================================
+
+
+class TestVectorStoreRemoveByFile:
+    """Tests for VectorStore.remove_by_file()."""
+
+    def _make_store(self) -> VectorStore:
+        store = VectorStore(dimension=4)
+        vecs = np.array([
+            [1.0, 0.0, 0.0, 0.0],
+            [0.0, 1.0, 0.0, 0.0],
+            [0.0, 0.0, 1.0, 0.0],
+            [0.0, 0.0, 0.0, 1.0],
+        ], dtype=np.float32)
+        meta = [
+            ChunkMetadata(file_path="/a.py", start_line=1, end_line=10,
+                          chunk_index=0, language="python", content="a1"),
+            ChunkMetadata(file_path="/b.py", start_line=1, end_line=5,
+                          chunk_index=0, language="python", content="b1"),
+            ChunkMetadata(file_path="/a.py", start_line=11, end_line=20,
+                          chunk_index=1, language="python", content="a2"),
+            ChunkMetadata(file_path="/c.py", start_line=1, end_line=15,
+                          chunk_index=0, language="python", content="c1"),
+        ]
+        store.add(vecs, meta)
+        return store
+
+    def test_remove_existing_file(self) -> None:
+        store = self._make_store()
+        assert store.size == 4
+        removed = store.remove_by_file("/a.py")
+        assert removed == 2
+        assert store.size == 2
+        assert all(m.file_path != "/a.py" for m in store.metadata)
+
+    def test_remove_nonexistent_file(self) -> None:
+        store = self._make_store()
+        removed = store.remove_by_file("/nonexistent.py")
+        assert removed == 0
+        assert store.size == 4
+
+    def test_remove_all_files(self) -> None:
+        store = self._make_store()
+        store.remove_by_file("/a.py")
+        store.remove_by_file("/b.py")
+        store.remove_by_file("/c.py")
+        assert store.size == 0
+        assert store.metadata == []
+
+    def test_search_after_removal(self) -> None:
+        store = self._make_store()
+        store.remove_by_file("/a.py")
+        query = np.array([0.0, 1.0, 0.0, 0.0], dtype=np.float32)
+        results = store.search(query, top_k=5)
+        assert len(results) == 2
+        assert results[0][0].file_path == "/b.py"
+
+    def test_add_after_removal(self) -> None:
+        store = self._make_store()
+        store.remove_by_file("/b.py")
+        new_vec = np.array([[0.5, 0.5, 0.0, 0.0]], dtype=np.float32)
+        new_meta = [ChunkMetadata(file_path="/d.py", start_line=1, end_line=3,
+                                  chunk_index=0, language="python", content="d1")]
+        store.add(new_vec, new_meta)
+        assert store.size == 4
+        paths = {m.file_path for m in store.metadata}
+        assert "/d.py" in paths
+        assert "/b.py" not in paths
+
+
+# ===================================================================
+# 2. Incremental indexing — stale removal
+# ===================================================================
+
+
+class TestIncrementalIndexing:
+    """Tests for indexing_service stale vector and deleted file handling."""
+
+    def test_deleted_paths_detected(self) -> None:
+        """HashStore tracks should be cleaned up for deleted files."""
+        from semantic_code_intelligence.storage.hash_store import HashStore
+
+        hs = HashStore()
+        hs.set("src/old.py", "abc123")
+        hs.set("src/keep.py", "def456")
+
+        scanned_paths = {"src/keep.py"}
+        deleted = [k for k in list(hs._hashes.keys()) if k not in scanned_paths]
+        assert deleted == ["src/old.py"]
+
+    def test_hash_store_remove(self) -> None:
+        from semantic_code_intelligence.storage.hash_store import HashStore
+
+        hs = HashStore()
+        hs.set("a.py", "h1")
+        hs.set("b.py", "h2")
+        hs.remove("a.py")
+        assert hs.get("a.py") is None
+        assert hs.get("b.py") == "h2"
+
+    def test_remove_by_file_preserves_other_vectors(self) -> None:
+        """Incremental re-index removes stale vectors but keeps others."""
+        store = VectorStore(dimension=2)
+        vecs = np.array([[1.0, 0.0], [0.0, 1.0]], dtype=np.float32)
+        meta = [
+            ChunkMetadata(file_path="/changed.py", start_line=1, end_line=5,
+                          chunk_index=0, language="python", content="old"),
+            ChunkMetadata(file_path="/unchanged.py", start_line=1, end_line=5,
+                          chunk_index=0, language="python", content="keep"),
+        ]
+        store.add(vecs, meta)
+        store.remove_by_file("/changed.py")
+
+        assert store.size == 1
+        assert store.metadata[0].file_path == "/unchanged.py"
+
+        # Add updated file
+        new_vec = np.array([[0.5, 0.5]], dtype=np.float32)
+        new_meta = [ChunkMetadata(file_path="/changed.py", start_line=1, end_line=8,
+                                  chunk_index=0, language="python", content="new")]
+        store.add(new_vec, new_meta)
+        assert store.size == 2
+
+
+# ===================================================================
+# 3. HF_TOKEN configuration
+# ===================================================================
+
+
+class TestHFTokenConfig:
+    """Tests for _configure_hf_token() in embeddings/generator.py."""
+
+    def test_hf_token_already_set(self) -> None:
+        from semantic_code_intelligence.embeddings.generator import _configure_hf_token
+
+        with patch.dict(os.environ, {"HF_TOKEN": "existing"}, clear=False):
+            _configure_hf_token()
+            assert os.environ["HF_TOKEN"] == "existing"
+
+    def test_hugging_face_hub_token_propagated(self) -> None:
+        from semantic_code_intelligence.embeddings.generator import _configure_hf_token
+
+        env = {"HUGGING_FACE_HUB_TOKEN": "hub_tok"}
+        with patch.dict(os.environ, env, clear=False):
+            os.environ.pop("HF_TOKEN", None)
+            _configure_hf_token()
+            assert os.environ.get("HF_TOKEN") == "hub_tok"
+
+    def test_huggingface_token_propagated(self) -> None:
+        from semantic_code_intelligence.embeddings.generator import _configure_hf_token
+
+        env = {"HUGGINGFACE_TOKEN": "hf_tok"}
+        with patch.dict(os.environ, env, clear=False):
+            os.environ.pop("HF_TOKEN", None)
+            os.environ.pop("HUGGING_FACE_HUB_TOKEN", None)
+            _configure_hf_token()
+            assert os.environ.get("HF_TOKEN") == "hf_tok"
+
+    def test_no_token_set(self) -> None:
+        from semantic_code_intelligence.embeddings.generator import _configure_hf_token
+
+        with patch.dict(os.environ, {}, clear=False):
+            os.environ.pop("HF_TOKEN", None)
+            os.environ.pop("HUGGING_FACE_HUB_TOKEN", None)
+            os.environ.pop("HUGGINGFACE_TOKEN", None)
+            _configure_hf_token()
+            assert "HF_TOKEN" not in os.environ
+
+
+# ===================================================================
+# 4. CallGraph regex matching
+# ===================================================================
+
+
+class TestCallGraphRegex:
+    """Tests for word-boundary call detection in context/engine.py."""
+
+    def test_exact_function_call_detected(self) -> None:
+        pattern = re.compile(r"\bprocess_data\s*[\(\.]")
+        body = "result = process_data(items)"
+        assert pattern.search(body) is not None
+
+    def test_method_call_detected(self) -> None:
+        pattern = re.compile(r"\bprocess_data\s*[\(\.]")
+        body = "obj.process_data(x)"
+        assert pattern.search(body) is not None
+
+    def test_substring_not_matched(self) -> None:
+        pattern = re.compile(r"\bprocess_data\s*[\(\.]")
+        body = "unprocess_data_handler = 1"
+        assert pattern.search(body) is None
+
+    def test_partial_name_not_matched(self) -> None:
+        pattern = re.compile(r"\bget\s*[\(\.]")
+        body = "get_user()"
+        # "get" at word boundary followed by "_" should not match \(|\.)
+        assert pattern.search(body) is None
+
+    def test_call_with_space(self) -> None:
+        pattern = re.compile(r"\bfoo\s*[\(\.]")
+        body = "foo (x)"
+        assert pattern.search(body) is not None
+
+
+# ===================================================================
+# 5. Web viz data format
+# ===================================================================
+
+
+class TestVizDataFormat:
+    """Tests for viz endpoint data format."""
+
+    def test_callgraph_has_edges_key(self) -> None:
+        from semantic_code_intelligence.bridge.context_provider import ContextProvider
+
+        provider = ContextProvider(_PROJECT_ROOT)
+        with patch.object(provider, "_ensure_indexed") as mock_idx:
+            mock_builder = MagicMock()
+            mock_idx.return_value = mock_builder
+            mock_builder.get_call_graph.return_value = MagicMock(
+                callers={"a": ["b"]},
+                callees={"a": ["c"]},
+            )
+            data = provider.get_call_graph(symbol_name="a")
+            assert "edges" in data
+            assert isinstance(data["edges"], list)
+
+
+# ===================================================================
+# 6. Silent exception logging
+# ===================================================================
+
+
+class TestSilentExceptionLogging:
+    """Verify that previously silent catches now log debug messages."""
+
+    @pytest.mark.parametrize("module_path", [
+        "semantic_code_intelligence.cli.commands.explain_cmd",
+        "semantic_code_intelligence.ci.hooks",
+        "semantic_code_intelligence.cli.commands.quality_cmd",
+        "semantic_code_intelligence.cli.commands.pr_summary_cmd",
+        "semantic_code_intelligence.llm.investigation",
+        "semantic_code_intelligence.llm.cross_refactor",
+        "semantic_code_intelligence.cli.commands.chat_cmd",
+        "semantic_code_intelligence.cli.commands.cross_refactor_cmd",
+        "semantic_code_intelligence.ci.metrics",
+        "semantic_code_intelligence.cli.commands.hotspots_cmd",
+        "semantic_code_intelligence.cli.commands.impact_cmd",
+        "semantic_code_intelligence.cli.commands.trace_cmd",
+        "semantic_code_intelligence.docs",
+        "semantic_code_intelligence.llm.streaming",
+    ])
+    def test_module_has_no_bare_pass_in_except(self, module_path: str) -> None:
+        """Ensure no bare 'except Exception: pass' remains in the source.
+
+        JSON/TypeError fallback patterns (e.g. ``except (json.JSONDecodeError, TypeError): pass``)
+        are allowed because they represent intentional parse-fallback, not swallowed errors.
+        """
+        import importlib
+        mod = importlib.import_module(module_path)
+        src_path = Path(mod.__file__)
+        source = src_path.read_text(encoding="utf-8")
+        lines = source.splitlines()
+        for i, line in enumerate(lines):
+            stripped = line.strip()
+            if stripped == "pass" and i > 0:
+                prev = lines[i - 1].strip()
+                if prev.startswith("except"):
+                    # Allow JSON/type-error parse fallbacks
+                    if "JSONDecodeError" in prev or "TypeError" in prev:
+                        continue
+                    if "Exception" in prev or prev == "except:":
+                        pytest.fail(
+                            f"{module_path}:{i + 1} has bare 'except: pass' — "
+                            f"should use logger.debug()"
+                        )
+
+
+# ===================================================================
+# 7. Refactored CLI output helpers
+# ===================================================================
+
+
+class TestQualityCmdHelpers:
+    """Tests for extracted quality_cmd output helpers."""
+
+    def test_output_safety_json(self) -> None:
+        from semantic_code_intelligence.cli.commands.quality_cmd import _output_safety
+
+        safety = MagicMock()
+        safety.safe = True
+        safety.to_dict.return_value = {"safe": True, "issues": []}
+
+        from io import StringIO
+        from unittest.mock import patch as _patch
+        from click.testing import CliRunner
+
+        import click
+
+        @click.command()
+        def _test_cmd() -> None:
+            _output_safety(safety, 10, json_mode=True, pipe=False)
+
+        runner = CliRunner()
+        result = runner.invoke(_test_cmd)
+        assert '"safe": true' in result.output
+
+    def test_output_safety_pipe(self) -> None:
+        from semantic_code_intelligence.cli.commands.quality_cmd import _output_safety
+
+        safety = MagicMock()
+        safety.safe = True
+
+        import click
+        from click.testing import CliRunner
+
+        @click.command()
+        def _test_cmd() -> None:
+            _output_safety(safety, 5, json_mode=False, pipe=True)
+
+        runner = CliRunner()
+        result = runner.invoke(_test_cmd)
+        assert "PASS" in result.output
+        assert "5 files" in result.output
+
+    def test_output_report_pipe(self) -> None:
+        from semantic_code_intelligence.cli.commands.quality_cmd import _output_report_pipe
+
+        report = MagicMock()
+        report.files_analyzed = 10
+        report.symbol_count = 50
+        report.issue_count = 3
+        report.complexity_issues = []
+        report.dead_code = []
+        report.duplicates = []
+        report.bandit_issues = []
+        report.safety = None
+        report.maintainability_index = None
+
+        import click
+        from click.testing import CliRunner
+
+        @click.command()
+        def _test_cmd() -> None:
+            _output_report_pipe(report)
+
+        runner = CliRunner()
+        result = runner.invoke(_test_cmd)
+        assert "Files: 10" in result.output
+        assert "Issues: 3" in result.output
+
+
+class TestMetricsCmdHelpers:
+    """Tests for extracted metrics_cmd output helpers."""
+
+    def test_output_history_json(self) -> None:
+        from semantic_code_intelligence.cli.commands.metrics_cmd import _output_history
+
+        snap = MagicMock()
+        snap.to_dict.return_value = {"timestamp": 1000, "mi": 65.0}
+
+        import click
+        from click.testing import CliRunner
+
+        @click.command()
+        def _test_cmd() -> None:
+            _output_history([snap], 1, json_mode=True, pipe=False)
+
+        runner = CliRunner()
+        result = runner.invoke(_test_cmd)
+        assert '"snapshots"' in result.output
+
+    def test_output_trend_pipe(self) -> None:
+        from semantic_code_intelligence.cli.commands.metrics_cmd import _output_trend
+
+        trend = MagicMock()
+        trend.metric_name = "maintainability_index"
+        trend.direction = "improving"
+        trend.oldest_value = 60.0
+        trend.newest_value = 70.0
+        trend.delta = 10.0
+
+        import click
+        from click.testing import CliRunner
+
+        @click.command()
+        def _test_cmd() -> None:
+            _output_trend([trend], 5, json_mode=False, pipe=True)
+
+        runner = CliRunner()
+        result = runner.invoke(_test_cmd)
+        assert "TREND" in result.output
+        assert "maintainability_index" in result.output
+
+    def test_output_current_metrics_json(self) -> None:
+        from semantic_code_intelligence.cli.commands.metrics_cmd import _output_current_metrics
+
+        pm = MagicMock()
+        pm.to_dict.return_value = {"files_analyzed": 10, "mi": 65.0}
+
+        import click
+        from click.testing import CliRunner
+
+        @click.command()
+        def _test_cmd() -> None:
+            _output_current_metrics(pm, Path("."), None, json_mode=True, pipe=False)
+
+        runner = CliRunner()
+        result = runner.invoke(_test_cmd)
+        assert '"files_analyzed"' in result.output
+
+
+# ===================================================================
+# 8. Module imports and version
+# ===================================================================
+
+
+class TestImportsAndVersion:
+    """Verify that all Phase 25 modules import correctly."""
+
+    def test_version_0_25(self) -> None:
+        import semantic_code_intelligence
+        assert semantic_code_intelligence.__version__ == "0.4.0"
+
+    def test_vector_store_has_remove_by_file(self) -> None:
+        assert hasattr(VectorStore, "remove_by_file")
+
+    def test_hf_token_function_importable(self) -> None:
+        from semantic_code_intelligence.embeddings.generator import _configure_hf_token
+        assert callable(_configure_hf_token)
+
+    def test_quality_helpers_importable(self) -> None:
+        from semantic_code_intelligence.cli.commands.quality_cmd import (
+            _output_safety,
+            _output_report_pipe,
+            _output_report_rich,
+        )
+        assert callable(_output_safety)
+        assert callable(_output_report_pipe)
+        assert callable(_output_report_rich)
+
+    def test_metrics_helpers_importable(self) -> None:
+        from semantic_code_intelligence.cli.commands.metrics_cmd import (
+            _output_history,
+            _output_trend,
+            _output_current_metrics,
+        )
+        assert callable(_output_history)
+        assert callable(_output_trend)
+        assert callable(_output_current_metrics)
+
+    def test_indexing_helpers_importable(self) -> None:
+        from semantic_code_intelligence.services.indexing_service import (
+            _extract_symbols,
+            _compute_index_stats,
+        )
+        assert callable(_extract_symbols)
+        assert callable(_compute_index_stats)
+
+    def test_visualize_module_importable(self) -> None:
+        from semantic_code_intelligence.web.visualize import (
+            render_call_graph,
+            render_dependency_graph,
+        )
+        assert callable(render_call_graph)
+        assert callable(render_dependency_graph)