Fix tests and linter conflicts

Author: Dominikus Nold

.github/pull_request_template.md

@@ -70,7 +70,7 @@ Please check all that apply:
 
 ## Checklist
 
-- [ ] My code follows the style guidelines (PEP 8, Black, isort)
+- [ ] My code follows the style guidelines (PEP 8, ruff format, isort)
 - [ ] I have performed a self-review of my code
 - [ ] I have added/updated contracts (`@icontract`, `@beartype`)
 - [ ] I have added/updated docstrings (Google style)

pyproject.toml

@@ -72,7 +72,6 @@ dev = [
     "pytest-asyncio>=1.2.0",
     "pytest-xdist>=3.8.0",
     "basedpyright>=1.32.1",
-    "black>=25.9.0",
     "isort>=7.0.0",
     "pylint>=4.0.2",
     "ruff>=0.14.2",
@@ -118,7 +117,6 @@ dependencies = [
     "python-dotenv",
     "pre-commit",
     # Ensure format/lint tools are available in the hatch env
-    "black>=25.9.0",
     "isort>=7.0.0",
     "basedpyright>=1.32.1",
     "pylint>=4.0.2",
@@ -137,9 +135,9 @@ dependencies = [
 test = "pytest {args}"
 test-cov = "pytest --cov=src --cov-report=term-missing {args}"
 type-check = "basedpyright {args}"
-lint = "black . --line-length=120 && basedpyright && ruff check . && ruff format . --check && pylint src tests tools"
+lint = "ruff format . --check && basedpyright && ruff check . && pylint src tests tools"
 governance = "pylint src tests tools --reports=y --output-format=parseable"
-format = "black . --line-length=120 && ruff check . --fix && ruff format ."
+format = "ruff check . --fix && ruff format ."
 
 # Code scanning (Semgrep)
 scan = "semgrep --config tools/semgrep/async.yml {args}"
@@ -366,28 +364,8 @@ exclude = [
 # [tool.hatch.envs.default.env-vars] # Add if you have default env vars for hatch environments
 # MY_VAR = "value"
 
-[tool.black]
-line-length = 120
-target-version = ["py312"] # From your original config
-include = '''\.pyi?$''' # From template
-exclude = '''
-/(
-    \.eggs
-  | \.git
-  | \.hg
-  | \.mypy_cache
-  | \.tox
-  | \.venv
-  | _build
-  | buck-out
-  | build
-  | dist
-  # Add project-specific excludes if any
-)/
-'''
-
 [tool.isort]
-profile = "black"
+profile = "ruff"
 multi_line_output = 3 
 line_length = 120 # From your original config
 
@@ -517,8 +495,8 @@ disable = [
     "C0115",  # missing-class-docstring
     "C0116",  # missing-function-docstring
     "C0103",  # invalid-name (too restrictive for some cases)
-    "C0330",  # bad-continuation (handled by black)
-    "C0326",  # bad-whitespace (handled by black)
+    "C0330",  # bad-continuation (handled by ruff format)
+    "C0326",  # bad-whitespace (handled by ruff format)
     "R0903",  # too-few-public-methods
     "R0913",  # too-many-arguments (too restrictive for APIs)
     "R0912",  # too-many-branches
@@ -580,7 +558,7 @@ select = [
 ]
 
 ignore = [
-    "E501",  # line too long (handled by black)
+    "E501",  # line too long (handled by ruff format)
     "C901",  # too complex (leave to pylint for governance)
     "PLR0913",  # too many arguments (pylint handles this better)
     "PLR0912",  # too many branches (pylint handles this better)
@@ -638,8 +616,8 @@ ignore = [
 ]
 
 [tool.ruff.lint.isort]
-# Match isort black profile configuration
-# Black-compatible: multi_line_output = 3, combine_as_imports = true
+# Match isort ruff profile configuration
+# Ruff-compatible: multi_line_output = 3, combine_as_imports = true
 force-single-line = false
 force-wrap-aliases = false
 combine-as-imports = true

src/specfact_cli/agents/analyze_agent.py

@@ -294,9 +294,7 @@ def _load_codebase_context(self, repo_path: Path) -> dict[str, Any]:
         context["dependencies"] = dependencies
 
         # Generate summary
-        context[
-            "summary"
-        ] = f"""
+        context["summary"] = f"""
 Repository: {repo_path.name}
 Total code files: {len(filtered_files)}
 Languages detected: {", ".join({f.suffix for f in filtered_files[:20]})}

src/specfact_cli/commands/sync.py

@@ -33,6 +33,7 @@ def _is_test_mode() -> bool:
         return True
     # Check if running under pytest (common patterns)
     import sys
+
     return any("pytest" in arg or "test" in arg.lower() for arg in sys.argv) or "pytest" in sys.modules
 
 

tests/e2e/test_complete_workflow.py

@@ -1962,12 +1962,12 @@ def test_story_points_fibonacci_compliance(self):
 
         for feature in plan.features:
             for story in feature.stories:
-                assert (
-                    story.story_points in valid_fibonacci
-                ), f"Story {story.key} has invalid story points: {story.story_points}"
-                assert (
-                    story.value_points in valid_fibonacci
-                ), f"Story {story.key} has invalid value points: {story.value_points}"
+                assert story.story_points in valid_fibonacci, (
+                    f"Story {story.key} has invalid story points: {story.story_points}"
+                )
+                assert story.value_points in valid_fibonacci, (
+                    f"Story {story.key} has invalid value points: {story.value_points}"
+                )
 
         print("✅ All stories use valid Fibonacci numbers")
 

tests/e2e/test_watch_mode_e2e.py

@@ -180,7 +180,6 @@ def run_watch_mode() -> None:
     @pytest.mark.timeout(10)
     def test_watch_mode_bidirectional_sync(self) -> None:
         """Test that watch mode handles bidirectional sync with changes on both sides."""
-        import shutil
 
         with TemporaryDirectory() as tmpdir:
             repo_path = Path(tmpdir)
@@ -280,10 +279,10 @@ def run_watch_mode() -> None:
                             if file.is_file():
                                 file.unlink()
                         # Try to remove the directory
-                        try:
-                            gates_results.rmdir()
-                        except OSError:
-                            pass  # Directory might not be empty, that's okay
+                        from contextlib import suppress
+
+                        with suppress(OSError):
+                            gates_results.rmdir()  # Directory might not be empty, that's okay
 
     def test_watch_mode_detects_repository_changes(self) -> None:
         """Test that watch mode detects and syncs repository code changes."""

tests/integration/analyzers/test_code_analyzer_integration.py

@@ -426,16 +426,16 @@ def __init__(self):
                 if module_b_name and module_a_name:
                     # Module B should depend on Module A
                     # Edge direction: module_b -> module_a (B imports A, so B depends on A)
-                    assert analyzer.dependency_graph.has_edge(
-                        module_b_name, module_a_name
-                    ), f"Missing edge from {module_b_name} to {module_a_name}. Available edges: {edges}"
+                    assert analyzer.dependency_graph.has_edge(module_b_name, module_a_name), (
+                        f"Missing edge from {module_b_name} to {module_a_name}. Available edges: {edges}"
+                    )
 
                 if module_c_name and module_b_name and module_c_name != module_b_name:
                     # Module C should depend on Module B
                     # Edge direction: module_c -> module_b (C imports B, so C depends on B)
-                    assert analyzer.dependency_graph.has_edge(
-                        module_c_name, module_b_name
-                    ), f"Missing edge from {module_c_name} to {module_b_name}. Available edges: {edges}"
+                    assert analyzer.dependency_graph.has_edge(module_c_name, module_b_name), (
+                        f"Missing edge from {module_c_name} to {module_b_name}. Available edges: {edges}"
+                    )
             else:
                 # If no edges, at least verify we have the nodes
                 assert len(module_names) >= 3, f"Expected at least 3 modules, got: {module_names}"