refactor(docs): code analysis engine

tom-sapletta-com · tom-sapletta-com · commit cd480257250d · 2026-03-26T10:09:19.000+01:00
changes:
  - file: prompt.py
    area: cli
    modified: [_export_prompt_txt, _export_chunked_prompt_txt]
    removed: [_build_source_files_section]
  - file: test_prompt_txt.py
    area: test
    modified: [TestPromptTxtGeneration]
    removed: [test_prompt_txt_includes_orchestrator_source_file_description, temp_source_dir]

stats:
  lines: "+239/-291 (net -52)"
  files: 10
  complexity: "Large structural change (normalized)"
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,22 @@
 ## [Unreleased]
 
+## [0.5.95] - 2026-03-26
+
+### Docs
+- Update project/context.md
+
+### Test
+- Update tests/test_prompt_txt.py
+
+### Other
+- Update code2llm/cli_exports/prompt.py
+- Update project/analysis.toon.yaml
+- Update project/calls.mmd
+- Update project/calls.png
+- Update project/flow.png
+- Update project/index.html
+- Update project/map.toon.yaml
+
 ## [0.5.94] - 2026-03-26
 
 ### Docs
diff --git a/VERSION b/VERSION
@@ -1 +1 @@
-0.5.94
+0.5.95
diff --git a/code2llm/__init__.py b/code2llm/__init__.py
@@ -8,7 +8,7 @@
 and entity resolution with multilingual support.
 """
 
-__version__ = "0.5.94"
+__version__ = "0.5.95"
 __author__ = "STTS Project"
 
 # Core analysis components (lightweight, always needed)
diff --git a/code2llm/cli_exports/prompt.py b/code2llm/cli_exports/prompt.py
@@ -14,7 +14,6 @@ def _export_prompt_txt(args, output_dir: Path, formats: list[str], source_path:
     lines = _build_prompt_header(project_path)
     lines.extend(_build_main_files_section(output_dir, output_rel_path))
     lines.extend(_build_optional_files_section(output_dir, output_rel_path))
-    lines.extend(_build_source_files_section(source_path))
 
     missing = _get_missing_files(output_dir)
     if missing:
@@ -45,7 +44,6 @@ def _export_chunked_prompt_txt(args, output_dir: Path, formats: list[str], sourc
     lines = _build_prompt_header(project_path)
     lines.extend(_build_main_files_section(output_dir, output_rel_path))
     lines.extend(_build_optional_files_section(output_dir, output_rel_path))
-    lines.extend(_build_source_files_section(source_path))
 
     if subprojects:
         lines.extend(_build_subprojects_section(subprojects, output_dir, output_rel_path))
@@ -102,11 +100,6 @@ def _get_prompt_paths(source_path: Optional[Path], output_dir: Path) -> Tuple[st
 ]
 
 
-_SOURCE_FILES = [
-    ('code2llm/cli_exports/orchestrator.py', 'Export orchestration - coordinates single-project and chunked export flows, including prompt generation'),
-]
-
-
 def _build_prompt_header(project_path: str) -> List[str]:
     """Build header section of prompt."""
     return [
@@ -162,27 +155,6 @@ def _build_optional_files_section(output_dir: Path, output_rel_path: str) -> Lis
     return ["", "Optional files:", ""] + lines
 
 
-def _build_source_files_section(source_path: Optional[Path]) -> List[str]:
-    """Build source files section with size metrics."""
-    if source_path is None:
-        return []
-
-    source_root = source_path if source_path.is_dir() else source_path.parent
-    lines = []
-    for rel_path, desc in _SOURCE_FILES:
-        file_path = source_root / rel_path
-        if not file_path.exists():
-            continue
-
-        size_str = _format_size(file_path.stat().st_size)
-        lines.append(f"- {rel_path}  ({desc}) [{size_str}]")
-
-    if not lines:
-        return []
-
-    return ["", "Source files:", ""] + lines
-
-
 def _format_size(size_bytes: int) -> str:
     """Format file size in human readable format."""
     if size_bytes < 1024:
diff --git a/code2llm/nlp/__init__.py b/code2llm/nlp/__init__.py
@@ -4,7 +4,7 @@
 with multilingual support and fuzzy matching.
 """
 
-__version__ = "0.5.94"
+__version__ = "0.5.95"
 
 from .pipeline import NLPPipeline
 from .normalization import QueryNormalizer
diff --git a/project/analysis.toon.yaml b/project/analysis.toon.yaml
@@ -26,7 +26,7 @@ PIPELINES[637]:
       PURITY: 100% pure
   [4] Src [main]: main → load_file → is_toon_file
       PURITY: 100% pure
-  [5] Src [main]: main → create_html → get_shield_url
+  [5] Src [__init__]: __init__
       PURITY: 100% pure
 
 LAYERS:
diff --git a/project/calls.mmd b/project/calls.mmd
diff --git a/project/calls.png b/project/calls.png
diff --git a/project/context.md b/project/context.md
@@ -82,15 +82,15 @@
 - **Classes**: 2
 - **File**: `side_effects.py`
 
-### code2llm.core.file_analyzer
+### code2llm.analysis.cfg
 - **Functions**: 17
 - **Classes**: 1
-- **File**: `file_analyzer.py`
+- **File**: `cfg.py`
 
-### code2llm.analysis.cfg
+### code2llm.core.file_analyzer
 - **Functions**: 17
 - **Classes**: 1
-- **File**: `cfg.py`
+- **File**: `file_analyzer.py`
 
 ### code2llm.exporters.evolution_exporter
 - **Functions**: 17
@@ -425,10 +425,10 @@ Sections: PIPELINES, TRANSFORMS, CONTRACTS, DATA_TY
 - **Key Methods**: code2llm.analysis.call_graph.CallGraphExtractor.__init__, code2llm.analysis.call_graph.CallGraphExtractor.extract, code2llm.analysis.call_graph.CallGraphExtractor._calculate_metrics, code2llm.analysis.call_graph.CallGraphExtractor.visit_Import, code2llm.analysis.call_graph.CallGraphExtractor.visit_ImportFrom, code2llm.analysis.call_graph.CallGraphExtractor.visit_ClassDef, code2llm.analysis.call_graph.CallGraphExtractor.visit_FunctionDef, code2llm.analysis.call_graph.CallGraphExtractor.visit_AsyncFunctionDef, code2llm.analysis.call_graph.CallGraphExtractor.visit_Call, code2llm.analysis.call_graph.CallGraphExtractor._qualified_name
 - **Inherits**: ast.NodeVisitor
 
-### code2llm.nlp.normalization.QueryNormalizer
-> Normalize queries for consistent processing.
+### code2llm.nlp.intent_matching.IntentMatcher
+> Match queries to intents using fuzzy and keyword matching.
 - **Methods**: 13
-- **Key Methods**: code2llm.nlp.normalization.QueryNormalizer.__init__, code2llm.nlp.normalization.QueryNormalizer.normalize, code2llm.nlp.normalization.QueryNormalizer._unicode_normalize, code2llm.nlp.normalization.QueryNormalizer._lowercase, code2llm.nlp.normalization.QueryNormalizer._remove_punctuation, code2llm.nlp.normalization.QueryNormalizer._normalize_whitespace, code2llm.nlp.normalization.QueryNormalizer._remove_stopwords, code2llm.nlp.normalization.QueryNormalizer._tokenize, code2llm.nlp.normalization.QueryNormalizer.step_1a_lowercase, code2llm.nlp.normalization.QueryNormalizer.step_1b_remove_punctuation
+- **Key Methods**: code2llm.nlp.intent_matching.IntentMatcher.__init__, code2llm.nlp.intent_matching.IntentMatcher.match, code2llm.nlp.intent_matching.IntentMatcher._fuzzy_match, code2llm.nlp.intent_matching.IntentMatcher._keyword_match, code2llm.nlp.intent_matching.IntentMatcher._apply_context, code2llm.nlp.intent_matching.IntentMatcher._combine_matches, code2llm.nlp.intent_matching.IntentMatcher._resolve_multi_intent, code2llm.nlp.intent_matching.IntentMatcher._calculate_similarity, code2llm.nlp.intent_matching.IntentMatcher.step_2a_fuzzy_match, code2llm.nlp.intent_matching.IntentMatcher.step_2b_semantic_match
 
 ## Data Transformation Functions
 
@@ -494,6 +494,10 @@ Checks:
 > Generate all format outputs and evaluate them.
 - **Output to**: format_configs.items, __import__, getattr, exporter_cls, time.time
 
+### code2llm.core.repo_files._get_gitignore_parser
+> Load gitignore parser for project if available.
+- **Output to**: code2llm.core.gitignore.load_gitignore_patterns
+
 ### code2llm.core.large_repo.HierarchicalRepoSplitter._process_large_dirs
 > Process large directories with file-level chunking.
 - **Output to**: self._chunk_by_files, chunks.extend
@@ -502,10 +506,6 @@ Checks:
 > Process Python files directly in level1 directory.
 - **Output to**: code2llm.core.repo_files._get_gitignore_parser, len, chunks.append, self._chunk_by_files, chunks.extend
 
-### code2llm.core.repo_files._get_gitignore_parser
-> Load gitignore parser for project if available.
-- **Output to**: code2llm.core.gitignore.load_gitignore_patterns
-
 ### code2llm.core.gitignore.GitIgnoreParser._parse_pattern
 > Parse a single gitignore pattern into regex.
 - **Output to**: pattern.startswith, pattern.endswith, pattern.startswith, self._wildcard_to_regex, re.compile
@@ -520,6 +520,10 @@ Returns list of (module_name, start_line, end_line).
 > Check if file should be processed.
 - **Output to**: file_path.lower, Path, self._gitignore_parser.is_ignored, any, fnmatch.fnmatch
 
+### code2llm.analysis.cfg.CFGExtractor._format_except
+> Format except handler.
+- **Output to**: self._expr_to_str
+
 ### code2llm.core.file_analyzer.FileAnalyzer._process_class
 > Process class definition.
 - **Output to**: ClassInfo, None.classes.append, isinstance, isinstance, methods.append
@@ -532,10 +536,6 @@ Returns list of (module_name, start_line, end_line).
 > Process a block of statements for CFG with depth limiting.
 - **Output to**: None.append, isinstance, None.append, FlowEdge, self._process_if_stmt
 
-### code2llm.core.file_analyzer.FileAnalyzer._process_if_stmt
-> Process if statement for CFG.
-- **Output to**: FlowNode, func_info.cfg_nodes.append, None.append, self._process_cfg_block, FlowNode
-
 ## Behavioral Patterns
 
 ### recursion__is_excluded
diff --git a/project/flow.png b/project/flow.png
diff --git a/project/index.html b/project/index.html
diff --git a/project/map.toon.yaml b/project/map.toon.yaml
@@ -398,6 +398,9 @@ Sections: P...
   code2llm/analysis/type_inference.py:
     e: TypeInferenceEngine
     TypeInferenceEngine: __init__(0),enrich_function(1),get_arg_types(1),get_return_type(1),get_typed_signature(1),extract_all_types(1),_get_ast(1),_find_function_node(3),_extract_from_node(2),_extract_args(1),_annotation_to_str(1),_ann_constant(1),_ann_name(1),_ann_attribute(1),_ann_subscript(1),_ann_tuple(1),_ann_binop(1),_infer_from_name(1),_infer_arg_type(1)  # Extract and infer type information from Python source files...
+  code2llm/analysis/call_graph.py:
+    e: CallGraphExtractor
+    CallGraphExtractor(ast.NodeVisitor): __init__(1),extract(3),_calculate_metrics(0),visit_Import(1),visit_ImportFrom(1),visit_ClassDef(1),visit_FunctionDef(1),visit_AsyncFunctionDef(1),visit_Call(1),_qualified_name(1),_resolve_call(1),_resolve_with_astroid(1),_expr_to_str(1)  # Extract call graph from AST...
   code2llm/core/large_repo.py:
     e: SubProject,HierarchicalRepoSplitter,should_use_chunking,get_analysis_plan
     SubProject:  # Represents a sub-project within a larger repository...
@@ -406,9 +409,6 @@ Sections: P...
 Stra...
     should_use_chunking(project_path;size_threshold_kb)
     get_analysis_plan(project_path;size_limit_kb)
-  code2llm/analysis/call_graph.py:
-    e: CallGraphExtractor
-    CallGraphExtractor(ast.NodeVisitor): __init__(1),extract(3),_calculate_metrics(0),visit_Import(1),visit_ImportFrom(1),visit_ClassDef(1),visit_FunctionDef(1),visit_AsyncFunctionDef(1),visit_Call(1),_qualified_name(1),_resolve_call(1),_resolve_with_astroid(1),_expr_to_str(1)  # Extract call graph from AST...
   code2llm/core/refactoring.py:
     e: RefactoringAnalyzer
     RefactoringAnalyzer: __init__(2),perform_refactoring_analysis(1),_build_call_graph(1),_calculate_centrality(2),_detect_cycles(2),_detect_communities(2),_analyze_coupling(1),_detect_smells(1),_detect_dead_code(1),_map_dead_code_to_items(2),_mark_reachable_items(1)  # Performs refactoring analysis on code...
diff --git a/pyproject.toml b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "code2llm"
-version = "0.5.94"
+version = "0.5.95"
 description = "High-performance Python code flow analysis with optimized TOON format - CFG, DFG, call graphs, and intelligent code queries"
 readme = "README.md"
 requires-python = ">=3.8"
diff --git a/setup.py b/setup.py
@@ -4,7 +4,7 @@
 import os
 
 # Read version
-version = "0.5.93"
+version = "0.5.94"
 
 # Read long description
 def read_readme():
diff --git a/tests/test_prompt_txt.py b/tests/test_prompt_txt.py
@@ -17,12 +17,6 @@ def temp_output_dir(self):
         with tempfile.TemporaryDirectory() as tmpdir:
             yield Path(tmpdir)
 
-    @pytest.fixture
-    def temp_source_dir(self):
-        """Create temporary source directory."""
-        with tempfile.TemporaryDirectory() as tmpdir:
-            yield Path(tmpdir)
-    
     @pytest.fixture
     def mock_args(self):
         """Create mock args object with verbose flag."""
@@ -158,24 +152,6 @@ def test_prompt_txt_prioritizes_blockers_when_validation_and_duplication_exist(s
         assert "P1 — Split or simplify the highest-CC / god modules" in content, "Analysis-driven refactoring should follow blockers"
         assert "P2 — Keep the compact project overview" in content, "Project overview should be preserved as a later priority"
 
-    def test_prompt_txt_includes_orchestrator_source_file_description(self, temp_output_dir, temp_source_dir, mock_args):
-        """Test that prompt.txt includes the orchestrator.py source file description."""
-        formats = ['code2logic']
-        source_path = temp_source_dir / 'myproject'
-
-        orchestrator_file = source_path / 'code2llm' / 'cli_exports' / 'orchestrator.py'
-        orchestrator_file.parent.mkdir(parents=True, exist_ok=True)
-        orchestrator_file.write_text('"""Export orchestration."""\n')
-
-        _export_prompt_txt(mock_args, temp_output_dir, formats, source_path)
-
-        prompt_file = temp_output_dir / 'prompt.txt'
-        content = prompt_file.read_text()
-
-        assert "Source files:" in content, "Source files section should be present"
-        assert "code2llm/cli_exports/orchestrator.py" in content, "Orchestrator file should be listed"
-        assert "Export orchestration" in content, "Orchestrator description should be included"
-    
     def test_prompt_txt_content_structure(self, temp_output_dir, mock_args):
         """Test the overall structure of generated prompt.txt."""
         formats = ['code2logic']
